Siderius Type X-Linked Intellectual Disability

Genetic MONDO:0010286 Pathograph 18 Show in embeddings browser Neurodevelopmental Disorder X-linked Disorder Chromatinopathy

Siderius type syndromic X-linked intellectual disability (MRXSSD; OMIM 300263; ORPHA:85287) is a rare X-linked chromatinopathy caused by loss-of-function variants in PHF8 at Xp11.22. PHF8 encodes a PHD-finger plus JmjC-domain, Fe(II)/2-oxoglutarate-dependent histone demethylase that erases the repressive marks H4K20me1 and H3K9me1/me2 at promoters to which it is recruited by PHD-finger recognition of H3K4me3 and by direct contact with RNA polymerase II. Hemizygous males present with developmental delay and borderline to severe intellectual disability, mild facial dysmorphism, and — in a minority of contemporary cases — orofacial clefting, the feature on which the syndrome was originally delineated. Autism spectrum disorder and attention deficit hyperactivity disorder were added to the phenotype by the largest published series. The entity has a two-phase literature. The founding descriptions (Siderius 1999; Laumonnier 2005; Abidi 2007; Koivisto 2007) were ascertained through cleft clinics and multiplex X-linked pedigrees, so cleft lip/palate looked obligate. The 2022 multicentre series of 16 additional individuals, ascertained largely by exome sequencing for developmental delay, found orofacial clefting in only 3 of 16 and explicitly reported that the feature is less common than previously reported. Both ascertainment frames are curated here, with the frequency bands taken from the contemporary series. IMPORTANT SCOPE NOTE - this entity is defined by PHF8. It must not be conflated with (a) PHF6-related Borjeson-Forssman-Lehmann syndrome or PHF21A-related disorders, which are different PHD-finger genes with different diseases; (b) the other eponymous X-linked intellectual disability types, in particular Claes-Jensen type (KDM5C) and Lubs type (MECP2 duplication), which are separate MONDO entities with separate genes - note that the PHF8 and KDM5C literatures are genuinely entangled because PHF8, ZNF711 and KDM5C/JARID1C sit in one transcriptional module, but the diseases are distinct; or (c) the large somatic-cancer PHF8 literature, which is a different disease context entirely.

Ask OpenScientist

Ask a research question about Siderius Type X-Linked Intellectual Disability. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).

Submitting...

Do not include personal health information in your question. Questions and results are cached in your browser's local storage.

🏷

Classifications

Harrison's Chapter

👪

Inheritance

X-linked recessive inheritance HP:0001419

This is the authoritative OMIM/HPO inheritance annotation and correctly describes the reported pedigrees: affected individuals are hemizygous males, the founding families were multiplex X-linked pedigrees with transmission through unaffected carrier mothers, and the original linkage was to Xp11.3-q21.3. Penetrance is recorded as incomplete and expressivity as variable on the strength of two documented observations within families: in the original Siderius-Hamel family only two of three affected individuals had both intellectual disability and clefting while the third had mild intellectual disability alone, and in the Finnish F279S sibship the identical allele produced unilateral clefting in one brother and bilateral clefting in the other. No systematic study of heterozygous female carriers has been published for this entity, so nothing is asserted about female manifestation here.

X-linked recessive inheritance Penetrance: INCOMPLETE Expressivity: VARIABLE

Show evidence (4 references)

PMID:10398231 SUPPORT Human Clinical

"Linkage analysis showed a maximum LOD score of Z=2.78 at straight theta=0.0 for the DXS441 locus with flanking markers DXS337 and DXS990, defining the region Xp11.3-q21.3 with a linkage interval of 25 cM."

The original X-linkage evidence that established the inheritance pattern before the gene was known.

PMID:17594395 SUPPORT Human Clinical

"The second mutation was present in a family with four affected men, three of whom had MR and CL/P, while the fourth individual had mild MR without clefting."

Documents intrafamilial variability for the clefting component, supporting the VARIABLE expressivity annotation.

PMID:17661819 SUPPORT Human Clinical

"unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling"

Same-allele, same-family variation in cleft laterality, a second independent line of support for variable expressivity.

+ 1 more reference

◈

Mechanistic Hypotheses

mTOR pathway suppression as a disease-modifying strategy

mtor_targeting EMERGING

Evidence balance 1 support

Because the cognitive and synaptic deficits of the Phf8 knockout mouse are driven by RSK1-dependent mTOR hyperactivation and are reversed by rapamycin, mTOR inhibition has been proposed as a rational therapeutic target for PHF8-related intellectual disability. The hypothesis rests on a single mouse study. It has not been tested in any human with a PHF8 variant, the therapeutic window in a chronic paediatric neurodevelopmental indication is unknown, and rapamycin's immunosuppressive and metabolic toxicity is substantial. It is recorded here as an emerging research direction, not as a treatment.

Show evidence (1 reference)

PMID:29317619 SUPPORT Model Organism

"and provides a potential therapeutic drug target to treat XLID"

The authors' own framing of the finding as a drug-target hypothesis.

Oxygen-dependence of PHF8 catalysis as a link to hypoxia-associated clefting

oxygen_sensing_clefting EMERGING

Evidence balance 1 partial

PHF8 is a 2-oxoglutarate/Fe(II) oxygenase and therefore consumes molecular oxygen. This has prompted the suggestion that partial loss of PHF8 activity and gestational hypoxia converge on the same craniofacial endpoint, which would predict that maternal hypoxic exposures modify cleft risk in carriers. No human data address this. It is recorded because it is the only proposed gene-environment interaction for this disorder and because it would be testable.

Show evidence (1 reference)

PMID:19843542 PARTIAL In Vitro

"The dependence of PHF8 activity on oxygen availability is interesting because the occurrence of fetal cleft lip has been demonstrated to increase with maternal hypoxia in mouse studies."

The originating statement of the hypothesis. Marked PARTIAL because the paper juxtaposes a biochemical property with a separate mouse observation rather than demonstrating the interaction.

Discussions and Knowledge Gaps

What is the true frequency of orofacial clefting in PHF8-related intellectual disability once ascertainment bias is removed?

KNOWLEDGE GAP OPEN phf8_clefting_ascertainment

Attached to

pathophysiology#Orofacial Clefting and Craniofacial Dysmorphology

Every published frequency for clefting in this disorder is confounded by how the cases were found. The founding families came from cleft clinics and multiplex X-linked pedigrees, so clefting appeared obligate; the 2022 series came largely from exome sequencing for developmental delay and found clefting in 3 of 16. Neither denominator is unbiased. Because clefting is the feature that gives the syndrome its clinical identity, and because its presence or absence currently drives whether PHF8 is considered at all, this is the single most consequential unresolved number in the entry.

Proposed experiments

Cleft frequency in PHF8 carriers from an unselected developmental-delay cohort

unselected_dd_cohort_cleft_frequency

Ascertain PHF8 loss-of-function carriers from a large-scale sequencing cohort recruited on developmental delay alone, with no craniofacial inclusion criterion, then phenotype for orofacial clefting prospectively.

Decision criterion

The cleft frequency in the unselected cohort differs significantly from both the founding-cohort estimate and the 3-of-16 figure.

Would support

Clefting frequency in PHF8-related intellectual disability is ascertainment-determined rather than a fixed property of the disorder.

Would refute

Clefting is an intrinsically variable feature whose frequency is stable across ascertainment frames.

Reverse phenotyping of PHF8 variant carriers in population biobanks

biobank_reverse_phenotyping

Identify PHF8 loss-of-function carriers in population biobanks with no neurodevelopmental ascertainment at all, and retrieve their craniofacial and cognitive phenotypes from linked health records.

Decision criterion

A substantial fraction of biobank PHF8 carriers have neither clefting nor recorded intellectual disability.

Would support

The published phenotype is the severe tail of a broader and milder distribution.

Does the rapamycin rescue of the Phf8 knockout mouse predict anything about human PHF8-related intellectual disability, given that a second independent knockout line showed no cognitive impairment at all?

HUMAN MODEL MISMATCH OPEN phf8_mtor_rapamycin_translation

Attached to

pathophysiology#RSK1 Derepression and mTOR Pathway Hyperactivation

The mouse result is striking - mTOR hyperactivation is mechanistically specified down to RSK1, and rapamycin restores both LTP and cognition. But the rescue rests on a single knockout line, and a separately generated Phf8 knockout allele reported neither developmental defects nor cognitive impairment, so the very phenotype the rescue corrects was not reproducible between laboratories. Neither line reproduces the craniofacial phenotype, and no measurement of mTOR pathway activity has been reported in any human tissue from a person with a PHF8 variant. Whether the human disorder is an mTOR-opathy at all is therefore untested, and treating the mouse rescue as a therapeutic lead would be premature in a chronic paediatric indication where rapamycin carries real toxicity.

Proposed experiments

mTOR pathway activity in patient-derived PHF8-mutant neurons

patient_ipsc_mtor_activity

Measure S6K and S6 phosphorylation and RSK1 protein levels in iPSC-derived cortical neurons from individuals carrying PHF8 loss-of-function alleles, against isogenic corrected controls.

Decision criterion

PHF8-mutant human neurons show elevated mTOR pathway activity relative to isogenic controls.

Would support

The mouse mTOR mechanism operates in human PHF8-deficient neurons.

Would refute

The mTOR arm is a mouse-specific consequence of Phf8 loss.

RSK1 derepression by the catalytic-dead F279S allele

f279s_rsk1_derepression

Test whether the F279S catalytic-dead missense allele, and not only a complete null, derepresses RSK1, since several reported human alleles retain protein.

Decision criterion

F279S produces RSK1 derepression comparable to the null allele.

Would support

The mTOR arm is engaged by catalytic loss alone and therefore applies to the missense end of the human allele spectrum.

Side-by-side comparison of the two Phf8 knockout alleles

phf8_ko_line_comparison

Backcross both published Phf8 knockout alleles onto a common genetic background and run identical learning, memory and hippocampal LTP assays in the same laboratory.

Decision criterion

The cognitive phenotype tracks the allele rather than the background.

Would support

The published discordance is allele-driven and one allele is not a true null.

Would refute

The discordance is background- or protocol-driven, which would weaken both reports equally.

Why do Phf8-null mice fail to develop orofacial clefting when human PHF8 loss-of-function causes cleft lip/palate and zebrafish phf8 knockdown causes jaw malformation?

HUMAN MODEL MISMATCH OPEN phf8_mouse_no_clefting

Attached to

pathophysiology#Impaired Neural Crest and Craniofacial Midline Development

The craniofacial arm of this entry rests almost entirely on zebrafish jaw and msxb data, because both published Phf8 knockout mouse lines are reported without craniofacial malformation - one explicitly as having no obvious developmental defects, the other as having no gross morphological defects. That is a species mismatch at exactly the point where the human phenotype is most distinctive. Paralogue compensation by PHF2/KDM7A in mouse, or a species difference in how much lip and palate fusion depends on MSX1, would both explain it, and distinguishing them would materially change how the craniofacial mechanism should be modelled.

Proposed experiments

Phf8;Phf2 double-mutant mouse craniofacial phenotyping

phf8_phf2_double_mutant

Generate mice null for both Phf8 and its closest paralogue Phf2 and assess lip and palate fusion, to test whether paralogue compensation masks a craniofacial requirement in mouse.

Decision criterion

Clefting emerges in double mutants but not in either single mutant.

Would support

Paralogue compensation explains the absence of clefting in Phf8-null mice.

Msx1 expression in Phf8-null mouse facial primordia

msx1_facial_primordia_assay

Assay Msx1 transcript and protein in the facial primordia of Phf8-null mouse embryos across the lip and palate fusion window.

Decision criterion

Msx1 expression is reduced in Phf8-null primordia.

Would support

The zebrafish msxb mechanism is conserved in mammals even where the morphological endpoint is not.

Would refute

The MSX1 link is fish-specific and the human craniofacial mechanism is unexplained.

Are heterozygous female carriers of pathogenic PHF8 variants affected?

KNOWLEDGE GAP OPEN phf8_female_carrier_status

Attached to

pathophysiology#Pathogenic PHF8 Loss-of-Function Variant

The disorder is annotated X-linked recessive and every reported affected individual is male, but for several other X-linked chromatinopathies - KDM5C and USP9X among them - the recessive label has proved empirically wrong for females once carriers were systematically assessed. The only X-inactivation data in this disorder come from a single unaffected deletion-carrier mother with complete skewing. No systematic study of PHF8 carrier females has been published, so the absence of reported affected females may reflect absence of looking rather than absence of effect. This matters directly for genetic counselling.

Proposed experiments

Formal cognitive phenotyping of obligate PHF8 carrier mothers

carrier_mother_cognitive_phenotyping

Recruit obligate carrier mothers from the published PHF8 pedigrees and apply formal cognitive and adaptive-behaviour testing rather than clinical impression, with population norms as the comparator.

Decision criterion

Carrier mothers score below population norms on formal testing.

Would support

The X-linked recessive annotation understates female involvement.

Determine whether PHF8 escapes X-inactivation

phf8_xci_escape

Measure allele-specific PHF8 expression in female cells with informative heterozygous markers and quantified X-inactivation ratios.

Decision criterion

PHF8 shows measurable expression from the inactive X.

Would support

Females have a higher baseline PHF8 dose, which would predict milder or absent manifestation.

Does PHF8-related intellectual disability have a DNA-methylation episignature that could resolve its reported variants of uncertain significance?

KNOWLEDGE GAP OPEN phf8_episignature

Attached to

pathophysiology#Failure of RNA Polymerase II-Coupled Transcriptional Coactivation

Many chromatinopathies - Kabuki, Sotos and CHARGE among them - now have peripheral-blood DNA-methylation episignatures that reclassify missense variants of uncertain significance into pathogenic or benign. No such classifier exists for PHF8. This is not an academic gap: the largest series reported five individuals from four families with intellectual disability, a PHF8 variant of uncertain significance and no alternative explanation, which is precisely the population an episignature would resolve. PHF8 is a transcriptional coactivator acting on repressive histone marks, so a downstream methylation signature is mechanistically plausible.

Proposed experiments

Peripheral-blood methylation classifier for PHF8 loss of function

phf8_methylation_classifier

Generate genome-wide methylation array data from peripheral blood of individuals with confirmed PHF8 loss-of-function variants and train a classifier against matched controls and against other chromatinopathy signatures.

Decision criterion

A reproducible signature separates PHF8 loss-of-function carriers from controls with high sensitivity and specificity.

Would support

PHF8 belongs among the episignature-classifiable chromatinopathies.

Classification of reported PHF8 missense variants of uncertain significance

phf8_vus_classification

Apply the trained classifier to samples from the five reported individuals carrying PHF8 missense variants of uncertain significance.

Decision criterion

Each variant of uncertain significance is assigned to the affected or the control cluster.

Would support

Episignature testing can resolve the PHF8 variant-interpretation backlog.

Show evidence (1 reference)

PMID:35469323 SUPPORT Human Clinical

"We also present five individuals from four different families who have ID and a variant of unknown significance in PHF8 with no other explanatory variant in another gene."

Establishes that the variant-of-uncertain-significance population this gap concerns actually exists and is quantified.

⚙

Pathophysiology

Pathogenic PHF8 Loss-of-Function Variant

The shared upstream lesion is a hemizygous loss-of-function variant in PHF8 at Xp11.22. The reported allele spectrum is small but mechanistically coherent: nonsense and frameshift alleles that truncate the protein before or within the JmjC domain and remove the nuclear localisation signals, splice alleles, missense alleles clustered in the JmjC-encoding exons, and whole-gene microdeletions. All reported mechanisms converge on loss of function; no dominant-negative or gain-of-function mechanism has been demonstrated for the germline neurodevelopmental phenotype. Two consequences fan out from this node in parallel rather than in series - loss of JmjC catalysis, and, for the truncating alleles, loss of the PHD-finger/nuclear-targeting module - because the F279S missense allele abolishes catalysis while leaving the N-terminal PHD finger intact, and the K177X nonsense allele removes both.

PHF8 hgnc:20672

Downstream

Loss of JmjC-Domain Histone Demethylase Activity

Loss of PHD-Finger H3K4me3 Reading and Nuclear Targeting

Show evidence (3 references)

PMID:16199551 SUPPORT Human Clinical

"Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)."

The gene-discovery paper, and the strongest available identity anchor - it binds PHF8 directly to MIM 300263, which is the OMIM xref of MONDO:0010286.

PMID:35469323 SUPPORT Human Clinical

"Loss-of-function variants in PHD Finger Protein 8 (PHF8) cause Siderius X-linked intellectual disability (ID) syndrome, hereafter called PHF8-XLID."

States the loss-of-function mechanism and names the entity being curated here.

PMID:19843542 SUPPORT In Vitro

"Clinically observed mutations to the PHF8 gene cluster in exons encoding for the double stranded beta-helix fold and will therefore disrupt catalytic activity."

Establishes that the clinical allele spectrum is concentrated in the catalytic fold, which is why the catalytic arm is the dominant downstream consequence of this node.

Loss of JmjC-Domain Histone Demethylase Activity

PHF8 is an Fe(II)- and 2-oxoglutarate-dependent oxygenase of the double-stranded beta-helix class. Its JmjC domain removes mono- and dimethyl marks from lysine but not trimethyl marks. Patient alleles clustered in the JmjC-encoding exons abolish this catalysis: the F279S missense variant, which substitutes a residue in a conserved hydrophobic region of the JmjC domain, is catalytically inactive against both peptide and intact-histone substrates, and truncating alleles remove the domain outright. Because the reaction requires molecular oxygen, PHF8 activity is oxygen-dependent, a property that has been proposed - though not demonstrated in patients - to link the catalytic arm to the known association between maternal hypoxia and cleft lip in mouse studies.

histone H4K20 demethylase activity GO:0035575 ↓ DECREASED histone H3K9 demethylase activity GO:0032454 ↓ DECREASED

Downstream

Persistence of Repressive H4K20me1 at Target Promoters

Persistence of Repressive H3K9me1/me2 Marks

Show evidence (4 references)

PMID:19843542 SUPPORT In Vitro

"We report that recombinant PHF8 is an Fe(II) and 2-oxoglutarate-dependent N(epsilon)-methyl lysine demethylase, which acts on histone substrates."

Defines the enzyme class and cofactor dependence of the catalytic arm.

PMID:19843542 SUPPORT In Vitro

"This mutant encodes a F279S variant of PHF8 that modifies a conserved hydrophobic region; assays with both peptides and intact histones reveal this variant to be catalytically inactive."

Direct demonstration that a patient missense allele abolishes catalysis.

PMID:20101266 SUPPORT In Vitro

"The in vitro demethylation assay also showed that the F279S mutant observed in clinical patients possesses no demethylation activity, suggesting that loss of enzymatic activity is crucial for pathogenesis of PHF8 patients."

Independent replication of catalytic inactivity, with the authors explicitly assigning pathogenic weight to the catalytic loss.

+ 1 more reference

Loss of PHD-Finger H3K4me3 Reading and Nuclear Targeting

The N-terminal PHD finger of PHF8 binds H3K4me3/me2, the mark that decorates active transcription start sites, and this reader function is what recruits the enzyme to the promoters it acts on. Truncating patient alleles remove not only the JmjC domain but also the nuclear localisation signals, so the residual protein cannot reach chromatin at all. This is modelled as a sibling of the catalytic arm rather than a downstream consequence, because the F279S catalytic-dead allele leaves the PHD finger intact while the K177X nonsense allele removes both modules - the two arms are separable in the patient allele series.

histone H3K4me3 reader activity GO:0140002 ↓ DECREASED

Downstream

Failure of RNA Polymerase II-Coupled Transcriptional Coactivation

Show evidence (4 references)

PMID:20421419 SUPPORT In Vitro

"Our biochemical analysis revealed specific association of the PHF8 PHD with histone H3 trimethylated at lysine 4 (H3K4me3)."

Establishes the PHD finger as an H3K4me3 reader module.

PMID:20548336 SUPPORT In Vitro

"In addition, we show that PHF8 binds specifically to H3K4me3/2 peptides via an N-terminal PHD finger domain."

Independent confirmation of the reader specificity.

PMID:17594395 SUPPORT Human Clinical

"The mutation results in a truncated PHF8 protein lacking the Jumonji-like C terminus domain and five nuclear localization signals."

Documents that a patient nonsense allele removes the nuclear localisation signals in addition to the catalytic domain, which is the nuclear-targeting component of this node.

+ 1 more reference

Persistence of Repressive H4K20me1 at Target Promoters

PHF8 is the founding H4K20me1 demethylase. When catalysis is lost, H4K20me1 accumulates at the transcription start sites of PHF8 target genes and the genes stay repressed. Two H4K20me1-dependent target sets have been mapped directly: cytoskeletal and cell-adhesion regulators (RHOA, RAC1, GSK3B) whose promoters PHF8 demethylates, and RSK1, whose derepression is the entry point into the mTOR arm. In astrocytes the related H4K20me3 mark accumulates at synaptogenic genes when PHF8 is depleted, extending the same heterochromatinisation logic to glia.

Chromatin organization GO:0006325 ⚠ ABNORMAL

Downstream

Failure of RNA Polymerase II-Coupled Transcriptional Coactivation

RSK1 Derepression and mTOR Pathway Hyperactivation

Show evidence (3 references)

PMID:20622853 SUPPORT In Vitro

"Here we provide multiple lines of evidence establishing PHF8 as the first mono-methyl histone H4 lysine 20 (H4K20me1) demethylase, with additional activities towards histone H3K9me1 and me2."

Establishes H4K20me1 as a principal PHF8 substrate.

PMID:20622853 SUPPORT In Vitro

"PHF8 depletion resulted in upregulation of H4K20me1 and H3K9me1 at the TSS and H3K9me2 in the non-TSS sites, respectively, demonstrating differential substrate specificities at different target locations."

Direct demonstration that losing PHF8 causes the repressive marks to persist, and that the two marks accumulate at different genomic locations.

PMID:22850744 SUPPORT In Vitro

"Our data demonstrate that PHF8 directly regulates the expression of these genes by demethylating H4K20me1 at promoters."

Maps a specific H4K20me1-dependent target set (cytoskeletal and adhesion genes) to promoter demethylation by PHF8.

Persistence of Repressive H3K9me1/me2 Marks

PHF8 also erases mono- and dimethylated H3K9, a repressive mark associated with facultative heterochromatin. Recombinant PHF8 is selective for H3K9me2 and H3K9me1 and does not accept H3K9me3, so the enzyme trims the repressive mark toward the unmethylated state rather than resolving constitutive heterochromatin. In cells, PHF8 depletion raises H3K9me1 at transcription start sites and H3K9me2 at non-TSS sites. Loss of this activity is one of the two routes by which target promoters fail to be activated.

histone H3K9me/H3K9me2 demethylase activity GO:0140683 ↓ DECREASED

Downstream

Failure of RNA Polymerase II-Coupled Transcriptional Coactivation

Show evidence (3 references)

PMID:20101266 SUPPORT In Vitro

"Biochemical and structural studies reveal that PHF8 is a novel histone demethylase specific for di- and mono-methylated histone H3 lysine 9 (H3K9me2/1), but not for H3K9me3."

Defines the H3K9 substrate specificity, including the exclusion of the trimethyl state.

PMID:20346720 SUPPORT In Vitro

"Here, we show that the XLMR protein PHF8 and a C. elegans homolog F29B9.2 catalyze demethylation of di- and monomethylated lysine 9 of histone H3 (H3K9me2/me1)."

Independent confirmation of the H3K9me2/me1 demethylase activity, with conservation to an invertebrate orthologue.

PMID:20421419 SUPPORT In Vitro

"Here, we show that PHF8 is a histone demethylase that removes repressive histone H3 dimethyl lysine 9 marks."

Third independent demonstration of the H3K9me2 eraser activity.

Failure of RNA Polymerase II-Coupled Transcriptional Coactivation

The convergence point of the chromatin arms. PHF8 occupies the transcription start sites of thousands of active or poised genes, mirroring the distribution of RNA polymerase II and H3K4me3-marked nucleosomes, and contacts the RNAPII C-terminal domain directly; it therefore acts as a promoter-resident transcriptional coactivator whose activation function depends on PHD binding to H3K4me3. Patient alleles are defective in coactivation as well as in demethylation. PHF8 additionally partners with the XLID transcription factor ZNF711 at a shared target set, and co-occupies promoters with the REST/NRSF repressor, so the net transcriptional output at any locus can be activating or repressive; the dominant documented effect is loss of activation. Downstream of this node the disease diversifies into distinct neural and craniofacial cell-type programmes.

Regulation of transcription by RNA polymerase II GO:0006357 ⚠ ABNORMAL

transcription coactivator activity GO:0003713 ↓ DECREASED

Downstream

Impaired Neuronal Differentiation and Neurite Outgrowth

Impaired Astrocyte Differentiation and Synaptic Gene Silencing

Perturbed Oligodendroglial Development

Impaired Serine Biosynthesis in Neural Progenitors

Impaired Neural Crest and Craniofacial Midline Development

Show evidence (7 references)

PMID:20421419 SUPPORT In Vitro

"Chromatin immunoprecipitation followed by high-throughput sequencing indicated that PHF8 is enriched at the transcription start sites of many active or poised genes, mirroring the presence of RNA polymerase II (RNAPII) and of H3K4me3-bearing nucleosomes."

Places PHF8 physically at active promoters together with RNAPII.

PMID:20421419 SUPPORT In Vitro

"Importantly, a PHF8 disease mutant was defective in demethylation and in coactivation."

Directly links a patient allele to failure of the coactivation function, which is the claim of this node.

PMID:20622853 SUPPORT In Vitro

"PHF8 positively regulates gene expression, which is dependent on its H3K4me3-binding PHD and catalytic domains."

Shows that the activating output requires both upstream arms, which is why both converge on this node.

+ 4 more references

RSK1 Derepression and mTOR Pathway Hyperactivation

A specific, pharmacologically actionable branch of the H4K20me1 arm. In Phf8 knockout mice, failure to demethylate H4K20me1 at the RSK1 locus derepresses RSK1, and the mTOR signalling pathway becomes hyperactive in hippocampus. This is the one node in the entry with a demonstrated rescue: rapamycin normalises both long-term potentiation and cognitive performance in the knockout. The evidence is entirely from the mouse; no patient has been treated on this rationale.

TOR signaling GO:0031929 ↑ INCREASED

Downstream

Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation

Show evidence (2 references)

PMID:29317619 SUPPORT Model Organism

"We also show that mTOR signaling pathway is hyperactive in hippocampus in Phf8 knockout mouse."

Establishes mTOR hyperactivation as a consequence of Phf8 loss in vivo.

PMID:29317619 SUPPORT Model Organism

"Mechanistically, we show that demethylation of H4K20me1 by Phf8 results in transcriptional suppression of RSK1 and homeostasis of mTOR signaling."

