Isolated Sedoheptulokinase Deficiency (SHPK Deficiency) — Comprehensive Research Report

Prepared: 2026-08-08 · Target: Isolated sedoheptulokinase deficiency · MONDO:0014969 · Category: Mendelian (autosomal recessive inborn error of metabolism)

⚠️ Framing note for knowledge-base curation (read first)

This is an unusual entry: the existence of a clinical disease is explicitly disputed in the defining primary literature. The only paper reporting isolated (non-contiguous-gene) SHPK deficiency is titled "First two unrelated cases of isolated sedoheptulokinase deficiency: A benign disorder?" and states verbatim:

"It is questionable whether SHPK deficiency is a causal factor for the clinical phenotypes of our patients. This study illustrates the necessity of extensive functional and clinical workup for interpreting a novel variant, including nonsense variants." — Wamelink et al., J Inherit Metab Dis 2015 (PMID:25647543)

Three independent lines of evidence support the "benign biochemical phenotype" reading: (i) the two reported patients had discordant, non-overlapping clinical presentations; (ii) SHPK loss-of-function alleles are present in population databases at frequencies higher than expected for a pathogenic recessive allele; (iii) Shpk^-/- mice reproduce the biochemical abnormality with no clinical or histological phenotype. Every clinical phenotype below should be curated with that caveat attached, and any supports: field should reflect the uncertainty rather than asserting causation.

Also flagged for curators: no ontology term ID in this report has been passed through just validate-terms. IDs marked ✅ were retrieved directly from the authoritative source (OLS4/ChEBI, HPO API, UniProt, NCBI); IDs marked ⚠️ are suggestions requiring OAK verification before commit.

1. Disease Information

1.1 Overview

Isolated sedoheptulokinase deficiency is an autosomal recessive inborn error of the non-oxidative branch of the pentose phosphate pathway (PPP) caused by biallelic loss-of-function variants in SHPK (formerly CARKL). The enzyme sedoheptulokinase (EC 2.7.1.14) phosphorylates free sedoheptulose to sedoheptulose-7-phosphate; its loss produces a characteristic urinary metabolite signature — elevated sedoheptulose and elevated erythritol, with low-to-normal sedoheptulose-7-phosphate.

Orphanet definition (verbatim, ORPHA:440713):

"A rare, hereditary disorder of pentose phosphate metabolism characterized by increased urine levels of sedoheptulose and erythritol, and low-to-normal excretion of sedoheptulose-7P."

The term "isolated" is load-bearing. Sedoheptulokinase deficiency occurs far more often as part of a contiguous gene deletion: the common 57-kb founder deletion causing nephropathic cystinosis removes CTNS and the adjacent SHPK, plus the 5′ non-coding exons of TRPV1 (PMID:18186520, PMID:21546516). "Isolated" SHPK deficiency = SHPK loss without concurrent CTNS loss, and has been reported in only two patients worldwide.

1.2 Key identifiers

Source for the cross-reference set: Orphadata rd-cross-referencing API for ORPHA:440713 (CC-BY-4.0).

1.3 Synonyms

• Isolated SHPK deficiency
• Sedoheptulokinase deficiency (SHPKD)
• Deficiency of sedoheptulokinase / deficiency of heptulokinase
• CARKL deficiency (historic gene name: carbohydrate kinase-like)
• SHK deficiency

1.4 Nature of the evidence base

Individual case reports only — there is no registry, no cohort, and no EHR-derived dataset. A targeted PubMed query (SHPK[TIAB] OR sedoheptulokinase[TIAB] AND deficiency/patient[TIAB]) returns exactly 5 records, of which only one (PMID:25647543) reports isolated human disease. Aggregated resources (Orphanet, OMIM, GARD, MedGen) all trace back to that single 2015 paper. The much larger body of human data comes from cystinosis patients homozygous for the 57-kb deletion, in whom SHPK deficiency is a secondary, co-deleted trait.