Supplies the H4K20me1-to-RSK1-to-mTOR mechanism that makes this a child of the H4K20me1 node rather than a free-standing observation.

Impaired Neuronal Differentiation and Neurite Outgrowth

PHF8 is required for neuronal differentiation and for the cytoskeletal programme that builds neurites. Knockdown in P19 embryonal carcinoma cells impairs retinoic-acid-induced neuronal differentiation, and overexpression of wild-type but not the catalytically dead F279S variant pushes the cells toward a neuronal fate - a clean genotype-to-cell-phenotype link through a patient allele. In differentiated neurons, PHF8 depletion downregulates cytoskeletal genes and produces deficient neurite outgrowth, which the authors propose is the cellular substrate of the cognitive phenotype.

neuron CL:0000540

Neuron differentiation GO:0030182 ↓ DECREASED Neuron projection development GO:0031175 ↓ DECREASED

Downstream

Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation

Show evidence (3 references)

PMID:20548336 SUPPORT In Vitro

"Consistent with a role for PHF8 in neuronal differentiation, knockdown of PHF8 in mouse embryonic carcinoma P19 cells impairs RA-induced neuronal differentiation, whereas overexpression of the wild-type but not the F279S mutant PHF8 drives P19 cells toward neuronal differentiation."

Links the patient F279S allele specifically to failure of neuronal differentiation, connecting the molecular arms to a cellular phenotype.

PMID:22850744 SUPPORT In Vitro

"Further analysis in neurons shows that depletion of PHF8 results in down-regulation of cytoskeleton genes and leads to a deficient neurite outgrowth."

Documents the neurite-outgrowth defect that gives this node its second half.

PMID:22850744 SUPPORT In Vitro

"Overall, our results suggest that the mental retardation phenotype associated with loss of function of PHF8 could be due to abnormal neuronal connections as a result of alterations in cytoskeleton function."

The authors' own statement of the proposed cell-to-organism link, cited as a proposal rather than a demonstration.

Impaired Astrocyte Differentiation and Synaptic Gene Silencing

A non-cell-autonomous arm. PHF8 acts in astrocytes through a regulatory crosstalk with Notch signalling that balances expression of the master astrocytic transcription factor Nfia, and it keeps synaptogenic genes accessible by holding H4K20me3 low. Depleting PHF8 in astrocytes alone is sufficient to disrupt neuronal synapse formation and maturation in vitro, so part of the synaptic phenotype originates outside the neuron. Evidence is from mouse astrocytic culture.

astrocyte CL:0000127

Astrocyte differentiation GO:0048708 ⚠ ABNORMAL

Downstream

Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation

Show evidence (3 references)

PMID:34081130 SUPPORT In Vitro

"Using genome-wide analyses and biochemical assays in mouse astrocytic cultures, we reveal a regulatory crosstalk between PHF8 and the Notch signaling pathway that balances the expression of the master astrocytic gene Nfia."

Establishes the astrocyte-differentiation arm and its Notch/Nfia mechanism.

PMID:34081130 SUPPORT In Vitro

"Accordingly, astrocytic-PHF8 depletion has a striking effect on neuronal synapse formation and maturation in vitro."

Demonstrates the non-cell-autonomous effect on neuronal synapses, which is why this node feeds the synaptic-plasticity node.

PMID:34081130 SUPPORT In Vitro

"Moreover, PHF8 regulates key synaptic genes in astrocytes by maintaining low levels of H4K20me3."

Supplies the chromatin mechanism connecting this node back to the H4K20 methylation arm.

Perturbed Oligodendroglial Development

A third glial arm, and the least clinically anchored of the three. Phf8 promotes proliferation of oligodendrocyte progenitor cells and restrains their differentiation into oligodendrocytes, acting largely through Olig2, whose regulatory regions it occupies; ectopic Olig2 rescues the proliferation defect in Phf8-deficient cells. Notably, human iPSC-derived oligodendrocyte generation did not require PHF8 when Olig2 was force-expressed, so the human requirement is conditional. No white-matter abnormality has been documented in patients with this syndrome, so this node is marked provisional.

oligodendrocyte precursor cell CL:0002453

Oligodendrocyte differentiation GO:0048709 ⚠ ABNORMAL

Downstream

Impaired Cognitive Development

Show evidence (3 references)

PMID:38613395 SUPPORT In Vitro

"we show that Phf8 promotes the proliferation of rodent oligodendrocyte progenitor cells and impairs their differentiation to oligodendrocytes"

States the oligodendroglial phenotype of Phf8 loss and gain.

PMID:38613395 SUPPORT In Vitro

"Taking the influence of Olig2 levels on oligodendroglial proliferation and differentiation into account, Olig2 likely acts as an important downstream effector of Phf8 in these cells."

Supplies the Olig2 effector mechanism.

PMID:38613395 PARTIAL In Vitro

"Additionally, generation of human oligodendrocytes from induced pluripotent stem cells did not require PHF8 in a system that relies on forced expression of Olig2 during oligodendroglial induction."

The authors' own negative result in a human system, which is the reason this node carries PROVISIONAL confidence rather than being asserted as established human pathophysiology.

Impaired Serine Biosynthesis in Neural Progenitors

The most recently described arm, and the one that connects chromatin regulation to metabolism. In neural stem cells PHF8 drives the serine biosynthesis pathway by tuning chromatin accessibility at the promoters of metabolic genes, safeguarding the intracellular serine pool that progenitor proliferation depends on. Loss of PHF8 disrupts amino-acid metabolism, blocks autophagy and impairs vesicle formation, and culminates in replication defects, DNA damage and proliferation arrest; in mouse embryos PHF8 deficiency halts progenitor expansion and neuron generation in the developing brain. Evidence is from neural stem cells and mouse embryos, not from patients.

neural progenitor cell CL:0011020

L-serine biosynthetic process GO:0006564 ↓ DECREASED

Downstream

Impaired Cognitive Development

Show evidence (3 references)

PMID:41714361 SUPPORT In Vitro

"we identify PHF8 as a key driver of the serine biosynthesis pathway, safeguarding the intracellular serine pool essential for neural progenitor proliferation"

Establishes the serine-biosynthesis mechanism in neural stem cells.

PMID:41714361 SUPPORT Model Organism

"In vivo, PHF8 deficiency in mouse embryos halts neurogenesis, progenitor expansion, and neuron generation in the developing brain."

Extends the progenitor-proliferation defect to the intact developing brain, which is what licenses the edge to impaired cognitive development.

PMID:41714361 SUPPORT In Vitro

"PHF8 fine-tunes chromatin accessibility at promoters of metabolic genes, ensuring their activation during development."

Supplies the chromatin-to-metabolism link that makes this a child of the transcriptional-coactivation node.

Impaired Neural Crest and Craniofacial Midline Development

The craniofacial arm. PHF8 is required for jaw development in zebrafish, where it acts in part by directly regulating the homeodomain transcription factor MSX1/MSXB, an integrator of several craniofacial signalling pathways; Phf8 is independently implicated in neural crest defects in mice and humans. The clinical inference of a midline-formation role predates the molecular work and was drawn directly from the co-segregation of intellectual disability with cleft lip/palate in the original families. Because PHF8 catalysis consumes molecular oxygen, an oxygen-sensing link to the known maternal-hypoxia effect on murine cleft lip has been proposed, but this remains a hypothesis.

migratory neural crest cell CL:0000333

Embryonic cranial skeleton morphogenesis GO:0048701 ⚠ ABNORMAL Roof of mouth development GO:0060021 ⚠ ABNORMAL

Downstream

Orofacial Clefting and Craniofacial Dysmorphology

Show evidence (4 references)

PMID:20622853 SUPPORT Model Organism

"PHF8 regulates cell survival in the zebrafish brain and jaw development, thus providing a potentially relevant biological context for understanding the clinical symptoms associated with PHF8 patients."

The primary animal-model evidence for a craniofacial requirement.

PMID:20622853 SUPPORT Model Organism

"Lastly, genetic and molecular evidence supports a model whereby PHF8 regulates zebrafish neuronal cell survival and jaw development in part by directly regulating the expression of the homeodomain transcription factor MSX1/MSXB, which functions downstream of multiple signalling and developmental..."

Supplies the MSX1/MSXB effector that connects the transcriptional node to craniofacial morphogenesis.

PMID:38613395 SUPPORT Model Organism

"The plant homeodomain finger protein Phf8 is a histone demethylase implicated by mutation in mice and humans in neural crest defects and neurodevelopmental disturbances."

Independent statement placing PHF8 in neural crest development, supporting the cell-type annotation on this node.

+ 1 more reference

Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation

The convergence point of the neuronal, astrocytic and mTOR arms. Phf8 knockout mice show impaired hippocampal long-term potentiation together with impaired learning and memory, and - importantly for interpreting the human phenotype - without gross morphological brain defects, matching the absence of a consistent structural brain abnormality in patients. The synaptic deficit is reversible in the mouse: rapamycin restores the weakened LTP.

neuron CL:0000540

Long-term synaptic potentiation GO:0060291 ↓ DECREASED

Downstream

Impaired Cognitive Development

Show evidence (2 references)

PMID:29317619 SUPPORT Model Organism

"Here we report that Phf8 knockout mice displayed impaired learning and memory, and impaired hippocampal long-term potentiation (LTP) without gross morphological defects."

Establishes the LTP deficit and the absence of gross structural brain change.

PMID:29317619 SUPPORT Model Organism

"Pharmacological suppression of mTOR signaling with rapamycin in Phf8 knockout mice recovers the weakened LTP and cognitive deficits."

Demonstrates reversibility, which is the basis of the mTOR-targeting research hypothesis recorded under mechanistic_hypotheses.

Impaired Cognitive Development

The terminal neurodevelopmental output: developmental delay in essentially all affected individuals and borderline to severe intellectual disability in the large majority, with a behavioural profile that includes autism spectrum disorder and attention deficit hyperactivity disorder. The severity range is wide enough that two individuals in the largest cohort had no intellectual disability at all, one with dyscalculia and one with mild learning difficulties.

Learning or memory GO:0007611 ↓ DECREASED

Show evidence (2 references)

PMID:35469323 SUPPORT Human Clinical

"All affected individuals exhibited developmental delay and all but two had borderline to severe ID."

The quantitative human anchor for this terminal node.

PMID:29317619 SUPPORT Model Organism

"Together, our results indicate that loss of Phf8 in animals causes deficient learning and memory by epigenetic disruption of mTOR signaling"

Model-organism recapitulation of the cognitive endpoint, kept distinct from the human evidence by evidence_source.

Orofacial Clefting and Craniofacial Dysmorphology

The terminal craniofacial output: cleft lip and/or cleft palate, unilateral or bilateral, together with a mild dysmorphic gestalt that has been described as a long face with broad nasal tip, and in the contemporary series as hypertelorism, microcephaly, elongated face, ptosis and mild facial asymmetry. Clefting was near-universal in the cleft-clinic-ascertained founding families and is present in a minority of exome-ascertained cases, so the apparent frequency of this node is strongly ascertainment-dependent.

Roof of mouth development GO:0060021 ⚠ ABNORMAL

Show evidence (2 references)

PMID:35469323 SUPPORT Human Clinical

"Orofacial clefting was seen in three individuals from our cohort, suggesting that this feature is less common than previously reported."

The contemporary frequency anchor, and the explicit statement of the ascertainment shift.

PMID:20101266 SUPPORT Human Clinical

"Mutations and truncations in human plant homeodomain (PHD) finger protein 8 (PHF8) are associated with X-linked mental retardation and facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"

The classical dysmorphic gestalt as summarised in the structural literature.

⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.

●

Phenotypes

Eye 2

Hypertelorism OCCASIONAL Hypertelorism HP:0000316

Show evidence (1 reference)

PMID:35469323 SUPPORT Human Clinical

"Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."

Supports the association and the OCCASIONAL band by qualitative-term mapping ("in some affected individuals"); the band is applied to the craniofacial group as a whole because no per-feature count is given.

Ptosis OCCASIONAL Ptosis HP:0000508

Show evidence (1 reference)

PMID:35469323 SUPPORT Human Clinical

"Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."

Supports the association and the OCCASIONAL band by qualitative-term mapping applied to the craniofacial group.

Head and Neck 7

Orofacial Clefting OCCASIONAL Orofacial cleft HP:0000202

Show evidence (2 references)

PMID:35469323 SUPPORT Human Clinical

"Orofacial clefting was seen in three individuals from our cohort, suggesting that this feature is less common than previously reported."

Supports the association and the OCCASIONAL band: 3 of 16 (18.8%), which falls in the 5-29% range.

PMID:16199551 SUPPORT Human Clinical

"Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)."

The founding gene-disease report, in which clefting co-segregated with the PHF8 variant.

Cleft Lip Cleft lip HP:0410030

Show evidence (1 reference)

PMID:17661819 SUPPORT Human Clinical

"unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling"

Documents cleft lip and its laterality variation within one sibship.

Cleft Palate Cleft palate HP:0000175

Show evidence (1 reference)

PMID:17661819 SUPPORT Human Clinical

"unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling"

Documents cleft palate co-occurring with cleft lip in both affected siblings.

Microcephaly OCCASIONAL Microcephaly HP:0000252

Show evidence (1 reference)

PMID:35469323 SUPPORT Human Clinical

"Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."

Supports the association and the OCCASIONAL band by qualitative-term mapping applied to the craniofacial group.

Long Face OCCASIONAL Long face HP:0000276

Show evidence (2 references)

PMID:35469323 SUPPORT Human Clinical

"Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."

Supports the association and the OCCASIONAL band by qualitative-term mapping applied to the craniofacial group.

PMID:20101266 SUPPORT Human Clinical

"facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"

The classical description of the facial gestalt, which places long face first.

Facial Asymmetry OCCASIONAL Facial asymmetry HP:0000324

Show evidence (1 reference)

PMID:35469323 SUPPORT Human Clinical

"Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."

Supports the association and the OCCASIONAL band by qualitative-term mapping applied to the craniofacial group.

Broad Nasal Tip Broad nasal tip HP:0000455

Show evidence (1 reference)

PMID:20101266 SUPPORT Human Clinical

"facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"

Documents broad nasal tip as part of the reported facial gestalt.

Nervous System 5

Global Developmental Delay VERY_FREQUENT Global developmental delay HP:0001263

Show evidence (1 reference)

PMID:35469323 SUPPORT Human Clinical

"All affected individuals exhibited developmental delay and all but two had borderline to severe ID."

Supports the association and the VERY_FREQUENT band: 16 of 16 individuals in the cohort. VERY_FREQUENT rather than OBLIGATE because ascertainment was largely on developmental delay, which makes 100% partly circular.

Intellectual Disability VERY_FREQUENT Intellectual disability HP:0001249

Show evidence (3 references)

PMID:35469323 SUPPORT Human Clinical

"All affected individuals exhibited developmental delay and all but two had borderline to severe ID."

Supports the association and the VERY_FREQUENT band: 14 of 16 (87.5%) had borderline to severe intellectual disability.

PMID:35469323 PARTIAL Human Clinical

"Of the two who did not have ID, one had dyscalculia and the other had mild learning difficulties."

Documents the two individuals at the mild end who did not meet criteria for intellectual disability, which is why the band is not OBLIGATE.

PMID:10398231 SUPPORT Human Clinical

"A family is described in which X-linked mild to borderline mental retardation (MR) is associated with cleft lip/palate."

The original clinical delineation, describing the mild end of the severity range.

Mild Intellectual Disability Mild intellectual disability HP:0001256

Show evidence (3 references)

PMID:17661819 SUPPORT Human Clinical

"The clinical phenotype of the male patients was characterized by mild MR, mild dysmorphic features, unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling."

Documents mild impairment in the Finnish F279S family.

PMID:19843542 SUPPORT Human Clinical

"The PHF8 missense mutation c.836C>T is associated with mild MR, mild dysmorphic features, and either unilateral or bilateral cleft lip and cleft palate in two male siblings."

Independent restatement of the mild-severity phenotype for the same allele.

PMID:17594395 SUPPORT Human Clinical

"One of the truncating mutations was found in the original family with Siderius-Hamel CL/P syndrome where only two of the three affected individuals had mental retardation (MR) with CL/P and one individual had mild MR."

Documents intrafamilial variability including a mildly affected individual in the original Siderius-Hamel family.

Autistic Behavior FREQUENT Autistic behavior HP:0000729

Show evidence (2 references)

PMID:35469323 SUPPORT Human Clinical

"Autism spectrum disorder and attention deficit hyperactivity disorder, which were not previously emphasized in PHF8-XLID, were frequently observed in affected individuals."

Supports the association and the FREQUENT band by the qualitative-term mapping ("frequently observed"); no numerator is given in the abstract, so the band is not derived from a count.

PMID:18498374 PARTIAL Human Clinical

"We describe two brothers with autistic disorder, intellectual disability (ID) and cleft lip/palate with a microdeletion of Xp11.22 detected through screening individuals with autism spectrum disorders (ASDs) for microdeletions and duplications using 1-Mb resolution array comparative genomic..."

Marked PARTIAL deliberately. These brothers carry a contiguous deletion of PHF8 plus FAM120C and WNK3, and the authors attribute the autism component to the larger deletion rather than to PHF8, so this corroborates the association without attributing autism to PHF8 alone.

Attention Deficit Hyperactivity Disorder FREQUENT Attention deficit hyperactivity disorder HP:0007018

Show evidence (1 reference)

PMID:35469323 SUPPORT Human Clinical

"Autism spectrum disorder and attention deficit hyperactivity disorder, which were not previously emphasized in PHF8-XLID, were frequently observed in affected individuals."

Supports the association and the FREQUENT band by qualitative-term mapping ("frequently observed").

Other 1

Large Hands Large hands HP:0001176

Show evidence (1 reference)

PMID:20101266 PARTIAL Human Clinical

"facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"

Marked PARTIAL: the statement is a secondary narrative summary of the phenotype rather than a primary clinical observation, and the feature is absent from the largest contemporary series.

🧬

Genetic Associations

PHF8 (Causative)

Gene: PHF8 hgnc:20672 variant_origin: GERMLINE

Show evidence (5 references)

PMID:16199551 SUPPORT Human Clinical

"Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)."

Gene-discovery paper establishing the PHF8 gene-disease relationship and binding it to OMIM 300263.

PMID:16199551 SUPPORT Human Clinical

"Expression studies showed that this gene is ubiquitously transcribed, with strong expression of the mouse orthologue Phf8 in embryonic and adult brain structures."

Sources the expression statement in the notes.

PMID:35469323 SUPPORT Human Clinical

"PHF8-XLID is an under-characterized disorder with only five previous reports describing different PHF8 predicted loss-of-function variants in eight individuals."

Quantifies how small the pre-2022 allele series was, which is the basis for the "small reported spectrum" claim.

+ 2 more references

💊

Medical Actions

Surgical Repair of Orofacial Cleft

Action: reconstructive surgery for orofacial cleft Ontology label: Reconstructive Surgery NCIT:C25351

Where a cleft is present, staged surgical repair is the standard of care and is the single most consequential intervention for these patients. It is generic cleft management rather than anything specific to PHF8, and is included because clefting is the defining malformation of the syndrome and because the founding Finnish cohort was assembled from patients operated on for cleft lip/palate. No PHF8-specific surgical outcome data exist.

Show evidence (1 reference)

PMID:17661819 PARTIAL Human Clinical

"the original cohort of 7712 patients operated on since 1 January 1950 for cleft lip/cleft palate in the Cleft Centre at the Helsinki University Hospital"

Marked PARTIAL: this establishes that affected individuals are managed surgically in a cleft centre, but it is not an outcome study and no PHF8-specific surgical evidence exists.

Developmental and Educational Intervention

Action: speech and language therapy Ontology label: Speech Language Therapy NCIT:C159273

Management of the neurodevelopmental phenotype is supportive and identical to that for other causes of developmental delay: early intervention, special education, and speech and language therapy directed at the communication delay. There is no disease-specific evidence base; this is recorded because it is the actual standard of care and because developmental delay was present in every individual in the largest series.

Show evidence (1 reference)

PMID:35469323 PARTIAL Human Clinical

"All affected individuals exhibited developmental delay and all but two had borderline to severe ID."

Marked PARTIAL: this establishes the universal need for developmental intervention but is not evidence for the efficacy of any specific therapy in this disorder.

Genetic Counselling

Action: genetic counseling Ontology label: Genetic Counseling NCIT:C15240

X-linked recessive counselling applies: carrier mothers have a 50% risk of transmitting the allele to each son, affected males transmit the allele to all daughters and no sons, and maternal carrier testing should follow a diagnosis in a proband. The founding pedigrees were multiplex X-linked families identified through linkage, which is the empirical basis for the counselling model.

Show evidence (1 reference)

PMID:10398231 SUPPORT Human Clinical

"A family is described in which X-linked mild to borderline mental retardation (MR) is associated with cleft lip/palate."

Establishes the X-linked familial transmission pattern that the counselling model rests on.

🔀

Differential Diagnoses

Conditions with similar clinical presentations that must be differentiated from Siderius Type X-Linked Intellectual Disability:

Claes-Jensen type syndromic X-linked intellectual disability (KDM5C) MONDO:0010355

Overlapping Features The most important named-entity distinction for this entry, because the two literatures are genuinely entangled rather than merely similarly named. Both are eponymous X-linked syndromic intellectual disability entities caused by JmjC-domain histone demethylases, both were delineated by the same era of X-chromosome mutation screening, and PHF8 and KDM5C/JARID1C are functionally linked in one transcriptional module through ZNF711 - so KDM5C is cited inside PHF8 mechanism papers and vice versa. They are nonetheless distinct MONDO entities with distinct genes, distinct OMIM numbers and distinct clinical profiles: Claes-Jensen is a KDM5C H3K4me3 eraser disorder with short stature, spasticity, hyperreflexia and seizures and is one of the commoner causes of X-linked intellectual disability, whereas Siderius type is a PHF8 H4K20me1/H3K9me1-2 eraser disorder with orofacial clefting and a milder, much rarer picture. Evidence must never be pooled across the two.

Show evidence (2 references)

PMID:20346720 SUPPORT In Vitro

"Taken together, our results functionally link the XLMR gene PHF8 to two other XLMR genes, ZNF711 and JARID1C, indicating that MR genes may be functionally linked in pathways, causing the complex phenotypes observed in patients developing MR."

Documents exactly the functional entanglement that makes this differential necessary: PHF8 and JARID1C/KDM5C share a pathway but cause separate diseases.

PMID:31691806 SUPPORT In Vitro

"Interestingly, mutations in all four genes (KDM5C, ARX, ZNF711 and PHF8) are associated with X-linked NDDs comprising intellectual disability as a core feature."

States explicitly that PHF8 and KDM5C are separate genes causing separate X-linked neurodevelopmental disorders while sharing one regulatory module - the precise reason the two literatures must be kept apart during curation.

Nonsyndromic cleft lip with or without cleft palate Not Yet Curated MONDO:0000358

Overlapping Features Orofacial clefting is one of the commonest birth defects, so the great majority of children presenting with cleft lip/palate do not have this syndrome. What should raise suspicion of PHF8 is the combination of clefting with intellectual disability in a male with an X-linked family history - the Finnish screen found one PHF8 family among 18 such selected patients drawn from a nationwide cleft cohort of 7712.

Show evidence (1 reference)

PMID:19843542 SUPPORT Human Clinical

"The prevalence of cleft lip with or without cleft palate is amongst the most common of all birth defects, averaging at 10.5 per 10,000 live births in the United States"

Quantifies the background clefting rate against which this rare syndrome must be distinguished.

Xp11.22 contiguous gene deletion syndrome

Overlapping Features Deletions at Xp11.22 can remove PHF8 together with neighbouring genes including FAM120C and part of WNK3. The reported brothers with such a microdeletion had the Siderius features plus autism spectrum disorder, hypertelorism and broad halluces, so a deletion-mediated presentation may be broader than the single-gene phenotype and the extra features should not be attributed to PHF8. Chromosomal microarray, not sequencing, is the test that resolves this.

Show evidence (3 references)

PMID:19843542 SUPPORT Human Clinical

"Recently, a microdeletion encompassing all of the PHF8 and FAM120C genes and parts of WNK3 was reported in two brothers with autism spectrum disorders, causing hypertelorism and broad halluces in addition to the Siderius XLMR features"

Documents the contiguous-gene deletion presentation and the extra features it carries.

PMID:18498374 SUPPORT Human Clinical

"Our findings show that in addition to point mutations, a complete deletion of the PHF8 gene is associated with the X-linked mental retardation Siderius-Hamel syndrome (OMIM 300263)"

The primary report of the deletion, and a second independent identity anchor binding PHF8 deletion to OMIM 300263, the OMIM xref of MONDO:0010286.

PMID:18498374 SUPPORT Human Clinical

"further suggest that the larger size of the Xp11.22 deletion including genes FAM120C and WNK3 may be involved in the pathogenesis of autism"

The authors themselves attribute the autism component to the extra deleted genes rather than to PHF8, which is why deletion cases must not be pooled with single-gene cases when curating the PHF8 phenotype.

{ }

Source YAML

click to show

name: Siderius Type X-Linked Intellectual Disability
creation_date: '2026-08-01T00:00:00Z'
category: Genetic
classifications:
  harrisons_chapter:
  - classification_value: NEUROLOGIC
    evidence:
    - reference: PMID:35469323
      reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
        disorders and facial dysmorphology.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: "All affected individuals exhibited developmental delay and all but two had borderline to severe ID."
      explanation: A neurodevelopmental phenotype in every affected individual
        supports classification under the neurologic chapter.
  - classification_value: GENETICS_ENVIRONMENT_DISEASE
    evidence:
    - reference: PMID:16199551
      reference_title: Mutations in PHF8 are associated with X linked mental retardation
        and cleft lip/cleft palate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: "Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)."
      explanation: A single-gene germline aetiology supports classification as a
        genetic disease.
parents:
- Neurodevelopmental Disorder
- X-linked Disorder
- Chromatinopathy
disease_term:
  preferred_term: syndromic X-linked intellectual disability Siderius type
  term:
    id: MONDO:0010286
    label: syndromic X-linked intellectual disability Siderius type
synonyms:
- MRXSSD
- Siderius X-linked intellectual disability syndrome
- Siderius-Hamel syndrome
- Siderius X-linked mental retardation syndrome
- PHF8-XLID
- X-linked intellectual disability Hamel type
- intellectual disability syndrome, X-linked, Siderius type
description: >
  Siderius type syndromic X-linked intellectual disability (MRXSSD; OMIM 300263;
  ORPHA:85287) is a rare X-linked chromatinopathy caused by loss-of-function
  variants in PHF8 at Xp11.22. PHF8 encodes a PHD-finger plus JmjC-domain,
  Fe(II)/2-oxoglutarate-dependent histone demethylase that erases the repressive
  marks H4K20me1 and H3K9me1/me2 at promoters to which it is recruited by
  PHD-finger recognition of H3K4me3 and by direct contact with RNA polymerase II.
  Hemizygous males present with developmental delay and borderline to severe
  intellectual disability, mild facial dysmorphism, and — in a minority of
  contemporary cases — orofacial clefting, the feature on which the syndrome was
  originally delineated. Autism spectrum disorder and attention deficit
  hyperactivity disorder were added to the phenotype by the largest published
  series.