2. Etiology

2.1 Causal factors

Genetic, monogenic, autosomal recessive. Two mechanisms produce SHPK deficiency:

(a) Isolated SHPK deficiency — biallelic intragenic loss-of-function variants in SHPK (17p13.2). Both reported patients were homozygous for a nonsense variant, and both came from families with consanguinity:

"Both patients had elevated excretion of erythritol and sedoheptulose, and each had a homozygous nonsense mutation in SHPK." (PMID:25647543)

(b) Contiguous-gene ("non-isolated") SHPK deficiency — homozygous 57-kb deletion at 17p13.2 removing CTNS + SHPK (± TRPV1 5′ exons). This is cystinosis with secondary SHPK deficiency, not this disease entity, and should be curated on the cystinosis entry with a cross-reference:

"Cystinosis patients with the common 57-kb deletion had strongly elevated urinary concentrations of sedoheptulose (28-451 mmol/mol creatinine; controls and other cystinosis patients <9) and erythritol (234-1110 mmol/mol creatinine; controls and other cystinosis patients <148)." (PMID:18186520)

2.2 Genetic risk factors

• Causal variants: NM_013276.4:c.355C>T (p.Arg119Ter) and NM_013276.4:c.211G>T (p.Glu71Ter) — see §4.
• Consanguinity is the dominant risk factor in both reported families (see §9).
• No susceptibility loci, GWAS signals, or modifier genes are known. No GWAS Catalog association exists for SHPK with this phenotype.
• Note the inverse relationship: SHPK has itself been proposed as a modifier of the cystinosis phenotype (NCBI Gene summary for GeneID 23729: "The gene resides in a chromosomal region frequently deleted in cystinosis patients, potentially serving as a disease modifier"). That hypothesis has now been tested and not supported for the HSPC/macrophage axis (§15).

2.3 Environmental risk factors

None identified. No toxin, occupational, infectious, or lifestyle exposure has been associated. Age, sex, and family history other than consanguinity carry no reported effect.

One mechanistically plausible (but unstudied in humans) modifier: dietary sedoheptulose intake, since the accumulating substrate is partly diet-derived:

"Testing plant extracts revealed sedoheptulose presence in carrots and fruits." (PMID:18775706)

2.4 Protective factors

None documented. No protective allele, dietary factor, or lifestyle exposure is described. gnomAD-frequency reasoning (§9.3) suggests the deficiency itself is largely non-deleterious, which makes "protective factor" an ill-posed question here.

2.5 Gene–environment interactions

No demonstrated GxE. A mechanistically predicted interaction, worth curating as a hypothesis rather than a fact: erythritol production in SHPK deficiency depends on a bypass route through ketohexokinase/fructokinase (KHK) and aldolase B (ALDOB) (§6.2). Dietary sedoheptulose load, and in principle KHK/ALDOB genotype (e.g., hereditary fructose intolerance alleles), would therefore modulate the erythritol biomarker. This is an inference from PMID:18775706, not an observed interaction — do not curate it as evidence-backed.

3. Phenotypes

3.1 The two index patients (the entire clinical evidence base)

Verbatim (PMID:25647543):

"The first patient presented with neonatal cholestasis, hypoglycemia, and anemia, while the second patient presented with congenital arthrogryposis multiplex, multiple contractures, and dysmorphisms."

Patient-level demographics are as abstracted by OMIM into ClinVar; for VCV000372203 (p.Glu71Ter) the ClinVar record states the variant was found in "a 2-year-old girl, born of consanguineous Turkish parents." Patient 1's age/ancestry come from the same OMIM abstraction chain and should be re-verified against the paywalled full text before being asserted in a KB entry.

The two presentations share no clinical feature. That absence of a recurrent phenotype is the single most important observation about this disease and is why the authors questioned causality.

3.2 HPO annotation set

Two annotation sources exist and they disagree in scope. Curators must know which they are using.

(a) HPOA / OMIM:617213 — the conservative set (2 terms, frequency 2/2):

This is the defensible core: the only feature present in both patients is the biochemical one.