  The entity has a two-phase literature. The founding descriptions
  (Siderius 1999; Laumonnier 2005; Abidi 2007; Koivisto 2007) were ascertained
  through cleft clinics and multiplex X-linked pedigrees, so cleft lip/palate
  looked obligate. The 2022 multicentre series of 16 additional individuals,
  ascertained largely by exome sequencing for developmental delay, found
  orofacial clefting in only 3 of 16 and explicitly reported that the feature is
  less common than previously reported. Both ascertainment frames are curated
  here, with the frequency bands taken from the contemporary series.

  IMPORTANT SCOPE NOTE - this entity is defined by PHF8. It must not be conflated
  with (a) PHF6-related Borjeson-Forssman-Lehmann syndrome or PHF21A-related
  disorders, which are different PHD-finger genes with different diseases;
  (b) the other eponymous X-linked intellectual disability types, in particular
  Claes-Jensen type (KDM5C) and Lubs type (MECP2 duplication), which are
  separate MONDO entities with separate genes - note that the PHF8 and KDM5C
  literatures are genuinely entangled because PHF8, ZNF711 and KDM5C/JARID1C sit
  in one transcriptional module, but the diseases are distinct; or (c) the large
  somatic-cancer PHF8 literature, which is a different disease context entirely.
references:
- reference: PMID:35469323
  title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental disorders
    and facial dysmorphology.
  findings: []
- reference: PMID:16199551
  title: Mutations in PHF8 are associated with X linked mental retardation and cleft
    lip/cleft palate.
  findings: []
- reference: PMID:10398231
  title: X-linked mental retardation associated with cleft lip/palate maps to Xp11.3-q21.3.
  findings: []
notes: >
  GeneReviews baseline: searched PubMed on 2026-08-01 with
  `GeneReviews[All Fields] AND (PHF8 OR Siderius)` (0 hits) and
  `Siderius[TI] GeneReviews[TI]` (0 hits). No GeneReviews chapter exists for
  Siderius-Hamel syndrome / PHF8-related X-linked intellectual disability. In the
  absence of a GeneReviews baseline the phenotype list here is anchored on the
  single largest systematic series (PMID:35469323, 16 newly reported individuals
  from 11 families plus review of the 8 previously reported) together with the
  four founding clinical reports (PMID:10398231, PMID:16199551, PMID:17594395,
  PMID:17661819).

  Frequency derivations, all from the 16-individual cohort in PMID:35469323
  unless stated: developmental delay "All affected individuals" = 16/16 (100%,
  VERY_FREQUENT); intellectual disability "all but two had borderline to severe
  ID" = 14/16 (87.5%, VERY_FREQUENT); orofacial clefting "seen in three
  individuals from our cohort" = 3/16 (18.8%, OCCASIONAL); autism spectrum
  disorder and ADHD "frequently observed" (qualitative, mapped to FREQUENT);
  the individual craniofacial findings are reported collectively as occurring
  "in some affected individuals", which maps to OCCASIONAL and is applied to the
  group rather than to each dysmorphism separately. Where the source is
  qualitative and does not license even a coarse band, `frequency:` is omitted.

  Known gap in phenotype coverage. The deep-research report surfaced a
  per-feature frequency table from the PMC full text of PMID:35469323 -
  retrognathia 10/16, infantile feeding difficulty 10/16, high-arched palate
  6/16, seizures 5/16, low-set ears 5/16, gross motor delay 12/16, fine motor
  delay 14/16, speech delay 16/16, and neuroimaging findings including
  polymicrogyria in one twin pair. Only the abstract of that paper is cached by
  `just fetch-reference`, so none of those counts can be verified as a byte-exact
  snippet and none of those phenotypes is curated here. This is a deliberate
  omission, not an oversight: a smaller correct entry is preferred. Anyone
  extending this entry should obtain the full text and add them with verified
  quotes.

  Structured-source citations (ORPHA:85287 for the Orphanet prevalence class,
  and any ClinGen gene-disease validity or dosage record for PHF8) are NOT
  included: neither is present in `references_cache/`, and generating them would
  require bumping a pinned data MANIFEST, which is out of scope for a curation PR
  and is blocked pending issue #7622. No population prevalence estimate is
  recorded for the same reason - the literature reports only case counts, and no
  numerator/denominator that would license a rate is available.

  Deliberately excluded: the extensive somatic-cancer PHF8 literature (breast,
  hepatocellular, triple-negative breast, gastric), which concerns a different
  disease context; and the PHF8 depression/anxiety-resistance mouse work
  (PMID:28485378), which reports a behavioural phenotype not documented in
  patients with this syndrome.
pathophysiology:
- name: Pathogenic PHF8 Loss-of-Function Variant
  biological_scale: MOLECULAR
  description: >
    The shared upstream lesion is a hemizygous loss-of-function variant in PHF8
    at Xp11.22. The reported allele spectrum is small but mechanistically
    coherent: nonsense and frameshift alleles that truncate the protein before or
    within the JmjC domain and remove the nuclear localisation signals, splice
    alleles, missense alleles clustered in the JmjC-encoding exons, and whole-gene
    microdeletions. All reported mechanisms converge on loss of function; no
    dominant-negative or gain-of-function mechanism has been demonstrated for the
    germline neurodevelopmental phenotype. Two consequences fan out from this node
    in parallel rather than in series - loss of JmjC catalysis, and, for the
    truncating alleles, loss of the PHD-finger/nuclear-targeting module - because
    the F279S missense allele abolishes catalysis while leaving the N-terminal PHD
    finger intact, and the K177X nonsense allele removes both.
  gene:
    preferred_term: PHF8
    term:
      id: hgnc:20672
      label: PHF8
  evidence:
  - reference: PMID:16199551
    reference_title: Mutations in PHF8 are associated with X linked mental retardation
      and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)."
    explanation: >
      The gene-discovery paper, and the strongest available identity anchor - it
      binds PHF8 directly to MIM 300263, which is the OMIM xref of MONDO:0010286.
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Loss-of-function variants in PHD Finger Protein 8 (PHF8) cause Siderius X-linked intellectual disability (ID) syndrome, hereafter called PHF8-XLID."
    explanation: States the loss-of-function mechanism and names the entity being
      curated here.
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Clinically observed mutations to the PHF8 gene cluster in exons encoding for the double stranded beta-helix fold and will therefore disrupt catalytic activity."
    explanation: >
      Establishes that the clinical allele spectrum is concentrated in the
      catalytic fold, which is why the catalytic arm is the dominant downstream
      consequence of this node.
  downstream:
  - target: Loss of JmjC-Domain Histone Demethylase Activity
  - target: Loss of PHD-Finger H3K4me3 Reading and Nuclear Targeting
- name: Loss of JmjC-Domain Histone Demethylase Activity
  biological_scale: MOLECULAR
  description: >
    PHF8 is an Fe(II)- and 2-oxoglutarate-dependent oxygenase of the
    double-stranded beta-helix class. Its JmjC domain removes mono- and dimethyl marks
    from lysine but not trimethyl marks. Patient alleles clustered in the
    JmjC-encoding exons abolish this catalysis: the F279S missense variant, which
    substitutes a residue in a conserved hydrophobic region of the JmjC domain, is
    catalytically inactive against both peptide and intact-histone substrates,
    and truncating alleles remove the domain outright. Because the reaction
    requires molecular oxygen, PHF8 activity is oxygen-dependent, a property that
    has been proposed - though not demonstrated in patients - to link the
    catalytic arm to the known association between maternal hypoxia and cleft
    lip in mouse studies.
  molecular_functions:
  - preferred_term: histone H4K20 demethylase activity
    term:
      id: GO:0035575
      label: histone H4K20 demethylase activity
    modifier: DECREASED
  - preferred_term: histone H3K9 demethylase activity
    term:
      id: GO:0032454
      label: histone H3K9 demethylase activity
    modifier: DECREASED
  evidence:
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "We report that recombinant PHF8 is an Fe(II) and 2-oxoglutarate-dependent N(epsilon)-methyl lysine demethylase, which acts on histone substrates."
    explanation: Defines the enzyme class and cofactor dependence of the catalytic
      arm.
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "This mutant encodes a F279S variant of PHF8 that modifies a conserved hydrophobic region; assays with both peptides and intact histones reveal this variant to be catalytically inactive."
    explanation: Direct demonstration that a patient missense allele abolishes
      catalysis.
  - reference: PMID:20101266
    reference_title: Structural insights into a novel histone demethylase PHF8.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "The in vitro demethylation assay also showed that the F279S mutant observed in clinical patients possesses no demethylation activity, suggesting that loss of enzymatic activity is crucial for pathogenesis of PHF8 patients."
    explanation: Independent replication of catalytic inactivity, with the authors
      explicitly assigning pathogenic weight to the catalytic loss.
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain
      and craniofacial development.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Importantly, patient mutations significantly compromised PHF8 catalytic function."
    explanation: Third independent line of evidence that patient alleles are
      catalytically defective.
  downstream:
  - target: Persistence of Repressive H4K20me1 at Target Promoters
  - target: Persistence of Repressive H3K9me1/me2 Marks
- name: Loss of PHD-Finger H3K4me3 Reading and Nuclear Targeting
  biological_scale: MOLECULAR
  description: >
    The N-terminal PHD finger of PHF8 binds H3K4me3/me2, the mark that decorates
    active transcription start sites, and this reader function is what recruits
    the enzyme to the promoters it acts on. Truncating patient alleles remove not
    only the JmjC domain but also the nuclear localisation signals, so the
    residual protein cannot reach chromatin at all. This is modelled as a sibling
    of the catalytic arm rather than a downstream consequence, because the F279S
    catalytic-dead allele leaves the PHD finger intact while the K177X nonsense
    allele removes both modules - the two arms are separable in the patient
    allele series.
  molecular_functions:
  - preferred_term: histone H3K4me3 reader activity
    term:
      id: GO:0140002
      label: histone H3K4me3 reader activity
    modifier: DECREASED
  evidence:
  - reference: PMID:20421419
    reference_title: PHF8 targets histone methylation and RNA polymerase II to activate
      transcription.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Our biochemical analysis revealed specific association of the PHF8 PHD with histone H3 trimethylated at lysine 4 (H3K4me3)."
    explanation: Establishes the PHD finger as an H3K4me3 reader module.
  - reference: PMID:20548336
    reference_title: The X-linked mental retardation gene PHF8 is a histone demethylase
      involved in neuronal differentiation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "In addition, we show that PHF8 binds specifically to H3K4me3/2 peptides via an N-terminal PHD finger domain."
    explanation: Independent confirmation of the reader specificity.
  - reference: PMID:17594395
    reference_title: A novel mutation in the PHF8 gene is associated with X-linked
      mental retardation with cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The mutation results in a truncated PHF8 protein lacking the Jumonji-like C terminus domain and five nuclear localization signals."
    explanation: >
      Documents that a patient nonsense allele removes the nuclear localisation
      signals in addition to the catalytic domain, which is the nuclear-targeting
      component of this node.
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "In contrast to the wild-type HA-PHF8, analogous studies revealed that the clinically observed F279S variant did not show clear nuclear localization, with apparent cytoplasmic localization."
    explanation: >
      Shows that even the F279S missense allele mislocalises out of the nucleus,
      so nuclear targeting is compromised by more than just the truncating
      alleles. This is why the node is framed as reading plus nuclear targeting.
  downstream:
  - target: Failure of RNA Polymerase II-Coupled Transcriptional Coactivation
- name: Persistence of Repressive H4K20me1 at Target Promoters
  biological_scale: MOLECULAR
  description: >
    PHF8 is the founding H4K20me1 demethylase. When catalysis is lost, H4K20me1
    accumulates at the transcription start sites of PHF8 target genes and the
    genes stay repressed. Two H4K20me1-dependent target sets have been mapped
    directly: cytoskeletal and cell-adhesion regulators (RHOA, RAC1, GSK3B) whose
    promoters PHF8 demethylates, and RSK1, whose derepression is the entry point
    into the mTOR arm. In astrocytes the related H4K20me3 mark accumulates at
    synaptogenic genes when PHF8 is depleted, extending the same
    heterochromatinisation logic to glia.
  biological_processes:
  - preferred_term: Chromatin organization
    term:
      id: GO:0006325
      label: chromatin organization
    modifier: ABNORMAL
  evidence:
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain
      and craniofacial development.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Here we provide multiple lines of evidence establishing PHF8 as the first mono-methyl histone H4 lysine 20 (H4K20me1) demethylase, with additional activities towards histone H3K9me1 and me2."
    explanation: Establishes H4K20me1 as a principal PHF8 substrate.
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain
      and craniofacial development.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "PHF8 depletion resulted in upregulation of H4K20me1 and H3K9me1 at the TSS and H3K9me2 in the non-TSS sites, respectively, demonstrating differential substrate specificities at different target locations."
    explanation: Direct demonstration that losing PHF8 causes the repressive marks
      to persist, and that the two marks accumulate at different genomic locations.
  - reference: PMID:22850744
    reference_title: The histone demethylase PHF8 is essential for cytoskeleton dynamics.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Our data demonstrate that PHF8 directly regulates the expression of these genes by demethylating H4K20me1 at promoters."
    explanation: Maps a specific H4K20me1-dependent target set (cytoskeletal and
      adhesion genes) to promoter demethylation by PHF8.
  downstream:
  - target: Failure of RNA Polymerase II-Coupled Transcriptional Coactivation
  - target: RSK1 Derepression and mTOR Pathway Hyperactivation
- name: Persistence of Repressive H3K9me1/me2 Marks
  biological_scale: MOLECULAR
  description: >
    PHF8 also erases mono- and dimethylated H3K9, a repressive mark associated
    with facultative heterochromatin. Recombinant PHF8 is selective for H3K9me2
    and H3K9me1 and does not accept H3K9me3, so the enzyme trims the repressive
    mark toward the unmethylated state rather than resolving constitutive
    heterochromatin. In cells, PHF8 depletion raises H3K9me1 at transcription
    start sites and H3K9me2 at non-TSS sites. Loss of this activity is one of the
    two routes by which target promoters fail to be activated.
  molecular_functions:
  - preferred_term: histone H3K9me/H3K9me2 demethylase activity
    term:
      id: GO:0140683
      label: histone H3K9me/H3K9me2 demethylase activity
    modifier: DECREASED
  evidence:
  - reference: PMID:20101266
    reference_title: Structural insights into a novel histone demethylase PHF8.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Biochemical and structural studies reveal that PHF8 is a novel histone demethylase specific for di- and mono-methylated histone H3 lysine 9 (H3K9me2/1), but not for H3K9me3."
    explanation: Defines the H3K9 substrate specificity, including the exclusion of
      the trimethyl state.
  - reference: PMID:20346720
    reference_title: A functional link between the histone demethylase PHF8 and the
      transcription factor ZNF711 in X-linked mental retardation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Here, we show that the XLMR protein PHF8 and a C. elegans homolog F29B9.2 catalyze demethylation of di- and monomethylated lysine 9 of histone H3 (H3K9me2/me1)."
    explanation: Independent confirmation of the H3K9me2/me1 demethylase activity,
      with conservation to an invertebrate orthologue.
  - reference: PMID:20421419
    reference_title: PHF8 targets histone methylation and RNA polymerase II to activate
      transcription.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Here, we show that PHF8 is a histone demethylase that removes repressive histone H3 dimethyl lysine 9 marks."
    explanation: Third independent demonstration of the H3K9me2 eraser activity.
  downstream:
  - target: Failure of RNA Polymerase II-Coupled Transcriptional Coactivation
- name: Failure of RNA Polymerase II-Coupled Transcriptional Coactivation
  biological_scale: MOLECULAR
  description: >
    The convergence point of the chromatin arms. PHF8 occupies the transcription
    start sites of thousands of active or poised genes, mirroring the distribution
    of RNA polymerase II and H3K4me3-marked nucleosomes, and contacts the RNAPII
    C-terminal domain directly; it therefore acts as a promoter-resident
    transcriptional coactivator whose activation function depends on PHD binding
    to H3K4me3. Patient alleles are defective in coactivation as well as in
    demethylation. PHF8 additionally partners with the XLID transcription factor
    ZNF711 at a shared target set, and co-occupies promoters with the REST/NRSF
    repressor, so the net transcriptional output at any locus can be activating or
    repressive; the dominant documented effect is loss of activation. Downstream
    of this node the disease diversifies into distinct neural and craniofacial
    cell-type programmes.
  molecular_functions:
  - preferred_term: transcription coactivator activity
    term:
      id: GO:0003713
      label: transcription coactivator activity
    modifier: DECREASED
  biological_processes:
  - preferred_term: Regulation of transcription by RNA polymerase II
    term:
      id: GO:0006357
      label: regulation of transcription by RNA polymerase II
    modifier: ABNORMAL
  evidence:
  - reference: PMID:20421419
    reference_title: PHF8 targets histone methylation and RNA polymerase II to activate
      transcription.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Chromatin immunoprecipitation followed by high-throughput sequencing indicated that PHF8 is enriched at the transcription start sites of many active or poised genes, mirroring the presence of RNA polymerase II (RNAPII) and of H3K4me3-bearing nucleosomes."
    explanation: Places PHF8 physically at active promoters together with RNAPII.
  - reference: PMID:20421419
    reference_title: PHF8 targets histone methylation and RNA polymerase II to activate
      transcription.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Importantly, a PHF8 disease mutant was defective in demethylation and in coactivation."
    explanation: Directly links a patient allele to failure of the coactivation
      function, which is the claim of this node.
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain
      and craniofacial development.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "PHF8 positively regulates gene expression, which is dependent on its H3K4me3-binding PHD and catalytic domains."
    explanation: Shows that the activating output requires both upstream arms, which
      is why both converge on this node.
  - reference: PMID:20346720
    reference_title: A functional link between the histone demethylase PHF8 and the
      transcription factor ZNF711 in X-linked mental retardation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Furthermore, PHF8 interacts with another XMLR protein, ZNF711, which binds to a subset of PHF8 target genes, including the XLMR gene JARID1C."
    explanation: >
      Documents the ZNF711 partnership. Note for scope: JARID1C is KDM5C, the gene
      of the separate Claes-Jensen entity - the two genes are functionally linked
      but the diseases are distinct.
  - reference: PMID:24852203
    reference_title: PHF8 and REST/NRSF co-occupy gene promoters to regulate proximal
      gene expression.
    supports: PARTIAL
    evidence_source: COMPUTATIONAL
    snippet: "Our analysis suggested that PHF8 not only activates but may also repress gene expression."
    explanation: >
      Qualifies the node - a reanalysis of published ChIP-seq indicates the
      transcriptional output is bidirectional, so "loss of activation" is the
      dominant but not the exclusive consequence.
  - reference: PMID:20548336
    reference_title: The X-linked mental retardation gene PHF8 is a histone demethylase
      involved in neuronal differentiation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Furthermore, we show that PHF8 interacts with RARalpha and functions as a coactivator for RARalpha."
    explanation: >
      Identifies a specific nuclear-receptor partner through which the
      coactivation function is exercised, and the one that connects this node to
      the retinoic-acid-dependent neuronal differentiation arm downstream.
  - reference: PMID:20208542
    reference_title: PHF8 activates transcription of rRNA genes through H3K4me3 binding
      and H3K9me1/2 demethylation.
    supports: PARTIAL
    evidence_source: IN_VITRO
    snippet: "PHF8 activates transcription of rRNA genes through H3K4me3 binding and H3K9me1/2 demethylation"
    explanation: >
      Extends the same reader-plus-eraser coactivation logic to RNA polymerase I
      transcription of rDNA. Marked PARTIAL because the nucleolar arm has no
      established connection to the clinical phenotype.
  downstream:
  - target: Impaired Neuronal Differentiation and Neurite Outgrowth
  - target: Impaired Astrocyte Differentiation and Synaptic Gene Silencing
  - target: Perturbed Oligodendroglial Development
  - target: Impaired Serine Biosynthesis in Neural Progenitors
  - target: Impaired Neural Crest and Craniofacial Midline Development
- name: RSK1 Derepression and mTOR Pathway Hyperactivation
  biological_scale: MOLECULAR
  description: >
    A specific, pharmacologically actionable branch of the H4K20me1 arm. In Phf8
    knockout mice, failure to demethylate H4K20me1 at the RSK1 locus derepresses
    RSK1, and the mTOR signalling pathway becomes hyperactive in hippocampus.
    This is the one node in the entry with a demonstrated rescue: rapamycin
    normalises both long-term potentiation and cognitive performance in the
    knockout. The evidence is entirely from the mouse; no patient has been
    treated on this rationale.
  biological_processes:
  - preferred_term: TOR signaling
    term:
      id: GO:0031929
      label: TOR signaling
    modifier: INCREASED
  evidence:
  - reference: PMID:29317619
    reference_title: Phf8 histone demethylase deficiency causes cognitive impairments
      through the mTOR pathway.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "We also show that mTOR signaling pathway is hyperactive in hippocampus in Phf8 knockout mouse."
    explanation: Establishes mTOR hyperactivation as a consequence of Phf8 loss in
      vivo.
  - reference: PMID:29317619
    reference_title: Phf8 histone demethylase deficiency causes cognitive impairments
      through the mTOR pathway.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Mechanistically, we show that demethylation of H4K20me1 by Phf8 results in transcriptional suppression of RSK1 and homeostasis of mTOR signaling."
    explanation: Supplies the H4K20me1-to-RSK1-to-mTOR mechanism that makes this a
      child of the H4K20me1 node rather than a free-standing observation.
  downstream:
  - target: Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation
- name: Impaired Neuronal Differentiation and Neurite Outgrowth
  biological_scale: CELLULAR
  description: >
    PHF8 is required for neuronal differentiation and for the cytoskeletal
    programme that builds neurites. Knockdown in P19 embryonal carcinoma cells
    impairs retinoic-acid-induced neuronal differentiation, and overexpression of
    wild-type but not the catalytically dead F279S variant pushes the cells toward
    a neuronal fate - a clean genotype-to-cell-phenotype link through a patient
    allele. In differentiated neurons, PHF8 depletion downregulates cytoskeletal
    genes and produces deficient neurite outgrowth, which the authors propose is
    the cellular substrate of the cognitive phenotype.
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  biological_processes:
  - preferred_term: Neuron differentiation
    term:
      id: GO:0030182
      label: neuron differentiation
    modifier: DECREASED
  - preferred_term: Neuron projection development
    term:
      id: GO:0031175
      label: neuron projection development
    modifier: DECREASED
  evidence:
  - reference: PMID:20548336
    reference_title: The X-linked mental retardation gene PHF8 is a histone demethylase
      involved in neuronal differentiation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Consistent with a role for PHF8 in neuronal differentiation, knockdown of PHF8 in mouse embryonic carcinoma P19 cells impairs RA-induced neuronal differentiation, whereas overexpression of the wild-type but not the F279S mutant PHF8 drives P19 cells toward neuronal differentiation."
    explanation: >
      Links the patient F279S allele specifically to failure of neuronal
      differentiation, connecting the molecular arms to a cellular phenotype.
  - reference: PMID:22850744
    reference_title: The histone demethylase PHF8 is essential for cytoskeleton dynamics.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Further analysis in neurons shows that depletion of PHF8 results in down-regulation of cytoskeleton genes and leads to a deficient neurite outgrowth."
    explanation: Documents the neurite-outgrowth defect that gives this node its
      second half.
  - reference: PMID:22850744
    reference_title: The histone demethylase PHF8 is essential for cytoskeleton dynamics.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Overall, our results suggest that the mental retardation phenotype associated with loss of function of PHF8 could be due to abnormal neuronal connections as a result of alterations in cytoskeleton function."
    explanation: The authors' own statement of the proposed cell-to-organism link,
      cited as a proposal rather than a demonstration.
  downstream:
  - target: Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation
- name: Impaired Astrocyte Differentiation and Synaptic Gene Silencing
  biological_scale: CELLULAR
  description: >
    A non-cell-autonomous arm. PHF8 acts in astrocytes through a regulatory
    crosstalk with Notch signalling that balances expression of the master
    astrocytic transcription factor Nfia, and it keeps synaptogenic genes
    accessible by holding H4K20me3 low. Depleting PHF8 in astrocytes alone is
    sufficient to disrupt neuronal synapse formation and maturation in vitro,
    so part of the synaptic phenotype originates outside the neuron. Evidence is
    from mouse astrocytic culture.
  cell_types:
  - preferred_term: astrocyte
    term:
      id: CL:0000127
      label: astrocyte
  biological_processes:
  - preferred_term: Astrocyte differentiation
    term:
      id: GO:0048708
      label: astrocyte differentiation
    modifier: ABNORMAL
  evidence:
  - reference: PMID:34081130
    reference_title: The histone demethylase PHF8 regulates astrocyte differentiation
      and function.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Using genome-wide analyses and biochemical assays in mouse astrocytic cultures, we reveal a regulatory crosstalk between PHF8 and the Notch signaling pathway that balances the expression of the master astrocytic gene Nfia."
    explanation: Establishes the astrocyte-differentiation arm and its Notch/Nfia
      mechanism.
  - reference: PMID:34081130
    reference_title: The histone demethylase PHF8 regulates astrocyte differentiation
      and function.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Accordingly, astrocytic-PHF8 depletion has a striking effect on neuronal synapse formation and maturation in vitro."
    explanation: Demonstrates the non-cell-autonomous effect on neuronal synapses,
      which is why this node feeds the synaptic-plasticity node.
  - reference: PMID:34081130
    reference_title: The histone demethylase PHF8 regulates astrocyte differentiation
      and function.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Moreover, PHF8 regulates key synaptic genes in astrocytes by maintaining low levels of H4K20me3."
    explanation: Supplies the chromatin mechanism connecting this node back to the
      H4K20 methylation arm.
  downstream:
  - target: Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation
- name: Perturbed Oligodendroglial Development
  biological_scale: CELLULAR
  description: >
    A third glial arm, and the least clinically anchored of the three. Phf8
    promotes proliferation of oligodendrocyte progenitor cells and restrains
    their differentiation into oligodendrocytes, acting largely through Olig2,
    whose regulatory regions it occupies; ectopic Olig2 rescues the proliferation
    defect in Phf8-deficient cells. Notably, human iPSC-derived oligodendrocyte
    generation did not require PHF8 when Olig2 was force-expressed, so the human
    requirement is conditional. No white-matter abnormality has been documented in
    patients with this syndrome, so this node is marked provisional.
  mechanism_confidence: PROVISIONAL
  cell_types:
  - preferred_term: oligodendrocyte precursor cell
    term:
      id: CL:0002453
      label: oligodendrocyte precursor cell
  biological_processes:
  - preferred_term: Oligodendrocyte differentiation
    term:
      id: GO:0048709
      label: oligodendrocyte differentiation
    modifier: ABNORMAL
  evidence:
  - reference: PMID:38613395
    reference_title: Transcription factor Olig2 is a major downstream effector of histone
      demethylase Phf8 during oligodendroglial development.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we show that Phf8 promotes the proliferation of rodent oligodendrocyte progenitor cells and impairs their differentiation to oligodendrocytes"
    explanation: States the oligodendroglial phenotype of Phf8 loss and gain.
  - reference: PMID:38613395
    reference_title: Transcription factor Olig2 is a major downstream effector of histone
      demethylase Phf8 during oligodendroglial development.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Taking the influence of Olig2 levels on oligodendroglial proliferation and differentiation into account, Olig2 likely acts as an important downstream effector of Phf8 in these cells."
    explanation: Supplies the Olig2 effector mechanism.
  - reference: PMID:38613395
    reference_title: Transcription factor Olig2 is a major downstream effector of histone
      demethylase Phf8 during oligodendroglial development.
    supports: PARTIAL
    evidence_source: IN_VITRO
    snippet: "Additionally, generation of human oligodendrocytes from induced pluripotent stem cells did not require PHF8 in a system that relies on forced expression of Olig2 during oligodendroglial induction."
    explanation: >
      The authors' own negative result in a human system, which is the reason this
      node carries PROVISIONAL confidence rather than being asserted as established
      human pathophysiology.
  downstream:
  - target: Impaired Cognitive Development
- name: Impaired Serine Biosynthesis in Neural Progenitors
  biological_scale: CELLULAR
  description: >
    The most recently described arm, and the one that connects chromatin
    regulation to metabolism. In neural stem cells PHF8 drives the serine
    biosynthesis pathway by tuning chromatin accessibility at the promoters of
    metabolic genes, safeguarding the intracellular serine pool that progenitor
    proliferation depends on. Loss of PHF8 disrupts amino-acid metabolism, blocks
    autophagy and impairs vesicle formation, and culminates in replication
    defects, DNA damage and proliferation arrest; in mouse embryos PHF8 deficiency
    halts progenitor expansion and neuron generation in the developing brain.
    Evidence is from neural stem cells and mouse embryos, not from patients.
  cell_types:
  - preferred_term: neural progenitor cell
    term:
      id: CL:0011020
      label: neural progenitor cell
  biological_processes:
  - preferred_term: L-serine biosynthetic process
    term:
      id: GO:0006564
      label: L-serine biosynthetic process
    modifier: DECREASED
  evidence:
  - reference: PMID:41714361
    reference_title: Epigenetic regulation of serine biosynthesis by PHF8 during neurogenesis.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we identify PHF8 as a key driver of the serine biosynthesis pathway, safeguarding the intracellular serine pool essential for neural progenitor proliferation"
    explanation: Establishes the serine-biosynthesis mechanism in neural stem cells.
  - reference: PMID:41714361
    reference_title: Epigenetic regulation of serine biosynthesis by PHF8 during neurogenesis.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "In vivo, PHF8 deficiency in mouse embryos halts neurogenesis, progenitor expansion, and neuron generation in the developing brain."
    explanation: Extends the progenitor-proliferation defect to the intact developing
      brain, which is what licenses the edge to impaired cognitive development.
  - reference: PMID:41714361
    reference_title: Epigenetic regulation of serine biosynthesis by PHF8 during neurogenesis.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "PHF8 fine-tunes chromatin accessibility at promoters of metabolic genes, ensuring their activation during development."
    explanation: Supplies the chromatin-to-metabolism link that makes this a child of
      the transcriptional-coactivation node.
  downstream:
  - target: Impaired Cognitive Development
- name: Impaired Neural Crest and Craniofacial Midline Development
  biological_scale: TISSUE
  description: >
    The craniofacial arm. PHF8 is required for jaw development in zebrafish, where
    it acts in part by directly regulating the homeodomain transcription factor
    MSX1/MSXB, an integrator of several craniofacial signalling pathways; Phf8 is
    independently implicated in neural crest defects in mice and humans. The
    clinical inference of a midline-formation role predates the molecular work and
    was drawn directly from the co-segregation of intellectual disability with
    cleft lip/palate in the original families. Because PHF8 catalysis consumes
    molecular oxygen, an oxygen-sensing link to the known maternal-hypoxia effect
    on murine cleft lip has been proposed, but this remains a hypothesis.
  cell_types:
  - preferred_term: migratory neural crest cell
    term:
      id: CL:0000333
      label: migratory neural crest cell
  biological_processes:
  - preferred_term: Embryonic cranial skeleton morphogenesis
    term:
      id: GO:0048701
      label: embryonic cranial skeleton morphogenesis
    modifier: ABNORMAL
  - preferred_term: Roof of mouth development
    term:
      id: GO:0060021
      label: roof of mouth development
    modifier: ABNORMAL
  evidence:
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain
      and craniofacial development.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "PHF8 regulates cell survival in the zebrafish brain and jaw development, thus providing a potentially relevant biological context for understanding the clinical symptoms associated with PHF8 patients."
    explanation: The primary animal-model evidence for a craniofacial requirement.
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain
      and craniofacial development.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Lastly, genetic and molecular evidence supports a model whereby PHF8 regulates zebrafish neuronal cell survival and jaw development in part by directly regulating the expression of the homeodomain transcription factor MSX1/MSXB, which functions downstream of multiple signalling and developmental pathways."
    explanation: Supplies the MSX1/MSXB effector that connects the transcriptional
      node to craniofacial morphogenesis.
  - reference: PMID:38613395
    reference_title: Transcription factor Olig2 is a major downstream effector of histone
      demethylase Phf8 during oligodendroglial development.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "The plant homeodomain finger protein Phf8 is a histone demethylase implicated by mutation in mice and humans in neural crest defects and neurodevelopmental disturbances."
    explanation: Independent statement placing PHF8 in neural crest development,
      supporting the cell-type annotation on this node.
  - reference: PMID:16199551
    reference_title: Mutations in PHF8 are associated with X linked mental retardation
      and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The association of XLMR and cleft lip/palate in these patients with mutations in PHF8 suggests an important function of PHF8 in midline formation and in the development of cognitive abilities, and links this gene to XLMR associated with cleft lip/palate."
    explanation: The human-genetic inference of a midline-formation role, made before
      any of the model-organism work.
  downstream:
  - target: Orofacial Clefting and Craniofacial Dysmorphology
- name: Impaired Synaptic Plasticity and Hippocampal Long-Term Potentiation
  biological_scale: CELLULAR
  description: >
    The convergence point of the neuronal, astrocytic and mTOR arms. Phf8
    knockout mice show impaired hippocampal long-term potentiation together with
    impaired learning and memory, and - importantly for interpreting the human
    phenotype - without gross morphological brain defects, matching the absence of
    a consistent structural brain abnormality in patients. The synaptic deficit is
    reversible in the mouse: rapamycin restores the weakened LTP.
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  biological_processes:
  - preferred_term: Long-term synaptic potentiation
    term:
      id: GO:0060291
      label: long-term synaptic potentiation
    modifier: DECREASED
  evidence:
  - reference: PMID:29317619
    reference_title: Phf8 histone demethylase deficiency causes cognitive impairments
      through the mTOR pathway.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Here we report that Phf8 knockout mice displayed impaired learning and memory, and impaired hippocampal long-term potentiation (LTP) without gross morphological defects."
    explanation: Establishes the LTP deficit and the absence of gross structural
      brain change.
  - reference: PMID:29317619
    reference_title: Phf8 histone demethylase deficiency causes cognitive impairments
      through the mTOR pathway.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Pharmacological suppression of mTOR signaling with rapamycin in Phf8 knockout mice recovers the weakened LTP and cognitive deficits."
    explanation: Demonstrates reversibility, which is the basis of the mTOR-targeting
      research hypothesis recorded under mechanistic_hypotheses.
  downstream:
  - target: Impaired Cognitive Development
- name: Impaired Cognitive Development
  biological_scale: ORGANISM
  description: >
    The terminal neurodevelopmental output: developmental delay in essentially all
    affected individuals and borderline to severe intellectual disability in the
    large majority, with a behavioural profile that includes autism spectrum
    disorder and attention deficit hyperactivity disorder. The severity range is
    wide enough that two individuals in the largest cohort had no intellectual
    disability at all, one with dyscalculia and one with mild learning
    difficulties.
  biological_processes:
  - preferred_term: Learning or memory
    term:
      id: GO:0007611
      label: learning or memory
    modifier: DECREASED
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "All affected individuals exhibited developmental delay and all but two had borderline to severe ID."
    explanation: The quantitative human anchor for this terminal node.
  - reference: PMID:29317619
    reference_title: Phf8 histone demethylase deficiency causes cognitive impairments
      through the mTOR pathway.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Together, our results indicate that loss of Phf8 in animals causes deficient learning and memory by epigenetic disruption of mTOR signaling"
    explanation: Model-organism recapitulation of the cognitive endpoint, kept
      distinct from the human evidence by evidence_source.
- name: Orofacial Clefting and Craniofacial Dysmorphology
  biological_scale: ORGANISM
  description: >
    The terminal craniofacial output: cleft lip and/or cleft palate, unilateral or
    bilateral, together with a mild dysmorphic gestalt that has been described as
    a long face with broad nasal tip, and in the contemporary series as
    hypertelorism, microcephaly, elongated face, ptosis and mild facial asymmetry.
    Clefting was near-universal in the cleft-clinic-ascertained founding families
    and is present in a minority of exome-ascertained cases, so the apparent
    frequency of this node is strongly ascertainment-dependent.
  biological_processes:
  - preferred_term: Roof of mouth development
    term:
      id: GO:0060021
      label: roof of mouth development
    modifier: ABNORMAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Orofacial clefting was seen in three individuals from our cohort, suggesting that this feature is less common than previously reported."
    explanation: The contemporary frequency anchor, and the explicit statement of the
      ascertainment shift.
  - reference: PMID:20101266
    reference_title: Structural insights into a novel histone demethylase PHF8.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Mutations and truncations in human plant homeodomain (PHD) finger protein 8 (PHF8) are associated with X-linked mental retardation and facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"
    explanation: The classical dysmorphic gestalt as summarised in the structural
      literature.
mechanistic_hypotheses:
- hypothesis_group_id: mtor_targeting
  hypothesis_label: mTOR pathway suppression as a disease-modifying strategy
  status: EMERGING
  description: >
    Because the cognitive and synaptic deficits of the Phf8 knockout mouse are
    driven by RSK1-dependent mTOR hyperactivation and are reversed by rapamycin,
    mTOR inhibition has been proposed as a rational therapeutic target for
    PHF8-related intellectual disability. The hypothesis rests on a single mouse
    study. It has not been tested in any human with a PHF8 variant, the
    therapeutic window in a chronic paediatric neurodevelopmental indication is
    unknown, and rapamycin's immunosuppressive and metabolic toxicity is
    substantial. It is recorded here as an emerging research direction, not as a
    treatment.
  evidence:
  - reference: PMID:29317619
    reference_title: Phf8 histone demethylase deficiency causes cognitive impairments
      through the mTOR pathway.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "and provides a potential therapeutic drug target to treat XLID"
    explanation: The authors' own framing of the finding as a drug-target hypothesis.
- hypothesis_group_id: oxygen_sensing_clefting
  hypothesis_label: Oxygen-dependence of PHF8 catalysis as a link to hypoxia-associated
    clefting
  status: EMERGING
  description: >
    PHF8 is a 2-oxoglutarate/Fe(II) oxygenase and therefore consumes molecular
    oxygen. This has prompted the suggestion that partial loss of PHF8 activity
    and gestational hypoxia converge on the same craniofacial endpoint, which
    would predict that maternal hypoxic exposures modify cleft risk in carriers.
    No human data address this. It is recorded because it is the only proposed
    gene-environment interaction for this disorder and because it would be
    testable.
  evidence:
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: PARTIAL
    evidence_source: IN_VITRO
    snippet: "The dependence of PHF8 activity on oxygen availability is interesting because the occurrence of fetal cleft lip has been demonstrated to increase with maternal hypoxia in mouse studies."
    explanation: >
      The originating statement of the hypothesis. Marked PARTIAL because the
      paper juxtaposes a biochemical property with a separate mouse observation
      rather than demonstrating the interaction.
phenotypes:
- name: Global Developmental Delay
  category: Neurologic
  description: >
    Developmental delay is the presenting feature and the one on which
    contemporary cases are ascertained. It was present in every affected
    individual in the largest series, including the two who did not meet criteria
    for intellectual disability.
  phenotype_term:
    preferred_term: Global developmental delay
    term:
      id: HP:0001263
      label: Global developmental delay
  frequency: VERY_FREQUENT
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "All affected individuals exhibited developmental delay and all but two had borderline to severe ID."
    explanation: >
      Supports the association and the VERY_FREQUENT band: 16 of 16 individuals in
      the cohort. VERY_FREQUENT rather than OBLIGATE because ascertainment was
      largely on developmental delay, which makes 100% partly circular.
- name: Intellectual Disability
  category: Neurologic
  description: >
    Intellectual disability is the defining feature but its severity spans
    borderline to severe, and a minority of individuals with a pathogenic PHF8
    variant do not meet criteria for intellectual disability at all. The founding
    families were described as having mild to borderline impairment; the modern
    exome-ascertained cohort reaches into the severe range.
  phenotype_term:
    preferred_term: Intellectual disability
    term:
      id: HP:0001249
      label: Intellectual disability
  frequency: VERY_FREQUENT
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "All affected individuals exhibited developmental delay and all but two had borderline to severe ID."
    explanation: >
      Supports the association and the VERY_FREQUENT band: 14 of 16 (87.5%) had
      borderline to severe intellectual disability.
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "Of the two who did not have ID, one had dyscalculia and the other had mild learning difficulties."
    explanation: >
      Documents the two individuals at the mild end who did not meet criteria for
      intellectual disability, which is why the band is not OBLIGATE.
  - reference: PMID:10398231
    reference_title: X-linked mental retardation associated with cleft lip/palate maps
      to Xp11.3-q21.3.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "A family is described in which X-linked mild to borderline mental retardation (MR) is associated with cleft lip/palate."
    explanation: The original clinical delineation, describing the mild end of the
      severity range.
- name: Mild Intellectual Disability
  category: Neurologic
  description: >
    The founding reports consistently describe the impairment as mild, and mild
    intellectual disability remains the modal severity across the older literature.
    Curated separately from the parent term because the mild-end presentation is
    what distinguishes this entity clinically from several of the other eponymous
    X-linked intellectual disability syndromes.
  phenotype_term:
    preferred_term: Mild intellectual disability
    term:
      id: HP:0001256
      label: Mild intellectual disability
  evidence:
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The clinical phenotype of the male patients was characterized by mild MR, mild dysmorphic features, unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling."
    explanation: Documents mild impairment in the Finnish F279S family.
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The PHF8 missense mutation c.836C>T is associated with mild MR, mild dysmorphic features, and either unilateral or bilateral cleft lip and cleft palate in two male siblings."
    explanation: Independent restatement of the mild-severity phenotype for the same
      allele.
  - reference: PMID:17594395
    reference_title: A novel mutation in the PHF8 gene is associated with X-linked mental
      retardation with cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "One of the truncating mutations was found in the original family with Siderius-Hamel CL/P syndrome where only two of the three affected individuals had mental retardation (MR) with CL/P and one individual had mild MR."
    explanation: Documents intrafamilial variability including a mildly affected
      individual in the original Siderius-Hamel family.
- name: Orofacial Clefting
  category: Craniofacial
  description: >
    Cleft lip with or without cleft palate, unilateral or bilateral, is the
    feature that gave the syndrome its original clinical identity. Its apparent
    frequency has fallen sharply with the shift from cleft-clinic ascertainment to
    exome-first ascertainment: it was present in essentially all of the founding
    families but in only 3 of 16 individuals in the largest contemporary series.
    Absence of clefting therefore does not exclude the diagnosis.
  phenotype_term:
    preferred_term: Orofacial cleft
    term:
      id: HP:0000202
      label: Orofacial cleft
  frequency: OCCASIONAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Orofacial clefting was seen in three individuals from our cohort, suggesting that this feature is less common than previously reported."
    explanation: >
      Supports the association and the OCCASIONAL band: 3 of 16 (18.8%), which
      falls in the 5-29% range.
  - reference: PMID:16199551
    reference_title: Mutations in PHF8 are associated with X linked mental retardation
      and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)."
    explanation: The founding gene-disease report, in which clefting co-segregated
      with the PHF8 variant.
- name: Cleft Lip
  category: Craniofacial
  description: >
    Cleft lip specifically, reported as unilateral in some affected males and
    bilateral in others within the same sibship, indicating variable expressivity
    of the clefting phenotype even for an identical allele.
  phenotype_term:
    preferred_term: Cleft lip
    term:
      id: HP:0410030
      label: Cleft lip
  evidence:
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling"
    explanation: Documents cleft lip and its laterality variation within one sibship.
- name: Cleft Palate
  category: Craniofacial
  description: >
    Cleft palate accompanies the cleft lip in the reported families rather than
    occurring in isolation, consistent with a defect in midline fusion rather than
    an isolated palatal shelf failure.
  phenotype_term:
    preferred_term: Cleft palate
    term:
      id: HP:0000175
      label: Cleft palate
  evidence:
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling"
    explanation: Documents cleft palate co-occurring with cleft lip in both affected
      siblings.
- name: Autistic Behavior
  category: Behavioral
  description: >
    Autism spectrum disorder was not emphasised in the founding descriptions but
    was frequently observed in the largest contemporary series, which is one of
    the two main phenotype expansions that series produced. An Xp11.22
    microdeletion removing PHF8 was independently reported in two brothers with
    autism spectrum disorders.
  phenotype_term:
    preferred_term: Autistic behavior
    term:
      id: HP:0000729
      label: Autistic behavior
  frequency: FREQUENT
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Autism spectrum disorder and attention deficit hyperactivity disorder, which were not previously emphasized in PHF8-XLID, were frequently observed in affected individuals."
    explanation: >
      Supports the association and the FREQUENT band by the qualitative-term
      mapping ("frequently observed"); no numerator is given in the abstract, so
      the band is not derived from a count.
  - reference: PMID:18498374
    reference_title: Autism-associated familial microdeletion of Xp11.22.
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "We describe two brothers with autistic disorder, intellectual disability (ID) and cleft lip/palate with a microdeletion of Xp11.22 detected through screening individuals with autism spectrum disorders (ASDs) for microdeletions and duplications using 1-Mb resolution array comparative genomic hybridization."
    explanation: >
      Marked PARTIAL deliberately. These brothers carry a contiguous deletion of
      PHF8 plus FAM120C and WNK3, and the authors attribute the autism component
      to the larger deletion rather than to PHF8, so this corroborates the
      association without attributing autism to PHF8 alone.
- name: Attention Deficit Hyperactivity Disorder
  category: Behavioral
  description: >
    Attention deficit hyperactivity disorder, like autism, was added to the
    phenotype by the 2022 series and was not emphasised in the founding
    descriptions.
  phenotype_term:
    preferred_term: Attention deficit hyperactivity disorder
    term:
      id: HP:0007018
      label: Attention deficit hyperactivity disorder
  frequency: FREQUENT
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Autism spectrum disorder and attention deficit hyperactivity disorder, which were not previously emphasized in PHF8-XLID, were frequently observed in affected individuals."
    explanation: >
      Supports the association and the FREQUENT band by qualitative-term mapping
      ("frequently observed").
- name: Hypertelorism
  category: Craniofacial
  description: >
    One of the craniofacial findings reported in the contemporary series, and also
    documented in the two brothers with the Xp11.22 microdeletion encompassing
    PHF8.
  phenotype_term:
    preferred_term: Hypertelorism
    term:
      id: HP:0000316
      label: Hypertelorism
  frequency: OCCASIONAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."
    explanation: >
      Supports the association and the OCCASIONAL band by qualitative-term mapping
      ("in some affected individuals"); the band is applied to the craniofacial
      group as a whole because no per-feature count is given.
- name: Microcephaly
  category: Craniofacial
  description: >
    Reduced head circumference is reported in a subset of affected individuals in
    the contemporary series. It is not a feature of the founding descriptions.
  phenotype_term:
    preferred_term: Microcephaly
    term:
      id: HP:0000252
      label: Microcephaly
  frequency: OCCASIONAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."
    explanation: >
      Supports the association and the OCCASIONAL band by qualitative-term mapping
      applied to the craniofacial group.
- name: Long Face
  category: Craniofacial
  description: >
    An elongated face is one of the more consistently reported dysmorphic
    features, appearing both in the classical description of the gestalt and in
    the contemporary series.
  phenotype_term:
    preferred_term: Long face
    term:
      id: HP:0000276
      label: Long face
  frequency: OCCASIONAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."
    explanation: >
      Supports the association and the OCCASIONAL band by qualitative-term mapping
      applied to the craniofacial group.
  - reference: PMID:20101266
    reference_title: Structural insights into a novel histone demethylase PHF8.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"
    explanation: The classical description of the facial gestalt, which places long
      face first.
- name: Ptosis
  category: Craniofacial
  description: >
    Drooping of the upper eyelid is among the craniofacial findings in the
    contemporary series.
  phenotype_term:
    preferred_term: Ptosis
    term:
      id: HP:0000508
      label: Ptosis
  frequency: OCCASIONAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."
    explanation: >
      Supports the association and the OCCASIONAL band by qualitative-term mapping
      applied to the craniofacial group.
- name: Facial Asymmetry
  category: Craniofacial
  description: >
    Mild facial asymmetry is reported in the contemporary series. It is described
    as mild and is not on its own a discriminating feature.
  phenotype_term:
    preferred_term: Facial asymmetry
    term:
      id: HP:0000324
      label: Facial asymmetry
  frequency: OCCASIONAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Craniofacial findings such as hypertelorism, microcephaly, elongated face, ptosis, and mild facial asymmetry were found in some affected individuals."
    explanation: >
      Supports the association and the OCCASIONAL band by qualitative-term mapping
      applied to the craniofacial group.
- name: Broad Nasal Tip
  category: Craniofacial
  description: >
    A broad nasal tip forms part of the classical dysmorphic gestalt as summarised
    in the early molecular literature. No frequency band is assigned because the
    source is a narrative summary without a denominator.
  phenotype_term:
    preferred_term: Broad nasal tip
    term:
      id: HP:0000455
      label: Broad nasal tip
  evidence:
  - reference: PMID:20101266
    reference_title: Structural insights into a novel histone demethylase PHF8.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"
    explanation: Documents broad nasal tip as part of the reported facial gestalt.
- name: Large Hands
  category: Skeletal
  description: >
    Large hands appear in the classical summary of the PHF8 phenotype. This is the
    weakest-supported feature in the entry - it derives from a narrative summary in
    a structural-biology paper rather than from a systematic clinical series, and
    is not mentioned in the 2022 cohort. No frequency band is assigned.
  phenotype_term:
    preferred_term: Large hands
    term:
      id: HP:0001176
      label: Large hands
  evidence:
  - reference: PMID:20101266
    reference_title: Structural insights into a novel histone demethylase PHF8.
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "facial anomalies, such as a long face, broad nasal tip, cleft lip/cleft palate and large hands"
    explanation: >
      Marked PARTIAL: the statement is a secondary narrative summary of the
      phenotype rather than a primary clinical observation, and the feature is
      absent from the largest contemporary series.
genetic:
- name: PHF8
  association: Causative
  variant_origin: GERMLINE
  gene_term:
    preferred_term: PHF8
    term:
      id: hgnc:20672
      label: PHF8
  notes: >
    PHF8 (Xp11.22; aliases JHDM1F, KIAA1111, ZNF422) encodes a 1024-amino-acid
    PHD-finger and JmjC-domain histone demethylase. The gene is ubiquitously
    transcribed, with strong expression of the mouse orthologue in embryonic and
    adult brain. The reported disease-allele spectrum is small - the 2022 series
    counted only five previous reports describing PHF8 predicted loss-of-function
    variants in eight individuals before it added sixteen more - and comprises
    nonsense, frameshift, splice and JmjC-clustered missense alleles plus
    Xp11.22 microdeletions. Clinical missense alleles cluster in the exons
    encoding the double-stranded beta-helix catalytic fold. Mechanism is loss of
    function throughout; no dominant-negative or gain-of-function germline
    mechanism has been demonstrated.