(b) HPOA / ORPHA:440713 — the expansive set (28 terms):

Curation warning — the "Frequent"/"Obligate" labels here are artifacts. Each term derives from one of the two patients (the arthrogryposis cluster from Patient 2; the cholestasis/anemia cluster from Patient 1). "Obligate" applied to arthrogryposis is literally contradicted by Patient 1, who did not have it. Per the dismech frequency SOP (docs/frequency-evidence-guidelines.md), omit frequency: on these phenotypes rather than importing the Orphanet band. If frequency must be recorded, the honest statement is 1/2 for every clinical feature and 2/2 for the biochemical ones.

Two additional laboratory HPO terms belong on the entry but are absent from both annotation sets:

Note the internal inconsistency: the Orphanet set carries postprandial hyperglycemia (HP:0011998) while the source abstract reports hypoglycemia in Patient 1. Resolve against full text before curating either.

3.3 Phenotype characteristics

• Age of onset: congenital / neonatal in both patients (Orphanet onset category: antenatal and neonatal). The biochemical phenotype is present from birth and lifelong.
• Severity: not gradeable; n=2 with discordant features.
• Progression: unknown. No follow-up beyond ages 2–3 years has been published in the 11 years since.
• Frequency among affected individuals: see the §3.2 warning — effectively unmeasurable.

3.4 Quality of life

No data. No EQ-5D, SF-36, PROMIS, or disease-specific instrument has been applied. Given the likely benign nature of the isolated biochemical defect, the QoL burden in the two index patients is more plausibly attributable to their (unexplained) clinical syndromes than to SHPK loss.

4. Genetic / Molecular Information

4.1 Causal gene

Genomic context is central to this gene's story: SHPK sits immediately adjacent to CTNS, sharing a bidirectional promoter region, inside a segment dense in Alu repeats that mediate the recurrent 57-kb deletion.

"The CTNS promoter region shares 41 nucleotides with the promoter region [of the adjacent CARKL gene], though patient mutations did not affect CARKL activity." (PMID:11505338)

"sequence analysis detected the presence of a novel gene (CARKL) residing within the most common cystinosis-causing deletion." (PMID:10673275, Genome Res 2000;10:165-73 — the gene's discovery paper)

4.2 Pathogenic variants — and their contested classification

Variant 1 — p.Arg119Ter (Patient 1)

The submitter's interpretation is the most consequential text in this whole report for curation purposes:

"The current clinical and genetic evidence is not sufficient to establish whether loss-of-function variants in SHPK cause disease… This variant is present in population databases (rs144071313, gnomAD 0.09%), and has an allele count higher than expected for a pathogenic variant… The available evidence is currently insufficient to determine the role of this variant in disease." (Labcorp Genetics, SCV002219788.4)

Population distribution (dbSNP): highest in European and Latin American populations; minimal or absent in East Asian, South Asian, and African populations.

Variant 2 — p.Glu71Ter (Patient 2)

Neither causal variant is classified "Pathogenic" or "Likely pathogenic" in ClinVar. One is a VUS; the other carries only OMIM's "Affects" label — the ClinVar term reserved for variants that alter a measurable trait without established disease causation. Under ACMG/AMP rules, PM2 (absent from controls) actively fails for R119*, and PS3 (functional evidence) is satisfied only for the enzymatic phenotype, not a clinical one. Curating either as PATHOGENIC would misrepresent the primary sources.

4.3 Copy-number variants

ClinVar contains 363 variant records touching SHPK, of which 67 carry a pathogenic classification. Inspection of the top-ranked pathogenic records shows every one is a multi-gene copy-number loss involving CTNS (± TRPV1, ± dozens of 17p13 genes) — i.e., cystinosis deletions and larger 17p13.3–13.2 deletions. Representative records: VCV004851017, VCV004279276, VCV004075905 (all chr17:~3.50–3.56 Mb ×1, genes CTNS, SHPK, TRPV1), and VCV003243030 (NC_000017.10:g.(?_3520391)_(3558524_?)del, conditions: ocular cystinosis / juvenile nephropathic cystinosis).

There is no ClinVar pathogenic CNV restricted to SHPK alone. Isolated SHPK deficiency has never been reported from a CNV.