    Two scope cautions. First, PHF8 must not be confused with PHF6 (the
    Borjeson-Forssman-Lehmann gene) or PHF21A - all are PHD-finger chromatin
    genes with distinct diseases. Second, there is a large somatic-oncology PHF8
    literature (breast, hepatocellular and gastric carcinoma) in which PHF8 acts
    as an overexpressed tumour promoter; that is a different disease context and
    contributes nothing to this germline neurodevelopmental entity.
  evidence:
  - reference: PMID:16199551
    reference_title: Mutations in PHF8 are associated with X linked mental retardation
      and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)."
    explanation: Gene-discovery paper establishing the PHF8 gene-disease
      relationship and binding it to OMIM 300263.
  - reference: PMID:16199551
    reference_title: Mutations in PHF8 are associated with X linked mental retardation
      and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Expression studies showed that this gene is ubiquitously transcribed, with strong expression of the mouse orthologue Phf8 in embryonic and adult brain structures."
    explanation: Sources the expression statement in the notes.
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "PHF8-XLID is an under-characterized disorder with only five previous reports describing different PHF8 predicted loss-of-function variants in eight individuals."
    explanation: Quantifies how small the pre-2022 allele series was, which is the
      basis for the "small reported spectrum" claim.
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "In this report we present 16 additional individuals with PHF8-XLID from 11 different families of diverse ancestry."
    explanation: Establishes the size and family structure of the largest published
      cohort, which is the denominator for the frequency derivations.
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Mutations of human PHF8 cluster within its JmjC encoding exons and are linked to mental retardation (MR) and a cleft lip/palate phenotype."
    explanation: Sources the statement that clinical alleles cluster in the
      JmjC-encoding exons.
variants:
- name: PHF8 p.Phe279Ser
  description: >
    The c.836C>T (p.F279S) missense allele, identified in a Finnish family with
    multiply affected males. It is the reference functional allele for this
    disorder: it substitutes a conserved hydrophobic residue in the JmjC domain
    and is catalytically dead in assays against both peptide and intact-histone
    substrates, in two independent laboratories, while leaving the rest of the
    protein intact. It is also the allele used to show that catalytic activity is
    required for PHF8 to drive neuronal differentiation.
  gene:
    preferred_term: PHF8
    term:
      id: hgnc:20672
      label: PHF8
  type: MISSENSE
  evidence:
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "A novel missense mutation c.836C>T of the PHF8 gene was identified in a Finnish family with multiple-affected male patients."
    explanation: Identifies the allele and its clinical context.
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The mutation resides in exon 8 and changes phenylalanine to serine (F279S) in the functionally important Jmonji C domain of the protein."
    explanation: Localises the substitution to the catalytic domain.
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "This mutant encodes a F279S variant of PHF8 that modifies a conserved hydrophobic region; assays with both peptides and intact histones reveal this variant to be catalytically inactive."
    explanation: First functional demonstration of catalytic inactivity.
  - reference: PMID:20548336
    reference_title: The X-linked mental retardation gene PHF8 is a histone demethylase
      involved in neuronal differentiation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Importantly, a mutant PHF8 (phenylalanine at position 279 to serine) identified in the XLMR patients is defective in enzymatic activity, indicating that the loss of histone demethylase activity is causally linked with the onset of disease."
    explanation: Independent confirmation, with the authors drawing the causal
      inference from enzyme loss to disease.
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The mutation was not present in 200 anonymous blood donors (approximately 300 X-chromosomes)."
    explanation: Population-control data supporting pathogenicity at the time of
      report.
- name: PHF8 p.Lys177Ter
  description: >
    A nonsense allele (p.K177X) reported in a male with intellectual disability and
    cleft lip/palate. It truncates the protein upstream of the JmjC domain and
    removes five nuclear localisation signals, so it is the cleanest example of
    the null end of the allele spectrum and the reason the PHD/nuclear-targeting
    arm is modelled separately from the catalytic arm.
  gene:
    preferred_term: PHF8
    term:
      id: hgnc:20672
      label: PHF8
  type: NONSENSE
  evidence:
  - reference: PMID:17594395
    reference_title: A novel mutation in the PHF8 gene is associated with X-linked mental
      retardation with cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Here, we report a novel nonsense mutation (p.K177X) in a male patient who has MR associated with CL/P."
    explanation: Identifies the allele and the clinical presentation.
  - reference: PMID:17594395
    reference_title: A novel mutation in the PHF8 gene is associated with X-linked mental
      retardation with cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The mutation results in a truncated PHF8 protein lacking the Jumonji-like C terminus domain and five nuclear localization signals."
    explanation: Establishes that the allele removes both the catalytic domain and
      the nuclear localisation signals.
inheritance:
- name: X-linked recessive inheritance
  inheritance_term:
    preferred_term: X-linked recessive inheritance
    term:
      id: HP:0001419
      label: X-linked recessive inheritance
  penetrance: INCOMPLETE
  expressivity: VARIABLE
  description: >
    This is the authoritative OMIM/HPO inheritance annotation and correctly
    describes the reported pedigrees: affected individuals are hemizygous males,
    the founding families were multiplex X-linked pedigrees with transmission
    through unaffected carrier mothers, and the original linkage was to
    Xp11.3-q21.3. Penetrance is recorded as incomplete and expressivity as
    variable on the strength of two documented observations within families: in
    the original Siderius-Hamel family only two of three affected individuals had
    both intellectual disability and clefting while the third had mild
    intellectual disability alone, and in the Finnish F279S sibship the identical
    allele produced unilateral clefting in one brother and bilateral clefting in
    the other. No systematic study of heterozygous female carriers has been
    published for this entity, so nothing is asserted about female manifestation
    here.
  evidence:
  - reference: PMID:10398231
    reference_title: X-linked mental retardation associated with cleft lip/palate maps
      to Xp11.3-q21.3.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Linkage analysis showed a maximum LOD score of Z=2.78 at straight theta=0.0 for the DXS441 locus with flanking markers DXS337 and DXS990, defining the region Xp11.3-q21.3 with a linkage interval of 25 cM."
    explanation: The original X-linkage evidence that established the inheritance
      pattern before the gene was known.
  - reference: PMID:17594395
    reference_title: A novel mutation in the PHF8 gene is associated with X-linked mental
      retardation with cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The second mutation was present in a family with four affected men, three of whom had MR and CL/P, while the fourth individual had mild MR without clefting."
    explanation: Documents intrafamilial variability for the clefting component,
      supporting the VARIABLE expressivity annotation.
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "unilateral cleft lip and cleft palate in one and bilateral cleft lip and cleft palate in the other sibling"
    explanation: Same-allele, same-family variation in cleft laterality, a second
      independent line of support for variable expressivity.
  - reference: PMID:18498374
    reference_title: Autism-associated familial microdeletion of Xp11.22.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "shown to be inherited from their unaffected mother who had skewed (100%) X inactivation of the aberrant chromosome"
    explanation: >
      The only published X-inactivation data bearing on carrier status in this
      disorder: complete skewing in an unaffected transmitting mother. It is a
      single observation of a deletion carrier, so it is consistent with, but
      does not establish, non-penetrance in females generally.
prevalence:
- population: Worldwide
  measure_type: CASES_IN_LITERATURE
  prevalence_class: NOT_YET_DOCUMENTED
  notes: >-
    No population prevalence or incidence estimate has been published for this
    disorder, so only a literature case count is recorded. Derivation: 8
    individuals across five reports before 2022, plus 16 individuals from 11
    families in the 2022 series, giving 24 individuals with a PHF8
    loss-of-function variant. Five further individuals from four families
    carrying PHF8 variants of uncertain significance are deliberately excluded
    from the count. Individuals with Xp11.22 contiguous-gene deletions are also
    excluded because they are not single-gene cases. No rate is asserted: there
    is no denominator that would license one, and the Orphanet prevalence class
    could not be cited because ORPHA:85287 is not present in references_cache
    and refreshing it would require a MANIFEST bump (blocked pending #7622).
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "PHF8-XLID is an under-characterized disorder with only five previous reports describing different PHF8 predicted loss-of-function variants in eight individuals."
    explanation: Supplies the pre-2022 count of 8 individuals used in the
      derivation.
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "In this report we present 16 additional individuals with PHF8-XLID from 11 different families of diverse ancestry."
    explanation: Supplies the 16 additional individuals used in the derivation.
diagnosis:
- name: Exome or Genome Sequencing
  description: >
    There are no formal clinical diagnostic criteria and the gestalt is too mild
    to be reliably recognised prospectively; molecular testing is the diagnosis.
    Exome or genome sequencing, or a targeted X-linked intellectual disability
    gene panel, is the appropriate route for a male with unexplained developmental
    delay, with or without orofacial clefting. The ascertainment shift matters
    clinically: the largest series was assembled largely through sequencing of
    individuals with developmental delay rather than through cleft clinics, so
    PHF8 should not be reserved for patients who have a cleft. Chromosomal
    microarray covers the Xp11.22 microdeletion fraction, which sequencing may
    miss.
  evidence:
  - reference: PMID:31906484
    reference_title: Targeted Next-Generation Sequencing in Patients with Suggestive
      X-Linked Intellectual Disability.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Seven families could be re-contacted and variant segregation analysis of the respective eight candidate variants was performed: HUWE1, IQSEC2, MAOA, MED12, PHF8, SLC6A8, SLC9A6, and SYN1."
    explanation: Demonstrates PHF8 being returned as a candidate by a targeted
      X-linked intellectual disability sequencing panel in real diagnostic practice.
  - reference: PMID:31906484
    reference_title: Targeted Next-Generation Sequencing in Patients with Suggestive
      X-Linked Intellectual Disability.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Our results show the utility of targeted next-generation sequencing in unravelling the genetic origin of XLID, especially in retrospective cases."
    explanation: Supports targeted sequencing as a diagnostic strategy in this
      disease class.
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We also present five individuals from four different families who have ID and a variant of unknown significance in PHF8 with no other explanatory variant in another gene."
    explanation: >
      Documents the interpretation problem that follows from sequencing-first
      ascertainment - a substantial fraction of PHF8 findings in individuals with
      intellectual disability remain variants of uncertain significance.
- name: Targeted PHF8 Sequencing in Cleft Plus Intellectual Disability
  description: >
    The historical diagnostic route, and still reasonable when the combination is
    present. In a nationwide Finnish cleft cohort, restricting PHF8 sequencing to
    the 18 patients who had cleft lip/palate together with intellectual disability
    of unknown cause yielded one causative family. That yield illustrates both
    that the combination enriches strongly for PHF8 and that PHF8 accounts for
    only a small fraction of syndromic clefting.
  evidence:
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "From this nationwide material, 18 patients including one family with two male patients with cleft lip/cleft palate and unknown cause of mental retardation (MR) were sequenced for the coding regions and splice sites of the PHF8 gene."
    explanation: Describes the phenotype-directed screening strategy and its
      denominator.
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We investigated the prevalence of mutations in the PHD finger protein 8 (PHF8) gene in X-linked mental retardation (XLMR) and facial cleft starting from the original cohort of 7712 patients operated on since 1 January 1950 for cleft lip/cleft palate in the Cleft Centre at the Helsinki University Hospital."
    explanation: >
      Establishes the full cleft-cohort denominator (7712) from which the 18
      sequenced patients were selected, which is why no population frequency is
      derived from this study.
differential_diagnoses:
- name: Claes-Jensen type syndromic X-linked intellectual disability (KDM5C)
  disease_term:
    preferred_term: syndromic X-linked intellectual disability Claes-Jensen type
    term:
      id: MONDO:0010355
      label: syndromic X-linked intellectual disability Claes-Jensen type
  description: >
    The most important named-entity distinction for this entry, because the two
    literatures are genuinely entangled rather than merely similarly named. Both
    are eponymous X-linked syndromic intellectual disability entities caused by
    JmjC-domain histone demethylases, both were delineated by the same era of
    X-chromosome mutation screening, and PHF8 and KDM5C/JARID1C are functionally
    linked in one transcriptional module through ZNF711 - so KDM5C is cited
    inside PHF8 mechanism papers and vice versa. They are nonetheless distinct
    MONDO entities with distinct genes, distinct OMIM numbers and distinct
    clinical profiles: Claes-Jensen is a KDM5C H3K4me3 eraser disorder with short
    stature, spasticity, hyperreflexia and seizures and is one of the commoner
    causes of X-linked intellectual disability, whereas Siderius type is a PHF8
    H4K20me1/H3K9me1-2 eraser disorder with orofacial clefting and a milder,
    much rarer picture. Evidence must never be pooled across the two.
  evidence:
  - reference: PMID:20346720
    reference_title: A functional link between the histone demethylase PHF8 and the
      transcription factor ZNF711 in X-linked mental retardation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Taken together, our results functionally link the XLMR gene PHF8 to two other XLMR genes, ZNF711 and JARID1C, indicating that MR genes may be functionally linked in pathways, causing the complex phenotypes observed in patients developing MR."
    explanation: >
      Documents exactly the functional entanglement that makes this differential
      necessary: PHF8 and JARID1C/KDM5C share a pathway but cause separate
      diseases.
  - reference: PMID:31691806
    reference_title: Histone demethylase KDM5C is a SAHA-sensitive central hub at the
      crossroads of transcriptional axes involved in multiple neurodevelopmental
      disorders.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Interestingly, mutations in all four genes (KDM5C, ARX, ZNF711 and PHF8) are associated with X-linked NDDs comprising intellectual disability as a core feature."
    explanation: >
      States explicitly that PHF8 and KDM5C are separate genes causing separate
      X-linked neurodevelopmental disorders while sharing one regulatory module -
      the precise reason the two literatures must be kept apart during curation.
- name: Nonsyndromic cleft lip with or without cleft palate
  disease_term:
    preferred_term: orofacial cleft
    term:
      id: MONDO:0000358
      label: orofacial cleft
  description: >
    Orofacial clefting is one of the commonest birth defects, so the great
    majority of children presenting with cleft lip/palate do not have this
    syndrome. What should raise suspicion of PHF8 is the combination of clefting
    with intellectual disability in a male with an X-linked family history - the
    Finnish screen found one PHF8 family among 18 such selected patients drawn
    from a nationwide cleft cohort of 7712.
  evidence:
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The prevalence of cleft lip with or without cleft palate is amongst the most common of all birth defects, averaging at 10.5 per 10,000 live births in the United States"
    explanation: Quantifies the background clefting rate against which this rare
      syndrome must be distinguished.
- name: Xp11.22 contiguous gene deletion syndrome
  description: >
    Deletions at Xp11.22 can remove PHF8 together with neighbouring genes
    including FAM120C and part of WNK3. The reported brothers with such a
    microdeletion had the Siderius features plus autism spectrum disorder,
    hypertelorism and broad halluces, so a deletion-mediated presentation may be
    broader than the single-gene phenotype and the extra features should not be
    attributed to PHF8. Chromosomal microarray, not sequencing, is the test that
    resolves this.
  evidence:
  - reference: PMID:19843542
    reference_title: PHF8, a gene associated with cleft lip/palate and mental retardation,
      encodes for an Nepsilon-dimethyl lysine demethylase.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Recently, a microdeletion encompassing all of the PHF8 and FAM120C genes and parts of WNK3 was reported in two brothers with autism spectrum disorders, causing hypertelorism and broad halluces in addition to the Siderius XLMR features"
    explanation: Documents the contiguous-gene deletion presentation and the extra
      features it carries.
  - reference: PMID:18498374
    reference_title: Autism-associated familial microdeletion of Xp11.22.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Our findings show that in addition to point mutations, a complete deletion of the PHF8 gene is associated with the X-linked mental retardation Siderius-Hamel syndrome (OMIM 300263)"
    explanation: >
      The primary report of the deletion, and a second independent identity anchor
      binding PHF8 deletion to OMIM 300263, the OMIM xref of MONDO:0010286.
  - reference: PMID:18498374
    reference_title: Autism-associated familial microdeletion of Xp11.22.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "further suggest that the larger size of the Xp11.22 deletion including genes FAM120C and WNK3 may be involved in the pathogenesis of autism"
    explanation: >
      The authors themselves attribute the autism component to the extra deleted
      genes rather than to PHF8, which is why deletion cases must not be pooled
      with single-gene cases when curating the PHF8 phenotype.
treatments:
- name: Surgical Repair of Orofacial Cleft
  description: >
    Where a cleft is present, staged surgical repair is the standard of care and
    is the single most consequential intervention for these patients. It is
    generic cleft management rather than anything specific to PHF8, and is
    included because clefting is the defining malformation of the syndrome and
    because the founding Finnish cohort was assembled from patients operated on
    for cleft lip/palate. No PHF8-specific surgical outcome data exist.
  treatment_term:
    preferred_term: reconstructive surgery for orofacial cleft
    term:
      id: NCIT:C25351
      label: Reconstructive Surgery
  therapeutic_modality: SURGERY
  evidence:
  - reference: PMID:17661819
    reference_title: Screening of mutations in the PHF8 gene and identification of a
      novel mutation in a Finnish family with XLMR and cleft lip/cleft palate.
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "the original cohort of 7712 patients operated on since 1 January 1950 for cleft lip/cleft palate in the Cleft Centre at the Helsinki University Hospital"
    explanation: >
      Marked PARTIAL: this establishes that affected individuals are managed
      surgically in a cleft centre, but it is not an outcome study and no
      PHF8-specific surgical evidence exists.
- name: Developmental and Educational Intervention
  description: >
    Management of the neurodevelopmental phenotype is supportive and identical to
    that for other causes of developmental delay: early intervention, special
    education, and speech and language therapy directed at the communication
    delay. There is no disease-specific evidence base; this is recorded because
    it is the actual standard of care and because developmental delay was present
    in every individual in the largest series.
  treatment_term:
    preferred_term: speech and language therapy
    term:
      id: NCIT:C159273
      label: Speech Language Therapy
  therapeutic_modality: BEHAVIORAL
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "All affected individuals exhibited developmental delay and all but two had borderline to severe ID."
    explanation: >
      Marked PARTIAL: this establishes the universal need for developmental
      intervention but is not evidence for the efficacy of any specific therapy in
      this disorder.
- name: Genetic Counselling
  description: >
    X-linked recessive counselling applies: carrier mothers have a 50% risk of
    transmitting the allele to each son, affected males transmit the allele to all
    daughters and no sons, and maternal carrier testing should follow a diagnosis
    in a proband. The founding pedigrees were multiplex X-linked families
    identified through linkage, which is the empirical basis for the counselling
    model.
  treatment_term:
    preferred_term: genetic counseling
    term:
      id: NCIT:C15240
      label: Genetic Counseling
  therapeutic_modality: BEHAVIORAL
  evidence:
  - reference: PMID:10398231
    reference_title: X-linked mental retardation associated with cleft lip/palate maps
      to Xp11.3-q21.3.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "A family is described in which X-linked mild to borderline mental retardation (MR) is associated with cleft lip/palate."
    explanation: Establishes the X-linked familial transmission pattern that the
      counselling model rests on.
animal_models:
- species: Mus musculus
  genotype: Phf8 knockout (constitutive null; Chen et al. 2018 line)
  category: KNOCKOUT
  description: >
    The most informative model for the cognitive phenotype, but see the second
    mouse entry below for a direct contradiction. Phf8-null mice in this line
    have impaired learning and memory and impaired hippocampal long-term
    potentiation with no gross morphological brain defect, which parallels the
    absence of a consistent structural brain abnormality in patients. The
    deficits are driven by RSK1 derepression and mTOR hyperactivation and are
    rescued by rapamycin. Two limitations: the model does not reproduce the
    craniofacial arm of the human syndrome, and an independently generated
    knockout line found no cognitive impairment at all.
  evidence:
  - reference: PMID:29317619
    reference_title: Phf8 histone demethylase deficiency causes cognitive impairments
      through the mTOR pathway.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Here we report that Phf8 knockout mice displayed impaired learning and memory, and impaired hippocampal long-term potentiation (LTP) without gross morphological defects."
    explanation: Describes the model's core neurobehavioural phenotype.
  - reference: PMID:41714361
    reference_title: Epigenetic regulation of serine biosynthesis by PHF8 during neurogenesis.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "In vivo, PHF8 deficiency in mouse embryos halts neurogenesis, progenitor expansion, and neuron generation in the developing brain."
    explanation: A second, developmental-stage mouse phenotype from an independent
      group.
- species: Mus musculus
  genotype: Phf8 knockout (independently generated constitutive null; Walsh et al.
    2017 line)
  category: KNOCKOUT
  description: >
    An independently generated Phf8 knockout allele that directly contradicts the
    other line on the phenotype that matters most: these mice have neither
    obvious developmental defects nor signs of cognitive impairment. What they do
    show is resilience to stress-induced anxiety- and depression-like behaviour,
    attributable to misregulated serotonin signalling in prefrontal cortex, with
    Htr1a and Htr2a as direct PHF8 targets. Recorded here because omitting it
    would misrepresent the strength of the mouse evidence, and because no
    counterpart of the stress-resilience phenotype has been looked for in
    patients.
  evidence:
  - reference: PMID:28485378
    reference_title: Phf8 loss confers resistance to depression-like and anxiety-like
      behaviors in mice.
    supports: REFUTE
    evidence_source: MODEL_ORGANISM
    snippet: "Phf8 deficient mice neither display obvious developmental defects nor signs of cognitive impairment."
    explanation: >
      Marked REFUTE against the claim that a Phf8 knockout mouse models the human
      cognitive phenotype - this line did not reproduce it, which is the single
      most important caveat on the mouse literature for this disorder.
  - reference: PMID:28485378
    reference_title: Phf8 loss confers resistance to depression-like and anxiety-like
      behaviors in mice.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "However, we report a striking resiliency to stress-induced anxiety- and depression-like behaviour on loss of Phf8."
    explanation: The phenotype this line does show, which has no described human
      counterpart.
  - reference: PMID:28485378
    reference_title: Phf8 loss confers resistance to depression-like and anxiety-like
      behaviors in mice.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "We further observe misregulation of serotonin signalling within the prefrontal cortex of Phf8 deficient mice and identify the serotonin receptors Htr1a and Htr2a as direct targets of PHF8."
    explanation: Supplies the mechanism of the model-only behavioural phenotype.
- species: Danio rerio
  genotype: phf8 morpholino knockdown
  category: KNOCKDOWN
  description: >
    The model that captures the craniofacial arm the mouse misses. Loss of phf8
    in zebrafish impairs brain cell survival and jaw development, acting in part
    through direct regulation of the homeodomain transcription factor MSX1/MSXB.
    It remains the principal in vivo evidence connecting PHF8 catalysis to
    craniofacial morphogenesis.
  evidence:
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain and
      craniofacial development.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "PHF8 regulates cell survival in the zebrafish brain and jaw development, thus providing a potentially relevant biological context for understanding the clinical symptoms associated with PHF8 patients."
    explanation: Establishes the craniofacial and neuronal-survival phenotypes of the
      model.
  - reference: PMID:20622853
    reference_title: Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain and
      craniofacial development.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Our findings indicate that an imbalance of histone methylation dynamics has a critical role in XLMR."
    explanation: The authors' summary conclusion linking the model back to the human
      disease class.
- species: Caenorhabditis elegans
  genotype: F29B9.2 (jmjd-1.2) loss-of-function mutant
  category: MUTANT
  description: >
    The invertebrate orthologue demonstrates deep conservation of both the
    enzymatic activity and the neuronal requirement: F29B9.2 catalyses H3K9me2/me1
    demethylation, is highly expressed in neurons, and mutant animals have
    impaired locomotion. Useful for establishing conservation, but the behavioural
    readout is too distant from human cognition to model the clinical phenotype.
  evidence:
  - reference: PMID:20346720
    reference_title: A functional link between the histone demethylase PHF8 and the
      transcription factor ZNF711 in X-linked mental retardation.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Of interest, the C. elegans PHF8 homolog is highly expressed in neurons, and mutant animals show impaired locomotion."
    explanation: Documents the conserved neuronal expression and the mutant
      behavioural phenotype.
discussions:
- discussion_id: phf8_clefting_ascertainment
  kind: KNOWLEDGE_GAP
  status: OPEN
  prompt: >
    What is the true frequency of orofacial clefting in PHF8-related intellectual
    disability once ascertainment bias is removed?
  attaches_to:
  - pathophysiology#Orofacial Clefting and Craniofacial Dysmorphology
  rationale: >
    Every published frequency for clefting in this disorder is confounded by how
    the cases were found. The founding families came from cleft clinics and
    multiplex X-linked pedigrees, so clefting appeared obligate; the 2022 series
    came largely from exome sequencing for developmental delay and found clefting
    in 3 of 16. Neither denominator is unbiased. Because clefting is the feature
    that gives the syndrome its clinical identity, and because its presence or
    absence currently drives whether PHF8 is considered at all, this is the single
    most consequential unresolved number in the entry.
  proposed_experiments:
  - experiment_id: unselected_dd_cohort_cleft_frequency
    name: Cleft frequency in PHF8 carriers from an unselected developmental-delay
      cohort
    description: >
      Ascertain PHF8 loss-of-function carriers from a large-scale sequencing
      cohort recruited on developmental delay alone, with no craniofacial
      inclusion criterion, then phenotype for orofacial clefting prospectively.
    decision_criterion: >
      The cleft frequency in the unselected cohort differs significantly from
      both the founding-cohort estimate and the 3-of-16 figure.
    would_support:
    - Clefting frequency in PHF8-related intellectual disability is
      ascertainment-determined rather than a fixed property of the disorder.
    would_refute:
    - Clefting is an intrinsically variable feature whose frequency is stable
      across ascertainment frames.
  - experiment_id: biobank_reverse_phenotyping
    name: Reverse phenotyping of PHF8 variant carriers in population biobanks
    description: >
      Identify PHF8 loss-of-function carriers in population biobanks with no
      neurodevelopmental ascertainment at all, and retrieve their craniofacial
      and cognitive phenotypes from linked health records.
    decision_criterion: >
      A substantial fraction of biobank PHF8 carriers have neither clefting nor
      recorded intellectual disability.
    would_support:
    - The published phenotype is the severe tail of a broader and milder
      distribution.
- discussion_id: phf8_mtor_rapamycin_translation
  kind: HUMAN_MODEL_MISMATCH
  status: OPEN
  prompt: >
    Does the rapamycin rescue of the Phf8 knockout mouse predict anything about
    human PHF8-related intellectual disability, given that a second independent
    knockout line showed no cognitive impairment at all?
  attaches_to:
  - pathophysiology#RSK1 Derepression and mTOR Pathway Hyperactivation
  rationale: >
    The mouse result is striking - mTOR hyperactivation is mechanistically
    specified down to RSK1, and rapamycin restores both LTP and cognition. But
    the rescue rests on a single knockout line, and a separately generated Phf8
    knockout allele reported neither developmental defects nor cognitive
    impairment, so the very phenotype the rescue corrects was not reproducible
    between laboratories. Neither line reproduces the craniofacial phenotype, and
    no measurement of mTOR pathway activity has been reported in any human tissue
    from a person with a PHF8 variant. Whether the human disorder is an
    mTOR-opathy at all is therefore untested, and treating the mouse rescue as a
    therapeutic lead would be premature in a chronic paediatric indication where
    rapamycin carries real toxicity.
  proposed_experiments:
  - experiment_id: patient_ipsc_mtor_activity
    name: mTOR pathway activity in patient-derived PHF8-mutant neurons
    description: >
      Measure S6K and S6 phosphorylation and RSK1 protein levels in iPSC-derived
      cortical neurons from individuals carrying PHF8 loss-of-function alleles,
      against isogenic corrected controls.
    decision_criterion: >
      PHF8-mutant human neurons show elevated mTOR pathway activity relative to
      isogenic controls.
    would_support:
    - The mouse mTOR mechanism operates in human PHF8-deficient neurons.
    would_refute:
    - The mTOR arm is a mouse-specific consequence of Phf8 loss.
  - experiment_id: f279s_rsk1_derepression
    name: RSK1 derepression by the catalytic-dead F279S allele
    description: >
      Test whether the F279S catalytic-dead missense allele, and not only a
      complete null, derepresses RSK1, since several reported human alleles
      retain protein.
    decision_criterion: >
      F279S produces RSK1 derepression comparable to the null allele.
    would_support:
    - The mTOR arm is engaged by catalytic loss alone and therefore applies to
      the missense end of the human allele spectrum.
  - experiment_id: phf8_ko_line_comparison
    name: Side-by-side comparison of the two Phf8 knockout alleles
    description: >
      Backcross both published Phf8 knockout alleles onto a common genetic
      background and run identical learning, memory and hippocampal LTP assays
      in the same laboratory.
    decision_criterion: >
      The cognitive phenotype tracks the allele rather than the background.
    would_support:
    - The published discordance is allele-driven and one allele is not a true
      null.
    would_refute:
    - The discordance is background- or protocol-driven, which would weaken both
      reports equally.
- discussion_id: phf8_mouse_no_clefting
  kind: HUMAN_MODEL_MISMATCH
  status: OPEN
  prompt: >
    Why do Phf8-null mice fail to develop orofacial clefting when human PHF8
    loss-of-function causes cleft lip/palate and zebrafish phf8 knockdown causes
    jaw malformation?
  attaches_to:
  - pathophysiology#Impaired Neural Crest and Craniofacial Midline Development
  rationale: >
    The craniofacial arm of this entry rests almost entirely on zebrafish jaw and
    msxb data, because both published Phf8 knockout mouse lines are reported
    without craniofacial malformation - one explicitly as having no obvious
    developmental defects, the other as having no gross morphological defects.
    That is a species mismatch at exactly the point where the human phenotype is
    most distinctive. Paralogue compensation by PHF2/KDM7A in mouse, or a species
    difference in how much lip and palate fusion depends on MSX1, would both
    explain it, and distinguishing them would materially change how the
    craniofacial mechanism should be modelled.
  proposed_experiments:
  - experiment_id: phf8_phf2_double_mutant
    name: Phf8;Phf2 double-mutant mouse craniofacial phenotyping
    description: >
      Generate mice null for both Phf8 and its closest paralogue Phf2 and assess
      lip and palate fusion, to test whether paralogue compensation masks a
      craniofacial requirement in mouse.
    decision_criterion: >
      Clefting emerges in double mutants but not in either single mutant.
    would_support:
    - Paralogue compensation explains the absence of clefting in Phf8-null mice.
  - experiment_id: msx1_facial_primordia_assay
    name: Msx1 expression in Phf8-null mouse facial primordia
    description: >
      Assay Msx1 transcript and protein in the facial primordia of Phf8-null
      mouse embryos across the lip and palate fusion window.
    decision_criterion: >
      Msx1 expression is reduced in Phf8-null primordia.
    would_support:
    - The zebrafish msxb mechanism is conserved in mammals even where the
      morphological endpoint is not.
    would_refute:
    - The MSX1 link is fish-specific and the human craniofacial mechanism is
      unexplained.
- discussion_id: phf8_female_carrier_status
  kind: KNOWLEDGE_GAP
  status: OPEN
  prompt: >
    Are heterozygous female carriers of pathogenic PHF8 variants affected?
  attaches_to:
  - pathophysiology#Pathogenic PHF8 Loss-of-Function Variant
  rationale: >
    The disorder is annotated X-linked recessive and every reported affected
    individual is male, but for several other X-linked chromatinopathies - KDM5C
    and USP9X among them - the recessive label has proved empirically wrong for
    females once carriers were systematically assessed. The only X-inactivation
    data in this disorder come from a single unaffected deletion-carrier mother
    with complete skewing. No systematic study of PHF8 carrier females has been
    published, so the absence of reported affected females may reflect absence of
    looking rather than absence of effect. This matters directly for genetic
    counselling.
  proposed_experiments:
  - experiment_id: carrier_mother_cognitive_phenotyping
    name: Formal cognitive phenotyping of obligate PHF8 carrier mothers
    description: >
      Recruit obligate carrier mothers from the published PHF8 pedigrees and
      apply formal cognitive and adaptive-behaviour testing rather than clinical
      impression, with population norms as the comparator.
    decision_criterion: >
      Carrier mothers score below population norms on formal testing.
    would_support:
    - The X-linked recessive annotation understates female involvement.
  - experiment_id: phf8_xci_escape
    name: Determine whether PHF8 escapes X-inactivation
    description: >
      Measure allele-specific PHF8 expression in female cells with informative
      heterozygous markers and quantified X-inactivation ratios.
    decision_criterion: >
      PHF8 shows measurable expression from the inactive X.
    would_support:
    - Females have a higher baseline PHF8 dose, which would predict milder or
      absent manifestation.
- discussion_id: phf8_episignature
  kind: KNOWLEDGE_GAP
  status: OPEN
  prompt: >
    Does PHF8-related intellectual disability have a DNA-methylation episignature
    that could resolve its reported variants of uncertain significance?
  attaches_to:
  - pathophysiology#Failure of RNA Polymerase II-Coupled Transcriptional Coactivation
  rationale: >
    Many chromatinopathies - Kabuki, Sotos and CHARGE among them - now have
    peripheral-blood DNA-methylation episignatures that reclassify missense
    variants of uncertain significance into pathogenic or benign. No such
    classifier exists for PHF8. This is not an academic gap: the largest series
    reported five individuals from four families with intellectual disability, a
    PHF8 variant of uncertain significance and no alternative explanation, which
    is precisely the population an episignature would resolve. PHF8 is a
    transcriptional coactivator acting on repressive histone marks, so a
    downstream methylation signature is mechanistically plausible.
  proposed_experiments:
  - experiment_id: phf8_methylation_classifier
    name: Peripheral-blood methylation classifier for PHF8 loss of function
    description: >
      Generate genome-wide methylation array data from peripheral blood of
      individuals with confirmed PHF8 loss-of-function variants and train a
      classifier against matched controls and against other chromatinopathy
      signatures.
    decision_criterion: >
      A reproducible signature separates PHF8 loss-of-function carriers from
      controls with high sensitivity and specificity.
    would_support:
    - PHF8 belongs among the episignature-classifiable chromatinopathies.
  - experiment_id: phf8_vus_classification
    name: Classification of reported PHF8 missense variants of uncertain significance
    description: >
      Apply the trained classifier to samples from the five reported individuals
      carrying PHF8 missense variants of uncertain significance.
    decision_criterion: >
      Each variant of uncertain significance is assigned to the affected or the
      control cluster.
    would_support:
    - Episignature testing can resolve the PHF8 variant-interpretation backlog.
  evidence:
  - reference: PMID:35469323
    reference_title: Variants in PHF8 cause a spectrum of X-linked neurodevelopmental
      disorders and facial dysmorphology.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We also present five individuals from four different families who have ID and a variant of unknown significance in PHF8 with no other explanatory variant in another gene."
    explanation: Establishes that the variant-of-uncertain-significance population
      this gap concerns actually exists and is quantified.