4.4 Functional consequence

Both variants are loss of function via premature termination, truncating the 478-aa protein at residue 71 or 119 — before the FGGY kinase catalytic core is complete. Both are predicted NMD substrates. Labcorp's assessment: the variant "creates a premature translational stop signal… expected to result in an absent or disrupted protein product," and functional studies in PMID:25647543 confirmed impact on SHPK activity. No gain-of-function or dominant-negative mechanism is described.

For dismech schema purposes: GeneticContext.functional_impact_category: LOSS_OF_FUNCTION, allele_type nonsense, zygosity homozygous, variant_origin germline.

4.5 Modifier genes, epigenetics, chromosomal abnormalities

• Modifier genes: none identified for SHPK deficiency itself.
• Epigenetics: no methylation or chromatin study of SHPK in disease. The shared bidirectional CTNS/CARKL promoter (PMID:11505338) is a cis-regulatory, not epigenetic, finding — and importantly, cystinosis-causing CTNS promoter mutations were shown not to affect CARKL promoter activity.
• Chromosomal abnormalities: the Alu-mediated recurrent 57-kb deletion (see §9.4); larger 17p13.3–13.2 deletions encompassing PAFAH1B1/YWHAE (Miller–Dieker region) also remove SHPK incidentally.
• Transcriptional consequence of the deletion beyond SHPK: the deletion extends into TRPV1, producing measurable downstream dysfunction — relevant when attributing phenotypes in deletion patients:

  "72% reduction in PBMC TRPV1 mRNA levels in cystinosis individuals homozygous for the 57 kb deletion (n=6) compared to unaffected individuals without the deletion (n=6) (p=0.002)." (PMID:21546516) "cystinosis patients homozygous for the 57-kb deletion exhibit a strong reduction of TRPV1 function" — 60% reduction in capsaicin-evoked vasodilation and pain, increased heat detection threshold (PMID:27734949, Sci Rep 2016;6:35395)

5. Environmental Information

• Environmental factors: none. No CTD/TOXNET/EPA association. Not a toxicant-, radiation-, or pollution-related condition.
• Lifestyle factors: none established. The only dietary consideration is the exogenous supply of the accumulating substrate — sedoheptulose is present in carrots and fruits (PMID:18775706). No dietary intervention has been trialled and none is indicated.
• Infectious agents: not applicable.

(Adjacent but distinct: sedoheptulose has been proposed as an anti-inflammatory agent based on the CARKL immunometabolism work — US patent 9,694,026, "Use of sedoheptulose for prevention or treatment of inflammation." This is a therapeutic-development thread about the substrate, not an environmental risk factor for the disease.)

6. Mechanism / Pathophysiology

6.1 The primary enzymatic lesion (upstream)

Sedoheptulokinase catalyses a single, well-defined reaction:

sedoheptulose + ATP → D-sedoheptulose 7-phosphate + ADP + H⁺

The product, sedoheptulose-7-phosphate (CHEBI:15721 ✅), is a core intermediate of the non-oxidative branch of the pentose phosphate pathway (GO:0009052 ✅), feeding transketolase and transaldolase reactions. SHPK thus provides an entry point for free sedoheptulose — dietary or from intracellular hydrolysis — into central carbohydrate metabolism.

"The identification of sedoheptulose kinase demonstrates that free sedoheptulose serves as 'a relevant and accessible carbon source in humans.'" (PMID:23514175, Biochem Soc Trans 2013;41:674-80)

6.2 The causal chain to the biomarker phenotype (the well-established part)

This is the one mechanistic chain in this disease that is fully worked out, and it is the chain a dismech pathophysiology graph should encode:

SHPK biallelic LOF (MOLECULAR)
  → loss of sedoheptulokinase activity (GO:0050277 ↓)  [MOLECULAR]
    → ① reduced hepatic sedoheptulose-7-phosphate  [MOLECULAR/CELLULAR]
    → ② accumulation of free sedoheptulose (CHEBI:16802)  [ORGANISM]
 → renal excretion → increased urinary sedoheptulose (HP:0025157)
 → ③ bypass: fructokinase (KHK) phosphorylates sedoheptulose
      → sedoheptulose 1-phosphate (CHEBI:9082)
      → aldolase B (ALDOB) cleavage
   → dihydroxyacetone phosphate + erythrose
        → reduction of erythrose
             → erythritol (CHEBI:17113) accumulation
                  → increased urinary erythritol (HP:0034613)

Verbatim support for the erythritol arm:

"Sedoheptulose 1-phosphate is shown to be a substrate for aldolase B… the findings suggest that in sedoheptulose-7-kinase-deficient patients, fructokinase phosphorylates sedoheptulose to sedoheptulose 1-phosphate, which aldolase B cleaves, leading to erythrose reduction to erythritol." (PMID:18775706, FEBS Lett 2008;582:3330-4)

Enzymatic confirmation in human cells:

"Enzyme studies performed on fibroblast homogenates derived from patients carrying the 57-kb deletion revealed 80% reduction in their sedoheptulose phosphorylating activity compared to cystinosis patients with other mutations and controls." (PMID:18186520)

And in the mouse knockout: "Analysis of pentose phosphate pathway intermediates in livers demonstrated a reduction in sedoheptulose-7-phosphate and an increase in sedoheptulose and erythritol in the urine" (PMID:34823997).

Where the chain stops. There is no demonstrated causal link from this biochemical chain to any clinical manifestation. Neither sedoheptulose nor erythritol has documented toxicity at the concentrations observed, and the PPP-flux consequences appear compensated in vivo. A dismech pathograph should terminate the causal chain at the laboratory phenotypes and represent the clinical features as unexplained co-occurrence, not downstream consequences.

6.3 The immunometabolic arm (CARKL as a metabolic rheostat)

The richest mechanistic literature on this protein concerns not the inborn error but CARKL's role as a regulator of immune-cell metabolism. This matters for the disease entry because it is the strongest a priori reason to expect a phenotype — and its absence in patients and mice is informative.

Founding observation (PMID:22682222, Haschemi et al., Cell Metab 2012;15:813-26):

"We find that one of these, the carbohydrate kinase-like protein CARKL, is rapidly downregulated in vitro and in vivo upon LPS stimulation in both mice and humans. Interestingly, CARKL catalyzes an orphan reaction in the pentose phosphate pathway, refocusing cellular metabolism to a high-redox state upon physiological or artificial downregulation. We find that CARKL-dependent metabolic reprogramming is required for proper M1- and M2-like macrophage polarization and uncover a rate-limiting requirement for appropriate glucose flux in macrophage polarization."

Downstream and corroborating literature:

Cancer: SHPK is implicated in glioblastoma proliferation —

"SHPK expression in GBM shows a significant correlation with histology, prognosis, and survival. In particular, its increased expression is associated with a worse prognosis. Furthermore, its overexpression in GBM cells confirms an increase in cell proliferation." (PMID:35682658, Int J Mol Sci 2022;23:5978)

⚠️ This paper carries a published Correction (PMID:39941164, Int J Mol Sci 2025;26:1044). Cite the correction alongside the original.

Note the direction of these findings: they concern overexpression or acute downregulation of CARKL as a regulatory event in immune/tumour cells, not constitutional germline deficiency. The clean mouse-knockout phenotype (§15) shows the two do not straightforwardly translate.

6.4 Suggested ontology terms for the mechanism graph

No dismech mechanism module currently fits this disease well. metabolic_intoxication_decompensation does not apply — there is no toxic-metabolite crisis, no catabolic-stress decompensation, and no encephalopathy. Do not force a conforms_to.

7. Anatomical Structures Affected

7.1 Where the enzyme is expressed (the biologically grounded answer)

• Liver, kidney, pancreas — strong expression; placenta, heart — lower (UniProt Q9UHJ6).
• NCBI Gene expression profile: ubiquitous, with notable kidney (RPKM 7.3) and duodenum (RPKM 6.9) across 27 tissues.
• Subcellular: cytoplasm/cytosol. Not mitochondrial, lysosomal, or nuclear.

7.2 Organs implicated in the reported patients (attribution uncertain)

Lateralization: bilateral/symmetric where relevant (contractures, orbits). No asymmetric involvement described.

Important negative: In Shpk^-/- mice there were "no histologic anomalies in kidney and liver" — the two organs with highest SHPK expression (PMID:34823997). This is direct evidence against liver/kidney being target organs of the enzyme defect.

7.3 Cell types

No cell-type-specific pathology has been demonstrated in patients. The mechanistic literature implicates macrophages (M1/M2 polarization), T cells, and microglia as cells where CARKL levels matter functionally — but as regulatory biology, not as sites of disease lesion.

8. Temporal Development

• Onset: congenital. Orphanet records onset as antenatal and neonatal; GARD states symptoms "may appear during pregnancy and as a newborn." Patient 2 had congenital arthrogryposis (prenatal onset, with breech presentation and neonatal asphyxia); Patient 1 presented in the neonatal period with cholestasis.
• Onset pattern: the metabolic derangement is chronic and constitutive from birth. The clinical presentations in the two cases were acute-on-congenital.
• Stages: none defined. No staging system exists.
• Progression rate / course: unknown. Both patients were reported at ages 2–3 years with no published follow-up. There is no natural-history study, no registry, and no longitudinal cohort.
• Duration: the biochemical phenotype is lifelong and non-remitting (it is an enzyme absence).
• Remission: not applicable to the biochemical phenotype. No treatment-induced remission is possible or has been attempted.
• Critical periods: none identified. If the disorder is benign, the concept does not apply; if the neonatal presentations were causally related, the neonatal period would be the window — but that causality is precisely what is unestablished.

9. Inheritance and Population

9.1 Inheritance

Autosomal recessive (HP:0000007, HPOA frequency 2/2; Orphanet; GARD). Both index patients were homozygous for a nonsense variant, both from families with consanguinity. No compound heterozygote has been reported. No X-linked, mitochondrial, or digenic contribution.

• Penetrance: cannot be estimated, and the population-genetic data (§9.3) argue for markedly reduced or zero clinical penetrance. Biochemical penetrance (urinary metabolite elevation) appears complete.
• Expressivity: the two cases are maximally discordant — which is more parsimoniously read as coincidence than as variable expressivity.
• Anticipation: not applicable (no repeat expansion).
• Germline mosaicism: not reported.

9.2 Epidemiology

• Prevalence: not documented. Orphanet assigns no epidemiological class to ORPHA:440713. Only 2 cases have ever been reported (CASES_IN_LITERATURE = 2 is the honest structured value for a dismech Prevalence record; prevalence_class: NOT_YET_DOCUMENTED).
• Incidence: unknown.
• Sex ratio: 1 male : 1 female among reported cases — uninformative at n=2.
• Age distribution: both reported patients ascertained in infancy/early childhood; no adult cases reported (which does not mean none exist — untargeted urine metabolite screening in adults is rare).

9.3 Carrier frequency and the under-ascertainment argument

This calculation is worth recording explicitly in the KB because it is the strongest quantitative argument about the disorder's nature:

• p.Arg119Ter (rs144071313) gnomAD allele frequency ≈ 0.00057 (0.057%; Labcorp cites 0.09% for a subpopulation-inclusive figure).
• Predicted heterozygous carrier frequency for this single allele: ≈1 in 880.
• Predicted homozygote frequency for this single allele alone: p² ≈ 3.2 × 10⁻⁷ ≈ 1 in ~3.1 million — i.e. roughly 2,500 living homozygotes worldwide for R119* alone, before counting any other SHPK LoF allele.
• Observed reported cases: 2.

(This is my arithmetic from the cited gnomAD frequencies, presented as a derived estimate, not a published figure.) The gap of three-plus orders of magnitude is exactly the reasoning Labcorp applied — "an allele count higher than expected for a pathogenic variant" — and it supports either (a) the condition is clinically silent, or (b) it is severely under-ascertained because nobody measures urinary sedoheptulose. Both readings are compatible with the primary paper's own question mark.