📚

References & Deep Research

References

PMID:35469323

Variants in PHF8 cause a spectrum of X-linked neurodevelopmental disorders and facial dysmorphology.

No top-level findings curated for this source.

PMID:16199551

Mutations in PHF8 are associated with X linked mental retardation and cleft lip/cleft palate.

No top-level findings curated for this source.

PMID:10398231

X-linked mental retardation associated with cleft lip/palate maps to Xp11.3-q21.3.

No top-level findings curated for this source.

Deep Research

Claude Code ▸

1. Disease Information

claude-haiku-4-5-20251001, claude-opus-5[1m] 13 citations 2026-08-01T00:51:27.827351

1. Disease Information

1.1 Overview

Siderius type syndromic X‑linked intellectual developmental disorder (MRXSSD), increasingly called PHF8‑XLID in the recent literature, is a rare X‑linked recessive chromatinopathy caused by loss‑of‑function variants in PHF8 at Xp11.22. PHF8 encodes a JmjC‑domain, Fe(II)/2‑oxoglutarate‑dependent histone lysine demethylase (KDM7B). The classical clinical description — derived from the founding families — is mild-to-borderline intellectual disability with cleft lip and/or cleft palate and mild facial dysmorphism in affected males. The 2022 multicenter expansion (n = 16 new affected males) substantially reframed the phenotype: developmental delay is universal, ID is near-universal, autism spectrum disorder and ADHD are frequent and were previously under-recognized, and orofacial clefting is much less common than the founding-cohort literature implied (3/16, 19%).

Key framing quote (Sobering et al. 2022, HGG Adv, PMID:35469323) ✅:

"Loss-of-function variants in PHD Finger Protein 8 (PHF8) cause Siderius X-linked intellectual disability (ID) syndrome, hereafter called PHF8-XLID. PHF8 is a histone demethylase that is important for epigenetic regulation of gene expression. PHF8-XLID is an under-characterized disorder with only five previous reports describing different PHF8 predicted loss-of-function variants in eight individuals."

1.2 Identifiers

Resource	Identifier	Notes
MONDO	MONDO:0010286	syndromic X-linked intellectual disability Siderius type (matches draft entry)
OMIM (phenotype)	300263	INTELLECTUAL DEVELOPMENTAL DISORDER, X-LINKED, SYNDROMIC, SIDERIUS TYPE; MRXSSD
OMIM (gene)	300560	PHF8
Orphanet	ORPHA:85287	X-linked intellectual disability, Siderius type ⚠ (orpha.net blocked this session; code confirmed via GTR + GenCC)
MedGen / UMLS	C1846055 (MedGen UID 375779)	⚠ from GTR
HGNC	HGNC:20672 (`hgnc:20672` in dismech lowercase convention)	PHF8
NCBI Gene	23133	PHF8, Xp11.22
UniProt	Q9UPP1 (PHF8_HUMAN)	Histone lysine demethylase PHF8
Ensembl	ENSG00000172943	⚠ not independently re-verified
RefSeq transcript	NM_015107.3 (NM_015107.2 in older reports)
ICD-10 / ICD-11 / MeSH	❌ not verified	No specific MeSH descriptor exists; PubMed indexes under X-Linked Intellectual Disability (D038901) + Cleft Lip / Cleft Palate. Do not assert an ICD code without checking Orphanet's cross-reference table.

1.3 Synonyms

MRXSSD (OMIM preferred abbreviation)
Siderius X-linked mental retardation syndrome
Siderius–Hamel syndrome / Siderius‑Hamel cleft lip‑palate syndrome
Mental retardation, X-linked, syndromic, Siderius type (legacy OMIM title)
PHF8-XLID (preferred in Sobering et al. 2022 and increasingly the working name)
X-linked mental retardation with cleft lip/palate (descriptive, historical)
Gene aliases used as disease-adjacent labels: JHDM1F, KDM7B, ZNF422

1.4 Data provenance character

Information is aggregated disease-level and case-series derived, not EHR-derived. The entire literature base is ~29 affected individuals across ~15 families in 6 primary reports plus contiguous-deletion cases. There is no registry, no natural-history study, no EHR cohort, and no clinical trial for this disorder. Any prevalence, frequency, or prognostic statement in this report is therefore case-series-derived and subject to strong ascertainment bias (see §3.4).

2. Etiology

2.1 Primary causal factor

Monogenic, germline, X-linked recessive loss of PHF8 function. No environmental, infectious, or multifactorial etiology is established. The mechanism of pathogenicity is loss of function; both truncating variants and catalytically inactivating missense variants have been shown to abolish demethylase activity.

Laumonnier et al. 2005 (J Med Genet, PMID:16199551) ✅ — the founding molecular report:

"Truncating mutations were found in the PHF8 gene (encoding the PHD finger protein 8) in two unrelated families with X linked mental retardation (XLMR) associated with cleft lip/palate (MIM 300263)." "The association of XLMR and cleft lip/palate in these patients with mutations in PHF8 suggests an important function of PHF8 in midline formation and in the development of cognitive abilities, and links this gene to XLMR associated with cleft lip/palate."

Original linkage (Siderius et al. 1999, Am J Med Genet, PMID:10398231) ✅:

"A family is described in which X-linked mild to borderline mental retardation (MR) is associated with cleft lip/palate. Linkage analysis showed a maximum LOD score of Z=2.78 at straight theta=0.0 for the DXS441 locus with flanking markers DXS337 and DXS990, defining the region Xp11.3-q21.3 with a linkage interval of 25 cM."

2.2 Genetic risk factors

Causal: hemizygous PHF8 LoF variant in a male (see §4 for the variant catalogue).
Carrier status: heterozygous mother. In Sobering 2022, 10/12 tested mothers were unaffected carriers; 2 probands' mothers carried no variant (de novo) ⚠.
De novo occurrence: 3 of 11 LoF families in Sobering 2022 were de novo (c.596+1G>A, c.1627-1G>A, c.1965_1966dup) ⚠ — clinically important, since absence of family history does not exclude the diagnosis.
Contiguous gene deletions at Xp11.22 encompassing PHF8 ± FAM120C ± WNK3 produce an overlapping/expanded phenotype (Qiao et al. 2008, PMID:18498374, two brothers with ASD; De Wolf et al. 2014, PMID:25258334, "The deletion of PHF8 most likely explains the cleft palate and mild intellectual disability" ⚠; Huang et al. 2020, PMID:32219840, prenatal CMA detection of Xp11.22 deletion in a fetus with cleft lip and palate ⚠).
Modifier genes: none identified. A pathway-level candidate framework exists: PHF8 physically and functionally interacts with two other XLID proteins, ZNF711 and KDM5C/JARID1C (Kleine-Kohlbrecher et al. 2010, PMID:20346720) ✅ — "our results functionally link the XLMR gene PHF8 to two other XLMR genes, ZNF711 and JARID1C, indicating that MR genes may be functionally linked in pathways, causing the complex phenotypes observed in patients developing MR." Poeta et al. 2019 (PMID:31691806) place KDM5C at the crossroads of ARX, ZNF711 and PHF8 transcriptional axes ⚠. Whether variation in these partners modifies PHF8-XLID severity is untested.
X-inactivation as a female modifier: skewed XCI tested in 3 carrier mothers — 2 completely skewed, 1 uninformative ⚠ (Sobering 2022). This is the presumed explanation for carrier females being unaffected.

2.3 Environmental risk factors

None established for this disorder. There is one mechanistically motivated hypothesis worth curating as a knowledge gap, not as fact: because PHF8 is a 2‑oxoglutarate/Fe(II)/O₂-dependent oxygenase, its catalytic output is oxygen-dependent, and maternal hypoxia is an established modifier of cleft lip/palate risk in mice.

Loenarz et al. 2010 (Hum Mol Genet, PMID:19843542) ✅ (abstract):

"The dependence of PHF8 activity on oxygen availability is interesting because the occurrence of fetal cleft lip has been demonstrated to increase with maternal hypoxia in mouse studies. Cleft lip and other congenital anomalies are also linked indirectly to maternal hypoxia in humans, including from maternal smoking and maternal anti-hypertensive treatment."

and from the cached full text ✅:

"Episodes of maternal respiratory hypoxia in mice also correlate with increased incidence of cleft lip/palate (gestational day 10-11, 10% O2), while hyperoxia rescued mouse strains that are genetically susceptible for cleft lip/palate (gestational day 10-11, 50% O2) (5), suggesting that gestational oxygen levels mediate genetic and environmental factors."

Curation guidance: this is a plausible gene–environment interaction (hypomorphic PHF8 × gestational hypoxia → clefting penetrance) but has never been tested in PHF8-mutant humans or mice. Model it as a discussions entry with kind: KNOWLEDGE_GAP (or as a mechanistic_hypotheses group with status: EMERGING), not as an etiologic claim.

2.4 Protective factors

Genetic: none identified. No protective alleles, no gnomAD LoF-tolerant signal reported.
Environmental: none disease-specific. Periconceptional folic acid supplementation is protective for orofacial clefting in the general population but has not been studied in PHF8-related clefting — do not import that claim into this entry.
A counterintuitive "protective" finding in mice: Phf8-null mice show resilience to stress-induced anxiety/depression-like behavior (Walsh et al. 2017, PMID:28485378) ⚠ — "we report a striking resiliency to stress-induced anxiety- and depression-like behaviour on loss of Phf8." This has no demonstrated human counterpart and should be curated as evidence_source: MODEL_ORGANISM with an explicit HUMAN_MODEL_MISMATCH discussion (see §15.4).

3. Phenotypes

3.1 Cohort-derived frequency table (Sobering et al. 2022, n = 16 affected males with LoF variants) ⚠

All frequencies below are from the PMC full text of PMID:35469323 and must be re-verified against the paper before entering as frequency: with evidence (see docs/frequency-evidence-guidelines.md — a frequency band needs its own quantitative snippet).

Phenotype	Frequency (Sobering 2022)	Suggested HPO term	Suggested dismech `frequency` band
Developmental delay (global)	16/16 (100%)	HP:0001263 Global developmental delay	OBLIGATE / VERY_FREQUENT
Speech delay	16/16 (100%)	HP:0000750 Delayed speech and language development	OBLIGATE / VERY_FREQUENT
Intellectual disability (borderline→severe)	14/16 (88%)	HP:0001249 Intellectual disability	VERY_FREQUENT
Fine motor delay	14/16 (88%)	HP:0007010 Poor fine motor coordination ⚠ (term not verified)	VERY_FREQUENT
Gross motor delay (mean walking 20 mo)	12/16 (75%)	HP:0002194 Delayed gross motor development ⚠	FREQUENT
Hypertelorism	11/16 (69%)	HP:0000316 Hypertelorism ⚠	FREQUENT
Retrognathia	10/16 (63%)	HP:0000278 Retrognathia ⚠	FREQUENT
Infantile feeding difficulty	10/16 (63%)	HP:0011968 Feeding difficulties ⚠	FREQUENT
Elongated (long) face	8/16 (50%)	HP:0000276 Long face	FREQUENT
Autism spectrum disorder	7/16 (44%)	HP:0000717 Autism ⚠	FREQUENT
ADHD	7/16 (44%)	HP:0007018 Attention deficit hyperactivity disorder ✅ (label verified via OLS)	FREQUENT
Microcephaly	6/16 (38%)	HP:0000252 Microcephaly	OCCASIONAL/FREQUENT
High-arched palate	6/16 (38%)	HP:0000218 High palate ⚠	OCCASIONAL
Seizures	5/16 (31%)	HP:0001250 Seizure ⚠	OCCASIONAL
Low-set ears (4 posteriorly rotated)	5/16 (31%)	HP:0000369 Low-set ears ⚠	OCCASIONAL
Orofacial clefting	3/16 (19%)	HP:0410030 Cleft lip ✅ / HP:0000175 Cleft palate	OCCASIONAL

3.2 HPO disease annotations currently attached to OMIM:300263

Retrieved from the HPO annotation API (ontology.jax.org) ⚠ — note these derive from the older, clefting-ascertained families, which is why cleft frequencies are far higher than in Sobering:

HPO ID	Label	Annotated frequency
HP:0001256	Mild intellectual disability	7/7
HP:0000276	Long face	4/4
HP:0410030	Cleft lip	6/8
HP:0000175	Cleft palate	5/8
HP:0000455	Broad nasal tip	3/7
HP:0001176	Large hands	2/7
HP:0001763	Pes planus	1/7
HP:0000252	Microcephaly	1/1
HP:0001249	Intellectual disability	1/1
HP:0000750	Delayed speech and language development	—
HP:0000340	Sloping forehead	—
HP:0000582	Upslanted palpebral fissure	—
HP:0000336	Prominent supraorbital ridges	—
HP:0000664	Synophrys	—
HP:0002162	Low posterior hairline	—
HP:0001611	Hypernasal speech	—
HP:0002942	Thoracic kyphosis	—
HP:0001166	Arachnodactyly	—
HP:0010511	Long toe	—
HP:0001419	X-linked recessive inheritance	—

Additional features named in OMIM/GTR/MalaCards summaries but not in the HPO table above ⚠: cryptorchidism, preaxial polydactyly, broad nasal bridge. Treat these as low-confidence single-case observations.

3.3 Neuroimaging phenotypes (6 individuals imaged, Sobering 2022) ⚠

1 normal
1 mildly increased subarachnoid space + white matter changes
2 (identical twins) polymicrogyria and cortical dysplasia → HP:0002126 Polymicrogyria ⚠
1 abnormal striatal signal (caudate/globus pallidus)
1 cranio-occipital malformation + thin corpus callosum → HP:0033725 / HP:0002079 ⚠

The twin polymicrogyria observation is mechanistically interesting given the 2026 finding that PHF8 loss arrests neurogenesis in mouse embryos (§6.6), but n = 2 in one family — curate as OCCASIONAL at most.

3.4 The clefting discordance — flag this explicitly in the KB entry

The single most important curation nuance for this disease: earlier cohorts were ascertained because of clefting (Koivisto et al. 2007 screened 7,712 cleft-surgery patients), so cleft frequency in the pre-2022 literature (~70–75%) is inflated by design. Sobering et al. found 3/16 (19%) and concluded ⚠:

"Orofacial clefting was seen in three individuals from our cohort, suggesting that this feature is less common than previously reported."

Recommendation for the dismech entry: annotate cleft lip/palate as a characteristic but not obligate feature (frequency: OCCASIONAL per Sobering) with a notes: field recording the ascertainment-bias discordance, and cite both frequency sources.

3.5 Phenotype characteristics

Onset: clefting is congenital (HP:0003577 Congenital onset ⚠); developmental delay is apparent in infancy/early childhood (HP:0011463 Childhood onset ⚠); facial gestalt evolves with age (elongated face "tends to worsen with age" ⚠).
Severity: classically mild-to-borderline ID; Sobering broadened this to borderline through severe, with two individuals having no ID (one with dyscalculia, one with mild learning difficulties) ✅ — "All affected individuals exhibited developmental delay and all but two had borderline to severe ID."
Progression: static / non-progressive neurodevelopmental disorder. No neurodegeneration, no regression, and no progressive organ involvement has been reported in any of the ~29 published individuals. Use clinical_course: STABLE for the ID node.
Frequency among affected individuals: see tables above.

3.6 Quality-of-life impact

❌ No disease-specific QoL data exist (no EQ-5D, SF-36, PROMIS, or disease-specific instrument has been applied). Inferable, non-citable impacts: educational support needs (universal DD), speech intelligibility (hypernasal speech + clefting), feeding difficulty in infancy (63%), behavioral/social burden of ASD+ADHD (~44% each), and seizure burden (31%). Do not assert QoL numbers.

4. Genetic / Molecular Information

4.1 Gene

PHF8 (PHD finger protein 8), Xp11.22, NCBI Gene 23133, HGNC:20672, OMIM 300560. Aliases: JHDM1F, KDM7B, ZNF422, KIAA1111, MRXSSD.

Protein: UniProt Q9UPP1, canonical isoform 1,060 aa (Sobering et al. describe the disease-relevant isoform as 1,024 aa; multiple transcript variants exist — note the discrepancy when curating). Domains: N-terminal PHD finger (H3K4me3 reader), central JmjC catalytic domain (aa 231–387 per UniProt), multiple nuclear localization signals, and a serine-rich region. Subcellular localization: nucleus and nucleolus; "recruited to H3K4me3 sites on chromatin during interphase" and dissociates during mitosis ⚠ (UniProt).

4.2 Reported pathogenic variants

Previously reported (the "five previous reports," 8 individuals) — as tabulated in Sobering et al. Table 1 ⚠:

cDNA	Protein	Type	Original report
c.943_946+8del	p.(Thr315Leufs*25)	frameshift/splice-region deletion	Siderius et al. 1999 family (molecularly solved by Laumonnier 2005)
c.631C>T	p.(Arg211*)	nonsense	Laumonnier et al. 2005 (PMID:16199551)
c.529A>T	p.(Lys177*)	nonsense	Abidi et al. 2007 (PMID:17594395)
c.836C>T	p.(Phe279Ser)	missense, catalytically dead	Koivisto et al. 2007 (PMID:17661819)
c.144C>A	p.(Tyr48*)	nonsense	Ibarluzea et al. ⚠ (year cited as 2013 in the table; verify)

New in Sobering et al. 2022 (11 LoF variants / 11 families, 16 males) ⚠:

cDNA	Protein	Type	Inheritance
del exons 9–10	p.Gly316_Arg380del	intragenic deletion	maternal
c.294-1820_597-603del	p.Ser98_Thr198del	intragenic deletion	maternal
c.596+1G>A	—	splice donor	de novo
c.1627-1G>A	—	splice acceptor	de novo
c.1731-2A>G	—	splice acceptor	maternal
c.862C>T	p.(Gln288*)	nonsense	maternal
c.1030C>T	p.(Gln343*)	nonsense	maternal
c.738_739insT	p.(His247Serfs*3)	frameshift	maternal
c.1965_1966dup	p.(Glu656Valfs*174)	frameshift	de novo
c.1996delG	p.(Glu666Argfs*163)	frameshift	maternal
c.2760dupC	p.(Thr921Hisfs*19)	frameshift	maternal

VUS reported (5 individuals / 4 families, all maternally inherited missense) ⚠: c.143A>G p.(Tyr48Cys); c.257C>T p.(Thr86Met); c.808C>T p.(Arg270Cys); c.1150G>A p.(Glu384Lys). Sobering explicitly separates these from the LoF cohort — do not curate them as pathogenic.

A benign in-frame variant to know about: p.Ser969del — "it is abundantly found in gnomAD in both heterozygous males and homozygous females" ⚠ (Sobering 2022). Useful as a negative control / interpretation caveat.

4.3 Variant classification, spectrum, and origin

Origin: germline, exclusively. No somatic PHF8 disease variants relevant to this disorder (somatic/overexpression PHF8 biology is a cancer story — see §6.9 — and must be kept out of the disease mechanism graph).
Types: nonsense, frameshift, canonical splice-site, intragenic multi-exon deletions, contiguous Xp11.22 deletions, and one recurrent catalytically inactivating missense (F279S). The older literature emphasized that mutations cluster in the JmjC-encoding exons; the 2022 cohort shows LoF variants distributed across the gene, including C-terminal frameshifts.
Functional consequence: loss of function — via (a) truncation removing the JmjC domain and NLSs, (b) catalytic inactivation, and (c) mislocalization.

Loenarz et al. ✅ on both loss-of-catalysis and mislocalization:

"Clinically observed mutations to the PHF8 gene cluster in exons encoding for the double stranded beta-helix fold and will therefore disrupt catalytic activity." "This mutant encodes a F279S variant of PHF8 that modifies a conserved hydrophobic region; assays with both peptides and intact histones reveal this variant to be catalytically inactive." (full text) "In contrast to the wild-type HA-PHF8, analogous studies revealed that the clinically observed F279S variant did not show clear nuclear localization, with apparent cytoplasmic localization."

Abidi et al. 2007 ✅ on the truncation mechanism:

"The mutation results in a truncated PHF8 protein lacking the Jumonji-like C terminus domain and five nuclear localization signals."

Koivisto et al. 2007 ✅ on F279S:

"A novel missense mutation c.836C>T of the PHF8 gene was identified in a Finnish family with multiple-affected male patients. The mutation resides in exon 8 and changes phenylalanine to serine (F279S) in the functionally important Jmonji C domain of the protein." "The mutation was not present in 200 anonymous blood donors (approximately 300 X-chromosomes)."