⚠️ Not retrieved: gnomAD gene-level constraint metrics for SHPK (pLI, LOEUF, observed/expected pLoF, homozygous pLoF count). The gnomAD browser is a client-rendered application and its GraphQL API requires POST, neither of which was accessible from this session. These should be pulled manually — the LOEUF value and the presence/absence of homozygous pLoF individuals would materially strengthen or weaken the benign-disorder argument.

9.4 Founder effects, geography, and the far more common contiguous-gene form

Isolated form: p.Arg119Ter is enriched in European and Latin American populations and near-absent in East Asian, South Asian, and African populations (dbSNP/ALFA). p.Glu71Ter was found in a Turkish consanguineous family and is absent from TOPMed — consistent with a private or very rare regional allele. Neither constitutes an established founder mutation.

Contiguous-gene form (secondary SHPK deficiency): the 57-kb CTNS/SHPK deletion is a genuine northern European founder allele, thought to have originated in Germany, and: - occurs in ~60% of cystinosis patients in the US and northern Europe (PMID:15365816: "The most prevalent CTNS mutation, a 57-kb deletion, occurs in approximately 60% of patients"); - accounts for 50–70% of pathogenic CTNS alleles in those regions, and is homozygous in ~50% of northern-European cystinosis patients (GeneReviews, Cystinosis); - has not been reported in individuals from the Middle East, Asia, or Africa.

Cystinosis birth prevalence is 1:100,000 to 1:200,000 (1:26,000 in Brittany, France). Combining these figures gives an order-of-magnitude estimate that roughly 1 in 200,000–400,000 births in northern European populations carries biallelic SHPK loss as part of a cystinosis deletion — a population perhaps a thousand-fold larger than the reported isolated cases. (Derived estimate; label as such.) These individuals are, by definition, clinically dominated by cystinosis.

Consanguinity is the operative population factor for the isolated form: both reported families were consanguineous or suspected consanguineous.

10. Diagnostics

10.1 Biochemical (the diagnostic entry point)

Urinary sugar/polyol profiling by GC-MS or LC-MS/MS is the discriminating test. The characteristic pattern:

⚠️ Note these interval values were established in the cystinosis 57-kb-deletion cohort, not the two isolated-deficiency patients (whose exact values are in the paywalled full text of PMID:25647543 and should be extracted before curating a reference_ranges block). No LOINC code exists for urinary sedoheptulose.

Dried blood spot sedoheptulose (LC-MS/MS) — developed for cystinosis screening but directly applicable:

"Sedoheptulose concentrations in the deleted patients were 6 to 23 times above the upper limit for controls. No overlap existed between sedoheptulose levels in patients homozygous for the deletion versus those without it." (PMID:21195649, Mol Genet Metab 2011;102:339-42)

Untargeted metabolomics is an increasingly used unbiased route into non-oxidative PPP disorders:

"Targeted polyol testing and untargeted metabolomic testing methods were both able to identify specific biochemical patterns indicative of TKT and TALDO deficiency… untargeted analysis revealed novel biomarkers including ribonate, ribose, erythronate, and sedoheptulose 7-phosphate." (PMID:32828637, Mol Genet Metab 2020;131:147-154)

Enzyme assay: sedoheptulose-phosphorylating activity in cultured skin fibroblast homogenates — 80% reduction demonstrated in 57-kb-deletion patients vs controls (PMID:18186520). This is a research assay, not a routine clinical service.

10.2 Genetic testing

Critical interpretation caveat: because both known causal variants are ClinVar VUS/"Affects" and SHPK LoF alleles are relatively common in gnomAD, a homozygous SHPK nonsense finding on exome should not be reported as the explanation for a complex neonatal phenotype without biochemical confirmation and, crucially, without continuing the diagnostic search. This is the operational lesson of PMID:25647543.