Independent confirmation of F279S catalytic death: Qi et al. 2010 ✅ ("a point mutation in the catalytic domain (phenylalanine to serine, F279S) abolished PHF8 demethylase activities"); Qiu et al. 2010 ✅ ("a mutant PHF8 (phenylalanine at position 279 to serine) identified in the XLMR patients is defective in enzymatic activity, indicating that the loss of histone demethylase activity is causally linked with the onset of disease"); Fortschegger et al. 2010 ✅ ("a PHF8 disease mutant was defective in demethylation and in coactivation").

4.4 Population / database data

ClinVar: 537 total records for PHF8[gene]; 251 returned by a pathogenic[clinical significance] filter ⚠ — this filter over-counts (it captures likely-pathogenic and conflicting interpretations); do not quote 251 as "pathogenic variants." Get the exact breakdown from ClinVar Miner or a fresh ClinVar query before curating.
gnomAD constraint (pLI / LOEUF / o-e LoF): ❌ not retrievable this session (gnomAD's GraphQL endpoint requires POST; DECIPHER blocked). This must be filled in manually before curating any constraint claim.
Allele frequency of pathogenic variants: all reported pathogenic variants are private/family-specific. No founder allele is known. F279S absent from ~300 control X chromosomes (Koivisto) ✅.
ClinGen Gene-Disease Validity: reported as Definitive for PHF8 – syndromic X-linked intellectual disability Siderius type, X-linked, supported by ≥7 variants from 5 publications plus mouse, zebrafish and cell-culture models with rescue; mechanism = loss of function ⚠. Recommended action: rather than citing a web page, run just clingen-refresh and cite the CGGV: assertion as a structured-source evidence row.

4.5 Epigenetic information

Two distinct senses matter here and should not be conflated in the KB:

PHF8 is itself an epigenetic effector — its loss changes the histone-methylation landscape (see §6). This is the disease mechanism.
Episignature (methylation signature) for diagnostic classification: ❌ No published DNA-methylation episignature for PHF8-XLID was found. Many chromatinopathies (Kabuki, Sotos, CHARGE, etc.) have EpiSign classifiers; PHF8 does not appear among them in the retrieved literature. This is a genuine, curatable knowledge gap and a natural proposed_experiments item.

4.6 Chromosomal abnormalities

Xp11.22 microdeletions removing PHF8 (± FAM120C, WNK3) — Qiao 2008 (PMID:18498374, familial deletion in two brothers with ASD; "complete deletion of the plant homeodomain finger protein 8 (PHF8) gene" ⚠), De Wolf 2014 (PMID:25258334, syndromic autism ⚠), Huang 2020 (PMID:32219840, prenatal detection in a fetus with cleft lip and palate ⚠). These support CMA as a diagnostic modality and support haploinsufficiency/nullisomy as the mechanism.

5. Environmental Information

Environmental factors: none established. The only mechanistically grounded candidate is gestational hypoxia acting on an O₂-dependent 2‑OG oxygenase (§2.3) — hypothesis only.
Lifestyle factors: none disease-specific. Maternal smoking is cited by Loenarz et al. only as a general hypoxia-mediated cleft risk factor in the population, not as a PHF8 interaction.
Infectious agents: not applicable.
Toxicological note (model-organism only, do not curate as human etiology): chronic exposure to the PFAS replacement GenX induced transgenerational motor deficits via the C. elegans PHF8 ortholog jmjd-1.2 (PMID:40803444, Environ Pollut 2025) ⚠. Interesting as a conservation-of-mechanism datapoint (§14), not as a human risk factor.

6. Mechanism / Pathophysiology

6.1 The core enzymatic lesion (MOLECULAR scale)

PHF8 is a PHD-finger reader + JmjC-domain eraser: it binds H3K4me3 at active promoters through its PHD finger and removes repressive mono-/di-methyl marks through its Fe(II)/2‑OG-dependent JmjC domain, thereby acting as a transcriptional coactivator.

Loenarz et al. 2010 ✅:

"We report that recombinant PHF8 is an Fe(II) and 2-oxoglutarate-dependent N(epsilon)-methyl lysine demethylase, which acts on histone substrates. PHF8 is selective in vitro for N(epsilon)-di- and mono-methylated lysine residues and does not accept trimethyl substrates." (full text) "our results reveal that PHF8 is a 2OG oxygenase with selectivity for H3K9me2/me1, H3K27me2 and H3K36me2 residues."

Kleine-Kohlbrecher et al. 2010 ✅:

"the XLMR protein PHF8 and a C. elegans homolog F29B9.2 catalyze demethylation of di- and monomethylated lysine 9 of histone H3 (H3K9me2/me1). The PHD domain of PHF8 binds to H3K4me3 and colocalizes with H3K4me3 at transcription initiation sites."

Qi et al. 2010 (Nature) ✅ — the H4K20me1 activity and the genome-wide picture:

"Here we provide multiple lines of evidence establishing PHF8 as the first mono-methyl histone H4 lysine 20 (H4K20me1) demethylase, with additional activities towards histone H3K9me1 and me2." "PHF8 is located around the transcription start sites (TSS) of approximately 7,000 RefSeq genes and in gene bodies and intergenic regions (non-TSS). PHF8 depletion resulted in upregulation of H4K20me1 and H3K9me1 at the TSS and H3K9me2 in the non-TSS sites, respectively, demonstrating differential substrate specificities at different target locations." "Importantly, patient mutations significantly compromised PHF8 catalytic function."

Fortschegger et al. 2010 ✅ — the RNAPII coupling:

"Chromatin immunoprecipitation followed by high-throughput sequencing indicated that PHF8 is enriched at the transcription start sites of many active or poised genes, mirroring the presence of RNA polymerase II (RNAPII) and of H3K4me3-bearing nucleosomes... we present evidence for direct interaction of PHF8 with the C-terminal domain of RNAPII."

Feng et al. 2010 (PMID:20208542) adds the rDNA/nucleolar arm: PHF8 "activates transcription of rRNA genes through H3K4me3 binding and H3K9me1/2 demethylation" ⚠.

Suggested GO annotations (verify all with OAK before use):

GO ID	Label	Verification
GO:0032454	histone H3K9 demethylase activity	✅ label verified via OLS
GO:0035575	histone H4K20 demethylase activity	✅ label verified via OLS
GO:0140457 / GO:0071558	H3K27 demethylase activity	⚠ ID not verified — check
GO:0006338	chromatin remodeling	⚠
GO:0045893	positive regulation of DNA-templated transcription	⚠
GO:0006360	transcription by RNA polymerase I	⚠ (rDNA arm)
GO:0060021	roof of mouth development	✅ label verified via OLS
GO:0000082	G1/S transition of mitotic cell cycle	⚠
GO:0032008	positive regulation of TOR signaling	⚠
GO:0006564	L-serine biosynthetic process	⚠
GO:0006914	autophagy	⚠

CHEBI cofactors/substrates: 2-oxoglutarate (CHEBI:16810 ⚠), iron(2+) (CHEBI:29033 ⚠), dioxygen (CHEBI:15379 ⚠), L-serine (CHEBI:17115 ⚠), sirolimus/rapamycin (CHEBI:9168 ✅ verified).

6.2 Downstream arm 1 — neuronal/cognitive: the mTOR–RSK1 axis

Chen et al. 2018 (Nat Commun, PMID:29317619) ✅ is the strongest mechanistic chain from gene to cognition, and it is druggable:

"Here we report that Phf8 knockout mice displayed impaired learning and memory, and impaired hippocampal long-term potentiation (LTP) without gross morphological defects. We also show that mTOR signaling pathway is hyperactive in hippocampus in Phf8 knockout mouse. Mechanistically, we show that demethylation of H4K20me1 by Phf8 results in transcriptional suppression of RSK1 and homeostasis of mTOR signaling. Pharmacological suppression of mTOR signaling with rapamycin in Phf8 knockout mice recovers the weakened LTP and cognitive deficits."

Causal chain: PHF8 loss → failure to demethylate H4K20me1 at RSK1 → RSK1 de-repression → mTOR hyperactivation in hippocampus → impaired LTP → learning/memory deficit (rescued by rapamycin). Note the paper places PHF8 in the same "mTORopathy-adjacent" company as tuberous sclerosis, Fragile X and Down syndrome ✅ (intro).

6.3 Downstream arm 2 — neuronal differentiation and cytoskeleton

RAR coactivation / neuronal differentiation — Qiu et al. 2010 (Cell Res, PMID:20548336) ✅: "knockdown of PHF8 in mouse embryonic carcinoma P19 cells impairs RA-induced neuronal differentiation, whereas overexpression of the wild-type but not the F279S mutant PHF8 drives P19 cells toward neuronal differentiation. Furthermore, we show that PHF8 interacts with RARalpha and functions as a coactivator for RARalpha."
Cytoskeleton / neurite outgrowth — Asensio-Juan et al. 2012 (NAR, PMID:22850744) ✅: "PHF8 controls the expression of genes involved in cell adhesion and cytoskeleton organization such as RhoA, Rac1 and GSK3β... Further analysis in neurons shows that depletion of PHF8 results in down-regulation of cytoskeleton genes and leads to a deficient neurite outgrowth. Overall, our results suggest that the mental retardation phenotype associated with loss of function of PHF8 could be due to abnormal neuronal connections as a result of alterations in cytoskeleton function."
REST/NRSF co-occupancy at neuronal gene promoters — Wang et al. 2014 (Sci Rep, PMID:24852203) ✅ (title/abstract: "PHF8 and REST/NRSF co-occupy gene promoters to regulate proximal gene expression").

6.4 Downstream arm 3 — glial biology (newer, and under-appreciated)

Astrocytes — Iacobucci et al. 2021 (Development, PMID:34081130) ✅: "we investigate the contribution of the XLID-associated histone demethylase PHF8 to astrocyte differentiation and function. Using genome-wide analyses and biochemical assays in mouse astrocytic cultures, we reveal a regulatory crosstalk between PHF8 and the Notch signaling [pathway]…"
Oligodendrocytes — Kremp et al. 2024 (Glia, PMID:38613395) ✅: "Phf8 promotes the proliferation of rodent oligodendrocyte progenitor cells and impairs their differentiation to oligodendrocytes... Phf8 has a strong positive impact on Olig2 expression by acting on several regulatory regions of the gene and changing their histone modification profile... We conclude that Phf8 may impact nervous system development at least in part through its action in oligodendroglial cells." Companion review: Kremp & Wegner 2026, Neural Regen Res (PMID:40145966) ⚠.

This gives a non-neuronal (glial) contribution to the neurodevelopmental phenotype — worth its own pathophysiology node with CL:0002453 (oligodendrocyte precursor cell ✅ verified) and CL:0000127 (astrocyte ⚠).

6.5 Downstream arm 4 — craniofacial / midline development

Qi et al. 2010 ✅ provides the only direct in-vivo craniofacial mechanism:

"PHF8 regulates cell survival in the zebrafish brain and jaw development, thus providing a potentially relevant biological context for understanding the clinical symptoms associated with PHF8 patients. Lastly, genetic and molecular evidence supports a model whereby PHF8 regulates zebrafish neuronal cell survival and jaw development in part by directly regulating the expression of the homeodomain transcription factor MSX1/MSXB, which functions downstream of multiple signalling and developmental pathways. Our findings indicate that an imbalance of histone methylation dynamics has a critical role in XLMR."

MSX1 is a well-established human orofacial-clefting gene, which makes the PHF8→MSX1 link the most credible molecular explanation for the clefting phenotype. Kremp et al. 2024 ✅ independently characterize Phf8 as "implicated by mutation in mice and humans in neural crest defects and neurodevelopmental disturbances." Bone-forming relevance: Han et al. 2015 (PMID:25923143) ⚠ — "PHF8, a major H4K20/H3K9 demethylase, plays a critical role in craniofacial and bone development."

Causal chain: PHF8 loss → H4K20me1/H3K9me1-2 accumulation at MSX1/msxb → reduced MSX1 expression in cranial neural crest-derived facial primordia → impaired fusion of facial prominences / palatal shelves → cleft lip ± cleft palate, in parallel with increased apoptosis of brain cells.

6.6 Downstream arm 5 — metabolic control of neurogenesis (2026, newest)

Artés/Iacobucci et al. 2026 (EMBO Rep, PMID:41714361) ⚠ identifies "PHF8 as a key driver of the serine biosynthesis pathway, safeguarding the intracellular serine pool essential for neural progenitor proliferation," with PHF8 depletion causing disrupted metabolism, blocked autophagy, replication defects and proliferation arrest, and PHF8 deficiency halting neurogenesis and brain development in mouse embryos. This is the newest and arguably most important mechanistic advance for this disease — it supplies a proliferation/metabolism node upstream of the cognitive phenotype and is consistent with the polymicrogyria seen in the twins. Verify the abstract before quoting.

6.7 Downstream arm 6 — cell cycle, DNA damage, transcription recovery

Cell cycle: PHF8 controls the G1–S transition with E2F1/HCF-1/SET1A via H4K20me1 demethylation (Liu et al. 2010, Nature, PMID:20622854) ⚠.
DNA damage: Kim et al. 2024 (NAR, PMID:39087553) ⚠ — PHF8 is "the major demethylase that reverses transcriptionally repressive epigenetic modification laid down by the DYRK1B-EHMT2 pathway," concentrating at damage tracks and promoting "timely resolution of local H3K9me2 to facilitate the resumption of transcription," including at rDNA. Relevance to the neurodevelopmental phenotype is unestablished — curate as mechanistic context, not as disease mechanism.

6.8 Immune / metabolic / tissue-damage dimensions

Immune: no immunodeficiency or autoimmunity in patients. A 2026 paper (PMID:41709745) ⚠ reports "the Znf711-Phf8 complex functions as a transcriptional rheostat essential for neutrophil development" in a model system — no human hematologic phenotype has been reported in PHF8-XLID, so this is a prediction to watch, not a curatable human phenotype.
Metabolic: the serine-biosynthesis finding (§6.6) is the only metabolic axis. No metabolic decompensation phenotype; do not conform this entry to metabolic_intoxication_decompensation.
Tissue damage: none — this is a developmental, not a degenerative or inflammatory, disorder. No fibrosis, ischemia, oxidative-injury or necrosis mechanism applies.

6.9 ⚠ Scope guardrail: the PHF8 cancer literature

A large and growing body of work (≥20 papers in 2025–2026 alone: TNBC, gastric, prostate, hepatocellular, colorectal) treats PHF8 as an overexpressed oncogenic driver and drug target. This is the opposite direction of effect from the disease (LoF), involves somatic/expression biology rather than germline lesions, and must not be imported into this disease entry's pathophysiology graph. If any of it is curated at all, it belongs in a separate context with an explicit note that the disease is loss-of-function.

6.10 Suggested pathophysiology node chain for the dismech entry

1. PHF8 Loss-of-Function Variant (MOLECULAR)
     ↓ CAUSES
2. Loss of JmjC Histone Demethylase Activity (MOLECULAR)
     [H3K9me1/2, H4K20me1, H3K27me2 fail to be erased; PHD-H3K4me3 reading uncoupled from erasure]
     ↓ CAUSES
3. Failure of PHF8-Dependent Transcriptional Coactivation at Target Promoters (MOLECULAR)
     ↓ branches into:
   3a. RSK1 de-repression → mTOR hyperactivation (CELLULAR)  → impaired hippocampal LTP (CELLULAR/TISSUE) → ID, learning/memory deficit (ORGANISM)
   3b. Impaired serine biosynthesis / autophagy block in neural progenitors (CELLULAR) → arrested neurogenesis (TISSUE) → ID ± cortical malformation (ORGANISM)
   3c. Impaired RAR-dependent neuronal differentiation + RhoA/Rac1/GSK3β cytoskeletal gene loss (CELLULAR) → deficient neurite outgrowth/connectivity (CELLULAR) → ID/ASD (ORGANISM)
   3d. Impaired Olig2 induction in OPCs / Notch-crosstalk in astrocytes (CELLULAR) → altered glial development (TISSUE) → contributory NDD (ORGANISM)
   3e. Reduced MSX1/msxb in cranial neural crest derivatives (MOLECULAR/CELLULAR) → failed fusion of facial prominences and palatal shelves (TISSUE) → cleft lip/palate (ORGANISM)

Treatment link for the drug-target pattern: an mTOR-inhibitor node (target_mechanisms with INHIBITS on node 3a) is the natural place to hang rapamycin — but flag it as preclinical-only (mouse, no human data).

Module conformance assessment: no existing kb/modules/ module is a clean fit. pharyngeal_arch_patterning_serial_homology is tempting for the craniofacial arm but is the wrong mechanism — Siderius clefting is a midline fusion failure (lip/palate), not a serially homologous arch-derivative bundle (mandible+maxilla+zygoma+ear), and the lesion is chromatin-level, not an arch-identity code. Recommend no conforms_to rather than a forced fit. A future "chromatinopathy / histone-modifier neurodevelopmental disorder" module would be the right home; note this as a module-creation candidate.

7. Anatomical Structures Affected

7.1 Organ level

Structure	Involvement	UBERON (⚠ all need OAK verification)
Brain	Primary — cognition, LTP, neurogenesis	UBERON:0000955 brain
Hippocampus	Primary — LTP deficit locus (mouse)	UBERON:0002421 hippocampal formation
Cerebral cortex	Polymicrogyria/cortical dysplasia in 2/6 imaged	UBERON:0000956 cerebral cortex
Prefrontal cortex	Serotonin-signaling dysregulation (mouse)	UBERON:0000451 prefrontal cortex
Corpus callosum	Thin in 1/6 imaged	UBERON:0002336 corpus callosum
Striatum (caudate, globus pallidus)	Abnormal signal in 1/6	UBERON:0002435 striatum
Upper lip	Cleft lip	UBERON:0001834 upper lip
Palate / secondary palate	Cleft palate, high-arched palate	UBERON:0001716 secondary palate
Face / craniofacial skeleton	Dysmorphism, retrognathia, elongated face	UBERON:0001456 face
Ear (external)	Low-set, posteriorly rotated	UBERON:0001690 ear
Hands / feet	Large hands, arachnodactyly, long toes, pes planus	UBERON:0002398 manus / UBERON:0002387 pes
Vertebral column	Thoracic kyphosis	UBERON:0002415 vertebral column ⚠
Testis	Cryptorchidism (rare)	UBERON:0000473 testis

Body systems: nervous (primary), craniofacial/musculoskeletal (secondary), and — in a minority — genitourinary. No cardiac, renal, hepatic, hematologic, immune, endocrine or ophthalmologic involvement is reported.

Lateralization: clefting may be unilateral or bilateral (Koivisto's two siblings differed: one unilateral CL/P, one bilateral CL/P ✅) — a nice intrafamilial-variability datapoint. Facial asymmetry is noted as "mild" in Sobering ⚠.

7.2 Tissue and cell level

Cell type	Role	CL term
Central nervous system neuron	Target of impaired differentiation, neurite outgrowth, LTP	CL:2000029 ✅ verified
Neural progenitor / neural stem cell	Proliferation arrest via serine pathway (2026)	CL:0000047 neural stem cell ⚠
Oligodendrocyte precursor cell	Phf8 promotes proliferation, Olig2 effector	CL:0002453 ✅ verified
Oligodendrocyte	Differentiation altered	CL:0000128 ⚠
Astrocyte	Differentiation/function via Notch crosstalk	CL:0000127 ⚠
Cranial neural crest cell / migratory neural crest cell	Craniofacial/clefting arm (zebrafish, mouse)	CL:0000333 ⚠
Bone marrow stromal cell / osteoblast lineage	Craniofacial bone formation (Han 2015)	CL:0000134 / CL:0000062 ⚠
Neutrophil (model-only, no human phenotype)	Znf711-Phf8 rheostat	CL:0000775 ⚠

7.3 Subcellular level

Nucleus (GO:0005634 ⚠) — primary site of PHF8 action; chromatin (GO:0000785 ⚠); nucleolus (GO:0005730 ⚠) for the rDNA/RNA Pol I arm.
Pathogenic relevance: F279S and truncating variants that delete the NLSs cause cytoplasmic mislocalization — the subcellular compartment shift is itself part of the molecular pathology ✅ (Loenarz full text).

8. Temporal Development

Onset:
Prenatal/congenital: orofacial clefting (detectable on prenatal ultrasound/CMA — Huang 2020 ⚠); brain malformations where present.
Infancy: feeding difficulties (63%), hypotonia/motor delay; mean age at independent walking 20 months ⚠.
Early childhood: speech delay (100%), then ID; ASD/ADHD typically recognized in preschool/school years.
Onset pattern: congenital/insidious developmental, not acute.
Progression: static (non-progressive). There is no reported regression, neurodegeneration, or organ deterioration. The facial gestalt evolves (long face becomes more marked with age ⚠) — that is a morphologic evolution, not disease progression.
Course: chronic, lifelong.
Stages: no staging system exists or is applicable.
Remission: not applicable.
Critical periods:
Weeks 4–10 of gestation — lip and palatal fusion; the only window in which the clefting phenotype could theoretically be modified (and the window in which the hypoxia hypothesis would operate).
Perinatal-to-early-childhood — the intervention window for cleft repair, feeding support, and early developmental/speech intervention, where outcome is genuinely modifiable.
Preclinically, the Chen 2018 rapamycin rescue was performed in adult mice and restored LTP and cognition — implying, in mice, a post-developmental window of reversibility. That is a striking and citable claim, but strictly MODEL_ORGANISM evidence.

9. Inheritance and Population

9.1 Epidemiology

Prevalence: ❌ no formal estimate published. Orphanet classifies it in its rare-disease register; based on ~29 reported individuals worldwide, the appropriate dismech prevalence_class is BELOW_1_IN_1000000 or, more defensibly, NOT_YET_DOCUMENTED with measure_type: CASES_IN_LITERATURE and a count. Recommended structured record: ```yaml prevalence:
population: Worldwide measure_type: CASES_IN_LITERATURE prevalence_class: NOT_YET_DOCUMENTED notes: >- Approximately 29 affected males from ~15 families reported: 8 individuals in five reports before 2022, plus 16 individuals from 11 families in Sobering et al. 2022 (plus 5 individuals with VUS from 4 families). ```
Incidence: ❌ unknown.
Denominator context (do not attribute to this disorder): X-linked ID overall affects ~1–4 per 2,000 males — Qi et al. 2010 ✅: "XLMR affects 1–4 out of 2,000 males, causing intellectual disability (Intelligence Quotient (IQ) <70)". PHF8 is a rare cause within that.
Screening-based frequency estimate: Koivisto et al. sequenced PHF8 in 18 selected patients drawn from a nationwide cohort of 7,712 cleft-surgery patients and found one family ✅ — i.e., PHF8 is a very rare cause of syndromic clefting.

9.2 Genetic epidemiology

Inheritance: X-linked recessive (HP:0001419 ✅ in HPO annotations; GenCC/ClinGen submissions use HP:0001417 X-linked inheritance). Suggested dismech inheritance_term: HP:0001419 X-linked recessive inheritance (bind the term: — do not leave preferred_term alone).
Penetrance: appears complete in hemizygous males for developmental delay (16/16) and near-complete for ID (14/16); incomplete/variable for clefting (3/16 in the unbiased cohort). Carrier females: essentially non-penetrant.
Expressivity: highly variable, including within families. Abidi et al. ✅ documents intrafamilial variability directly: "One of the truncating mutations was found in the original family with Siderius-Hamel CL/P syndrome where only two of the three affected individuals had mental retardation (MR) with CL/P and one individual had mild MR. The second mutation was present in a family with four affected men, three of whom had MR and CL/P, while the fourth individual had mild MR without clefting."
Genotype–phenotype correlation: none identified. Sobering et al. ⚠: "The phenotypic variability does not appear to be linked to the variant location in individuals who harbor a null allele." Variants inside vs. outside the JmjC domain did not differ in severity.
Anticipation: not applicable (not a repeat-expansion disorder).
Germline mosaicism: ❌ not reported; cannot be excluded (relevant to recurrence counseling after an apparently de novo variant).
Founder effects: none. The Finnish F279S family is a single family, not a founder population effect.
Consanguinity: not a factor (X-linked recessive in males).
Carrier frequency: ❌ unknown; no population carrier-screening data.

9.3 Population demographics

Sex ratio: affected individuals are essentially all male. Carrier females are typically unaffected (skewed XCI). ❌ No symptomatic heterozygous female has been convincingly reported in the retrieved literature — this is worth stating explicitly and worth flagging as something to re-check as cohorts grow.
Ethnic/geographic distribution: no predilection. Sobering's 11 families were of diverse ancestry: 9 Western European, 3 Moroccan, 2 Asian-Indian, 1 Afro-Caribbean (family counts as reported) ⚠. Prior families were Dutch (Siderius), French/Italian (Laumonnier), American (Abidi), Finnish (Koivisto).
Age distribution: lifelong from birth; published individuals span infancy to adulthood.

10. Diagnostics

10.1 Genetic testing (the diagnostic backbone)

There is no biochemical or imaging test that establishes this diagnosis — it is molecular.

Modality	Utility
Exome sequencing (ES)	First-line. Most 2022-cohort diagnoses came from ES (many via GeneDx / UW Center for Mendelian Genomics) ⚠. Detects nonsense, frameshift, splice-site, and missense variants.
Genome sequencing (GS)	Adds deep-intronic and structural resolution; the c.294-1820_597-603del intragenic deletion illustrates why breakpoint-capable methods matter ⚠.
XLID / ID multigene panels	PHF8 is included on XLID and broad ID/ASD panels; GTR lists 17 clinical tests for this condition (14 sequence analysis, 10 del/dup, 2 homozygosity, 1 targeted variant) ⚠.
Single-gene PHF8 sequencing	Reasonable when the classic gestalt (male, ID + CL/P + long face) is present, or for targeted familial variant testing.
Deletion/duplication analysis (MLPA / exon-level array / read-depth)	Essential — intragenic multi-exon deletions (exons 9–10) occur and are missed by sequencing-only pipelines.
Chromosomal microarray (CMA)	Detects the Xp11.22 contiguous deletions; also the route to prenatal detection (PMID:32219840) ⚠.
Karyotype / FISH	Low yield; FISH used historically to confirm CMA calls (Qiao 2008) ⚠.
mtDNA testing, repeat-expansion testing	Not applicable.
RNA sequencing	Potentially useful for the three canonical-splice variants (functional confirmation of aberrant splicing) — not reported as used in any published case; a reasonable `proposed_experiments` item.
Methylation episignature (EpiSign)	❌ Not available for PHF8. Genuine diagnostic gap for VUS resolution — Sobering reported 5 individuals with unresolved missense VUS who would be the exact use case.
Proteomics / metabolomics / liquid biopsy	❌ not applicable/not developed.

Carrier testing: targeted familial variant testing in the mother; X-inactivation studies are informative but not diagnostic.

10.2 Clinical, imaging and functional testing (supportive, not diagnostic)

Brain MRI: variable and non-specific — normal in some, polymicrogyria/cortical dysplasia, thin corpus callosum, striatal signal change, or increased subarachnoid space in others (6 imaged) ⚠. Justified where seizures or focal findings exist.
EEG: indicated for the ~31% with seizures ⚠.
Developmental/neuropsychological assessment: to quantify ID severity and identify ASD/ADHD (formal ASD and ADHD assessment should be routine given ~44% each).
Audiology and speech assessment: required with clefting (velopharyngeal insufficiency → hypernasal speech, HP:0001611) and given universal speech delay.
Laboratory tests / biomarkers: ❌ none. No enzyme assay, no metabolite, no circulating biomarker. Histone-methylation levels in patient cells are a research measure, not a clinical test.
Biopsy / histopathology: not indicated.

10.3 Clinical criteria and differential diagnosis

Standardized diagnostic criteria: ❌ none published. Diagnosis = compatible phenotype + hemizygous PHF8 LoF variant.
Differential diagnosis:
Other XLID chromatinopathies, especially the functionally linked ones: KDM5C/JARID1C (Claes-Jensen; microcephaly, short stature) and ZNF711-related XLID — mechanistically intertwined with PHF8 (Kleine-Kohlbrecher 2010 ✅; Poeta 2019 ⚠).
Other syndromic clefting + ID disorders: MSX1-related (Witkop/orofacial cleft), IRF6 (Van der Woude — lip pits), TP63 (EEC), 22q11.2 deletion (velocardiofacial — cardiac, immune, hypocalcemia), Kabuki syndrome (KMT2D/KDM6A — another chromatinopathy with clefting).
Xp11.22 contiguous deletion syndrome (broader phenotype; distinguished by CMA).
Nonsyndromic cleft lip/palate with coincidental ID — distinguished only by molecular testing.
Because clefting is present in only ~19%, PHF8 should be considered in males with unexplained DD/ID + ASD/ADHD + subtle dysmorphism even without clefting — arguably the single most actionable clinical message of the 2022 paper.

10.4 Screening

Newborn screening: ❌ not applicable (no treatable metabolic marker).
Population carrier screening: ❌ not offered; PHF8 is not on standard expanded carrier-screening panels.
Cascade testing of at-risk female relatives after a proband diagnosis: standard of care and the main practical screening activity.

11. Outcome / Prognosis

Survival / life expectancy / mortality: ❌ No mortality or survival data have been published. No deaths attributable to the disorder are reported. Adults are described in multiple families (the original Siderius pedigree and Abidi's four affected men), implying survival to adulthood is expected. Do not assert a numeric life expectancy. The defensible statement is: "no evidence of reduced life expectancy has been reported; formal survival data are absent."
Morbidity: driven by (1) lifelong intellectual disability with educational and independent-living implications; (2) neurobehavioral comorbidity (ASD ~44%, ADHD ~44%); (3) epilepsy (~31%); (4) surgical and speech burden of clefting where present (~19%); (5) infant feeding difficulty/failure to thrive (~63%).
Disability outcomes: ID severity spans borderline to severe; most published individuals fall in the mild-to-moderate range. Two individuals in Sobering had no ID (dyscalculia; mild learning difficulty) ✅ — the mild tail of the spectrum is real.
QoL measures: ❌ none applied.
Complications: velopharyngeal insufficiency/hypernasal speech, feeding and growth issues in infancy, seizure-related morbidity, behavioral comorbidity. No progressive organ complications.
Recovery potential: none for the ID itself (static encephalopathy). Cleft repair, speech therapy and educational intervention meaningfully improve function.
Prognostic factors: ❌ none validated. Notably, variant type and position do not predict severity ⚠ (Sobering), which is itself a prognostically relevant negative finding — counseling cannot be refined by genotype. Sibling pairs were more concordant than the cohort at large ⚠, hinting at genetic-background modifiers.
Prognostic biomarkers: ❌ none.

12. Treatment

There is no disease-modifying therapy. Management is entirely supportive/symptomatic and follows generic best practice for syndromic ID with clefting. Sobering et al. explicitly offer no management recommendations beyond diagnosis and counseling ⚠.

12.1 Suggested treatment annotations (NCIT — ⚠ all IDs require OAK verification against `sqlite:obo:ncit`)

Treatment	`treatment_term` (NCIT)	`therapeutic_modality`	Notes
Cleft lip and palate surgical repair (cheiloplasty/palatoplasty, staged)	NCIT:C15329 Surgical Procedure ⚠ (check for a specific palatoplasty/cheiloplasty term)	`SURGERY`	Timing per standard cleft-team protocols
Speech and language therapy	NCIT:C159273 Speech Therapy ⚠	`BEHAVIORAL`	Universal indication (100% speech delay; hypernasality)
Physical therapy	NCIT:C15302 Physical Therapy ⚠	`BEHAVIORAL`	Motor delay (75%)
Occupational therapy	NCIT:C121351 Occupational Therapy ⚠	`BEHAVIORAL`	Fine motor delay (88%)
Special education / early developmental intervention	NCIT:C15747 Supportive Care ⚠	`BEHAVIORAL`
Feeding support (specialized cleft bottles, NG feeding, nutrition)	NCIT:C15433 Nutritional Support ⚠	needs per-case judgment — do not auto-tag `BEHAVIORAL` per CLAUDE.md guidance	Infantile feeding difficulty 63%
Antiseizure medication	NCIT:C15986 Pharmacotherapy ⚠ + `therapeutic_agent` per drug	`SMALL_MOLECULE`	No PHF8-specific ASM preference known
ADHD pharmacotherapy (e.g., methylphenidate)	NCIT:C15986 ⚠	`SMALL_MOLECULE`	Generic ADHD management; no disease-specific evidence
ASD behavioral intervention	NCIT:C15747 ⚠ / behavioral counseling NCIT:C181743 ⚠	`BEHAVIORAL`
Genetic counseling	NCIT:C15240 Genetic Counseling ⚠	n/a	X-linked recurrence risk; carrier and prenatal options
Audiology / hearing management	—	`DEVICE` (if hearing aid)	Cleft-associated otitis media risk (general cleft care)

12.2 Pharmacogenomics

❌ No PHF8-specific pharmacogenomic interactions are known. Standard CPIC guidance applies to any ASM or psychotropic used.

12.3 Advanced therapeutics

Gene therapy / gene editing / ASO / siRNA / cell therapy: ❌ none in development. No PHF8 program appears in the retrieved literature or trial registries. Note that ASO strategies are ill-suited here — the lesion is loss of a large multidomain nuclear enzyme, not a splice-correctable or knockdown-amenable target (a few canonical splice-site variants are conceivable splice-modulation targets, but this is speculative).
Targeted / repurposed pharmacotherapy — the one real lead: rapamycin/sirolimus (CHEBI:9168 ✅), based on Chen et al. 2018 ✅ ("Pharmacological suppression of mTOR signaling with rapamycin in Phf8 knockout mice recovers the weakened LTP and cognitive deficits... provides a potential therapeutic drug target to treat XLID"). Strictly preclinical, mouse-only, and complicated by the Walsh 2017 Phf8-KO line showing no cognitive impairment (§15.4). Curate as evidence_source: MODEL_ORGANISM with an explicit caveat; do not present as a treatment option.
Immunotherapy: not applicable.

12.4 Clinical trials

❌ No clinical trials for Siderius syndrome / PHF8-XLID were identified. No NCT identifiers to record.

12.5 Treatment outcomes, adverse events, algorithms

❌ No disease-specific response rates, adverse-event data, treatment algorithms, combination regimens, or genotype-guided strategies exist. Cleft repair outcomes follow general cleft-care literature and should not be attributed to this disorder.

13. Prevention

Primary prevention: not possible for a germline monogenic disorder. Genetic counseling is the operative intervention: a carrier mother has a 50% chance of transmitting the variant to each child — affected sons, carrier daughters. De novo occurrence in ~3/11 families ⚠ means recurrence risk is low but not zero after an apparently de novo event (germline mosaicism cannot be excluded).
Reproductive options: prenatal diagnosis (CVS/amniocentesis with targeted variant testing), preimplantation genetic testing for monogenic disorders (PGT-M), donor gametes. Prenatal ultrasound may detect clefting; Xp11.22 deletions are detectable by prenatal CMA (PMID:32219840) ⚠.
Secondary prevention (early detection): cascade carrier testing of maternal female relatives; early developmental surveillance in known at-risk males, enabling early speech/behavioral intervention before formal diagnosis.
Tertiary prevention (complication avoidance): timely cleft repair to prevent speech and feeding morbidity; audiologic surveillance (middle-ear disease in cleft palate); seizure control; proactive ASD/ADHD assessment (the 2022 paper's practical message is that these are under-recognized in this disorder).
Immunization / public-health / environmental interventions: not applicable. (General periconceptional folic acid is population cleft-prevention advice; it has no PHF8-specific evidence base.)
Prophylaxis: none.

14. Other Species / Natural Disease

14.1 Orthologs and taxonomy

Species	NCBI Taxon	Gene	Notes
Homo sapiens	NCBITaxon:9606	PHF8 (NCBI Gene 23133)	Xp11.22
Mus musculus	NCBITaxon:10090	Phf8, MGI:2444341, X chromosome ⚠	"strong expression of the mouse orthologue Phf8 in embryonic and adult brain structures" ✅ (Laumonnier 2005)
Danio rerio	NCBITaxon:7955	phf8	Morphant brain/jaw phenotype (Qi 2010) ✅
Caenorhabditis elegans	NCBITaxon:6239	F29B9.2 / jmjd-1.2	Catalytically conserved H3K9me2/me1 demethylase; neuronal expression; locomotion defect ✅
Rattus norvegicus	NCBITaxon:10116	Phf8	Primary oligodendroglial cultures (Kremp 2024) ✅
Teleost (mangrove rivulus)	—	Phf8 among 25 Kdm orthologues	Comparative Kdm family survey (PMID:30458291) ⚠

Conservation quote (Kleine-Kohlbrecher 2010, cached full text) ✅:

"Two closely related homologs are present in C. elegans showing conservation of the overall domain structure and of the primary amino acid sequence in the PHD and JmjC domains, indicating an evolutionary conserved role for these proteins."

14.2 Naturally occurring disease in other species

❌ No naturally occurring PHF8-related disease is recorded in OMIA or the veterinary literature (no companion-animal or livestock PHF8 disorder was found). No breed associations; no VBO identifiers apply. Veterinary relevance: none.

14.3 Comparative biology

The enzymatic function is deeply conserved (worm→human), and so is the neuronal requirement: the C. elegans homolog is "highly expressed in neurons, and mutant animals show impaired locomotion" ✅, zebrafish morphants show brain-cell apoptosis and jaw defects ✅, and mice show hippocampal LTP/memory deficits ✅ (in one line). The craniofacial arm is conserved from fish to human (jaw/msxb in zebrafish; cleft lip/palate in humans; "neural crest defects" in mice ✅ per Kremp 2024). The notable non-conservation is the mouse craniofacial phenotype: no mouse model reproduces cleft lip/palate (§15.4).

14.4 Transmission

Not applicable — non-infectious, non-zoonotic.

15. Model Organisms

15.1 Mouse (Mus musculus) — MGI:2444341

Two independently generated Phf8 knockout lines, with discordant cognitive phenotypes — this is the central caveat of the mouse literature:

Line	Key findings	Citation
Chen et al. 2018 (Phf8 KO)	"impaired learning and memory, and impaired hippocampal long-term potentiation (LTP) without gross morphological defects"; hippocampal mTOR hyperactivation via RSK1 de-repression; rapamycin rescues LTP and cognition ✅	PMID:29317619
Walsh et al. 2017 (Phf8 KO)	"Phf8 deficient mice neither display obvious developmental defects nor signs of cognitive impairment. However, we report a striking resiliency to stress-induced anxiety- and depression-like behaviour"; serotonin misregulation in PFC; Htr1a and Htr2a are direct PHF8 targets ⚠	PMID:28485378

MGI's own summary of the two alleles ⚠: one null allele → impaired learning/memory and impaired hippocampal LTP; another null allele → resiliency to depression-like behavior and decreased anxiety.

Mouse embryonic work (2026) adds a developmental phenotype: PHF8 deficiency halted neurogenesis and brain development in mouse embryos ⚠ (PMID:41714361) — potentially reconciling the adult-behavior discordance by shifting attention to embryonic stages and to background/allele differences.

Available model types: targeted knockouts (≥2 independent lines) ✅; conditional/humanized/knock-in models — ❌ none reported. IMPC data for Phf8 ❌ not retrievable this session (fetch attempt returned the wrong gene page — do not rely on it).

15.2 Zebrafish (Danio rerio)

The best craniofacial model. Qi et al. 2010 ✅: phf8 knockdown → apoptosis/reduced cell survival in the brain and defective jaw development, mediated at least in part by direct regulation of msxb (MSX1); catalytically dead PHF8 fails to rescue, tying the phenotype to demethylase activity. A separate zebrafish study extends the sensory phenotype: He et al. 2020 (PMID:33330448) ⚠ — "PHF8 knockdown significantly disrupted the development of the posterior lateral line system" and caused "severe malformation of the semicircular canal and otoliths." (Inner-ear involvement has not been described in humans — treat as model-only.)

15.3 Invertebrate and cellular models

C. elegans — F29B9.2/jmjd-1.2: neuronal expression, impaired locomotion, global H3K9me2/H3K27me2 increase in mutants ✅. Also the vehicle for the GenX toxicology study ⚠.
P19 mouse embryonal carcinoma cells — RA-induced neuronal differentiation assay; PHF8 knockdown impairs it, WT (but not F279S) rescues/drives differentiation ✅ (Qiu 2010). A 2024 follow-up: PMID:38463639 ⚠.
Primary rodent oligodendroglial cultures + oligodendroglial cell lines ✅ (Kremp 2024).
Mouse astrocyte cultures ✅ (Iacobucci 2021).
Human iPSC-derived oligodendrocytes — notably, "generation of human oligodendrocytes from induced pluripotent stem cells did not require PHF8 in a system that relies on forced expression of Olig2" ✅ (Kremp 2024) — a clean epistasis result placing Olig2 downstream.
HeLa / U2OS / HEK — biochemistry, ChIP-seq, localization ✅.
Patient-derived iPSC or organoid models: ❌ none published. Given the 2026 neurogenesis/serine finding, a patient-iPSC cortical organoid is the obvious next model and a strong proposed_experiments entry.

15.4 Phenotype recapitulation and limitations — curate as `HUMAN_MODEL_MISMATCH`

This disorder is a textbook case for dismech's HUMAN_MODEL_MISMATCH discussion kind (evidence exists in models but translational validity is the open question):

No mouse model reproduces the human clefting phenotype. Both published Phf8 KO lines are reported without craniofacial malformation ("neither display obvious developmental defects"; "without gross morphological defects"). The cleft phenotype is recapitulated only in zebrafish jaw development. Mechanistic implication: species differences in redundancy (PHF2/KDM7A may compensate in mouse) or in the MSX1-dependence of lip/palate fusion.
The two mouse KO lines disagree about cognition — one shows impaired learning/memory + LTP deficits, the other shows no cognitive impairment. Any KB claim that "Phf8 KO mice model the ID phenotype" must be qualified with both citations.
The mouse behavioral phenotype that is robust (stress resilience / altered serotonin signaling) has no reported human counterpart — no depression/anxiety-resilience phenotype has been described in PHF8-XLID patients. This is a genuinely testable, unaddressed clinical question.
The rapamycin rescue is single-line, mouse-only and rests on the line whose phenotype the other lab did not replicate.

Suggested discussion entries:

discussions:
- kind: HUMAN_MODEL_MISMATCH
  attaches_to: "pathophysiology#Reduced MSX1 Expression in Cranial Neural Crest"
  prompt: >-
    Why do Phf8-null mice fail to develop cleft lip/palate when human PHF8
    loss-of-function causes orofacial clefting and zebrafish phf8 knockdown
    causes jaw malformation?
  rationale: >-
    Both published Phf8 knockout mouse lines are reported without craniofacial
    malformation, while the craniofacial arm of the mechanism rests on zebrafish
    msxb data. Possible paralog compensation (PHF2/KDM7A) or species differences
    in the MSX1 dependence of lip/palate fusion.
- kind: HUMAN_MODEL_MISMATCH
  attaches_to: "pathophysiology#mTOR Hyperactivation via RSK1 De-repression"
  prompt: >-
    Is the mTOR-dependent cognitive phenotype (and its rapamycin rescue) a
    reliable model of human PHF8-XLID, given that a second independent Phf8
    knockout line showed no cognitive impairment?
- kind: KNOWLEDGE_GAP
  attaches_to: "pathophysiology#Loss of PHF8 Histone Demethylase Function"
  prompt: >-
    Does PHF8-XLID have a detectable DNA-methylation episignature that could
    resolve the reported missense variants of uncertain significance?

15.5 Model resources

MGI (informatics.jax.org, Phf8 = MGI:2444341), IMPC/KOMP (status ❌ unverified), ZFIN, WormBase, Alliance of Genome Resources, Cellosaurus (for HeLa/U2OS/P19).

Appendix A — Consolidated citation list

PMID	Short citation	Evidence type	Cached verbatim in repo?
10398231	Siderius et al. 1999, Am J Med Genet — original linkage	HUMAN_CLINICAL	✅
16199551	Laumonnier et al. 2005, J Med Genet — PHF8 identified	HUMAN_CLINICAL	✅
17594395	Abidi et al. 2007, Clin Genet — p.K177X	HUMAN_CLINICAL	✅
17661819	Koivisto et al. 2007, Clin Genet — F279S, Finnish family	HUMAN_CLINICAL	✅
18498374	Qiao et al. 2008, Clin Genet — Xp11.22 microdeletion, ASD	HUMAN_CLINICAL	❌ (fetch needed)
19843542	Loenarz et al. 2010, Hum Mol Genet — 2-OG demethylase; F279S dead	IN_VITRO	✅ (full text)
20208542	Feng et al. 2010, Nat Struct Mol Biol — rDNA activation	IN_VITRO	❌
20346720	Kleine-Kohlbrecher et al. 2010, Mol Cell — H3K9me2/1; ZNF711 link	IN_VITRO	✅ (full text)
20421419	Fortschegger et al. 2010, Mol Cell Biol — RNAPII coactivator	IN_VITRO	✅
20548336	Qiu et al. 2010, Cell Res — RAR coactivator, neuronal differentiation	IN_VITRO	✅
20622853	Qi et al. 2010, Nature — H4K20me1; zebrafish brain/jaw; MSX1	MODEL_ORGANISM	✅ (full text)
20622854	Liu et al. 2010, Nature — H4K20me1, G1-S	IN_VITRO	❌
22850744	Asensio-Juan et al. 2012, NAR — cytoskeleton, neurite outgrowth	IN_VITRO	✅
24852203	Wang et al. 2014, Sci Rep — PHF8/REST co-occupancy	IN_VITRO	✅ (full text)
25258334	De Wolf et al. 2014, AJMG A — Xp11.22 deletion, syndromic autism	HUMAN_CLINICAL	❌
25923143	Han et al. 2015, Tissue Eng A — craniofacial bone repair	MODEL_ORGANISM	❌
28485378	Walsh et al. 2017, Nat Commun — Phf8 KO, stress resilience, 5-HT	MODEL_ORGANISM	❌
29317619	Chen et al. 2018, Nat Commun — mTOR/RSK1, rapamycin rescue	MODEL_ORGANISM	✅ (full text)
31691806	Poeta et al. 2019, Hum Mol Genet — KDM5C/ARX/ZNF711/PHF8 hub	IN_VITRO	✅
32219840	Huang et al. 2020 — prenatal CMA, Xp11.22 deletion, CL/P	HUMAN_CLINICAL	❌
33330448	He et al. 2020, Front Cell Dev Biol — zebrafish inner ear/PLL	MODEL_ORGANISM	❌
34081130	Iacobucci et al. 2021, Development — astrocyte differentiation	IN_VITRO	✅
35469323	Sobering et al. 2022, HGG Adv — 16 new individuals; phenotype expansion	HUMAN_CLINICAL	✅
38613395	Kremp et al. 2024, Glia — Olig2 downstream of Phf8	IN_VITRO	✅
39087553	Kim et al. 2024, NAR — transcription recovery after DSB	IN_VITRO	❌
39311138	Fan et al. 2024, Epigenomes — PHF8/KDM7B review	review	❌
40145966	Kremp & Wegner 2026, Neural Regen Res — oligodendroglial review	review	❌
41709745	Tan et al. 2026, Haematologica — Znf711-Phf8 in neutrophils	MODEL_ORGANISM	❌
41714361	Artés/Iacobucci et al. 2026, EMBO Rep — serine biosynthesis, neurogenesis	IN_VITRO + MODEL_ORGANISM	❌

Also relevant, not fetched: an erratum to Sobering et al. exists (HGG Adv. 2022 Dec 20;4(1):100168) ✅ — check it before transcribing any variant nomenclature from Table 1.

Appendix B — Pre-commit checklist for the curator

just fetch-reference PMID:X for every PMID marked ❌ in Appendix A that you intend to cite, then re-verify each snippet as an exact substring.
Re-verify all ⚠ quotes (Sobering frequencies, Walsh abstract, Qiao, De Wolf, Huang, He, Kim, Fan, Tan, Artés) against freshly cached abstracts — the frequency table in §3.1 came from a summarizing fetch of the PMC full text, not from a verified cache.
Read the Sobering erratum (HGG Adv 4(1):100168) before entering variant nomenclature.
Run just validate-terms on the file — every ontology ID in this report marked ⚠ is unverified (OAK and OLS were both unavailable in this session; only GO:0032454, GO:0035575, GO:0060021, HP:0410030, HP:0007018, CL:2000029, CL:0002453, CHEBI:9168 were label-verified).
Fill the gnomAD constraint gap (pLI / LOEUF / o-e LoF) manually from gnomAD v4.
Pull the ClinGen assertion properly: just clingen-refresh + cite the CGGV: id rather than the web page, for the "Definitive" classification.
Consider adding an ORPHA:85287 structured-source evidence row via just structured-rebuild-orphanet --id 85287 for prevalence/definition (orpha.net was unreachable directly).
Bind inheritance_term to HP:0001419 with a full term: block (the common gap).
Do not add a conforms_to module reference — no existing module fits (see §6.10).
Record the clefting frequency discordance (§3.4) in notes:, and add the three suggested discussions entries (§15.4).
Add a history record: just new-history --kind disorder --slug Siderius_Type_X-Linked_Intellectual_Disability --event CREATE ....

Sources (web-retrieved, non-PubMed): OMIM 300263 · GTR: Syndromic X-linked intellectual disability Siderius type · Orphanet ORPHA:85287 · GenCC PHF8 submissions · ClinGen conditions: MONDO:0010286 · ClinVar Miner: PHF8 / Siderius · MedlinePlus Genetics: PHF8 · GARD: Siderius type · MGI Phf8 (MGI:2444341) · UniProt Q9UPP1 · NCBI Gene 23133 · HPO annotations, OMIM:300263 · EBI OLS4 · PubMed/PMC via NCBI E-utilities (PMIDs cited inline; DOIs listed in the cached reference files).

Ask OpenScientist

OpenScientist Report

Classifications

Inheritance

Mechanistic Hypotheses

Discussions and Knowledge Gaps

Pathophysiology

Pathograph

Phenotypes

Genetic Associations

Medical Actions

Differential Diagnoses

Source YAML

References & Deep Research

References

Deep Research

1. Disease Information

1.1 Overview

1.2 Identifiers

1.3 Synonyms

1.4 Data provenance character

2. Etiology

2.1 Primary causal factor

2.2 Genetic risk factors

2.3 Environmental risk factors

2.4 Protective factors

3. Phenotypes

3.1 Cohort-derived frequency table (Sobering et al. 2022, n = 16 affected males with LoF variants) ⚠

3.2 HPO disease annotations currently attached to OMIM:300263

3.3 Neuroimaging phenotypes (6 individuals imaged, Sobering 2022) ⚠

3.4 The clefting discordance — flag this explicitly in the KB entry

3.5 Phenotype characteristics

3.6 Quality-of-life impact

4. Genetic / Molecular Information

4.1 Gene

4.2 Reported pathogenic variants

4.3 Variant classification, spectrum, and origin

4.4 Population / database data

4.5 Epigenetic information

4.6 Chromosomal abnormalities

5. Environmental Information

6. Mechanism / Pathophysiology

6.1 The core enzymatic lesion (MOLECULAR scale)

6.2 Downstream arm 1 — neuronal/cognitive: the mTOR–RSK1 axis

6.3 Downstream arm 2 — neuronal differentiation and cytoskeleton

6.4 Downstream arm 3 — glial biology (newer, and under-appreciated)

6.5 Downstream arm 4 — craniofacial / midline development

6.6 Downstream arm 5 — metabolic control of neurogenesis (2026, newest)

6.7 Downstream arm 6 — cell cycle, DNA damage, transcription recovery

6.8 Immune / metabolic / tissue-damage dimensions

6.9 ⚠ Scope guardrail: the PHF8 cancer literature

6.10 Suggested pathophysiology node chain for the dismech entry

7. Anatomical Structures Affected

7.1 Organ level

7.2 Tissue and cell level

7.3 Subcellular level

8. Temporal Development

9. Inheritance and Population

9.1 Epidemiology

9.2 Genetic epidemiology

9.3 Population demographics

10. Diagnostics

10.1 Genetic testing (the diagnostic backbone)

10.2 Clinical, imaging and functional testing (supportive, not diagnostic)

10.3 Clinical criteria and differential diagnosis

10.4 Screening

11. Outcome / Prognosis

12. Treatment

12.1 Suggested treatment annotations (NCIT — ⚠ all IDs require OAK verification against sqlite:obo:ncit)

12.2 Pharmacogenomics

12.3 Advanced therapeutics

12.4 Clinical trials

12.5 Treatment outcomes, adverse events, algorithms

13. Prevention

14. Other Species / Natural Disease

14.1 Orthologs and taxonomy

14.2 Naturally occurring disease in other species

14.3 Comparative biology

14.4 Transmission

15. Model Organisms

12.1 Suggested treatment annotations (NCIT — ⚠ all IDs require OAK verification against `sqlite:obo:ncit`)

15.4 Phenotype recapitulation and limitations — curate as `HUMAN_MODEL_MISMATCH`