10.3 Other modalities

• Imaging: no disease-specific findings. Brain MRI abnormalities (myelination, ventriculomegaly, subcortical atrophy) were recorded in one patient; liver imaging for cholestasis in the other. Both are workup of the presenting syndrome, not of SHPK deficiency.
• Biopsy/histopathology: no characteristic finding. Mouse KO liver and kidney are histologically normal.
• Electrophysiology, functional tests: no role.

10.4 Clinical criteria and differential diagnosis

No standardized diagnostic criteria exist. A working definition is: biallelic SHPK LoF + elevated urinary sedoheptulose and erythritol + normal cystine (excluding cystinosis).

Differential diagnosis — causes of elevated urinary sedoheptulose/polyols:

(TALDO/RPI background: Disorders of the Pentose Phosphate Pathway and Polyol Metabolism; RPI deficiency original description in Am J Hum Genet.)

10.5 Screening

• Newborn screening: not performed and not indicated for isolated SHPK deficiency (no treatment, uncertain pathogenicity — fails Wilson–Jungner criteria on both counts).
• However, DBS sedoheptulose is a proposed NBS marker for cystinosis homozygous for the 57-kb deletion, where presymptomatic detection does change management (cysteamine): "The method enables fast pre-symptomatic detection of cystinosis patients homozygous for the 57-kb deletion, facilitating early treatment initiation" (PMID:21195649). Any such programme would incidentally detect isolated SHPK deficiency — an incidental-findings issue worth flagging.
• Carrier screening: not on any expanded carrier screening panel; not recommended.
• Cascade screening: reasonable within a family for genetic-counselling completeness, but the counselling message is one of uncertainty.

11. Outcome / Prognosis

No prognostic data exist. Both index patients were alive at ages 2 and 3 at publication; no follow-up has been published in the subsequent 11 years.

• Survival / life expectancy / mortality: unknown; no deaths attributed to SHPK deficiency. There is no SEER/registry/GBD entry.
• Disease-specific mortality: none reported.
• Morbidity / disability: the disabilities present in the two index patients (arthrogryposis-related motor impairment; neonatal liver disease) are substantial but of unestablished attribution.
• Complications: none attributable to the enzyme defect.
• Recovery potential: the enzyme deficiency is permanent; the biochemical phenotype does not remit.
• Prognostic factors / biomarkers: none. Urinary sedoheptulose and erythritol are diagnostic markers with no demonstrated prognostic value.

The best available prognostic evidence is indirect and reassuring: Shpk^-/- mice show "no obvious phenotypic abnormalities, including the absence of histologic anomalies in kidney and liver" (PMID:34823997), and IMPC phenotyping of the Shpk line reports 0 significant phenotypes across 20 of 24 physiological systems tested.

12. Treatment

There is no specific treatment, and none is currently indicated. No pharmacological, dietary, enzyme-replacement, gene-therapy, or cell-therapy approach targets SHPK deficiency.

Clinical trials: zero. A ClinicalTrials.gov API query for sedoheptulokinase OR SHPK returned 0 studies.

12.1 Management as actually practised

Dietary sedoheptulose restriction (NCIT:C15447 Dietary Intervention) is theoretically capable of lowering the biomarker but has never been trialled, has no rationale in the absence of demonstrated toxicity, and should not be curated as a treatment.

12.2 Pharmacogenomics, advanced therapeutics, surgery

• Pharmacogenomics: no PharmGKB/CPIC entry for SHPK.
• Gene therapy, cell therapy, RNA-based therapy, targeted therapy, immunotherapy: none developed or proposed for this indication.
• Surgery: no role for the metabolic disease.

12.3 The one therapeutically consequential finding — cystinosis gene therapy

The most clinically actionable result in this entire literature concerns whether secondary SHPK deficiency compromises HSPC gene therapy for cystinosis. It does not:

"Transplantation of Shpk-/- HSPCs into Ctns-/- mice resulted in significant reduction in tissue cystine load and restoration of Ctns expression, as well as improved kidney architecture comparable to WT-HSPC recipients. Altogether, these data demonstrate that absence of SHPK does not alter the ability of HSPCs to rescue cystinosis, and then patients homozygous for the 57-kb deletion should benefit from ex vivo gene therapy and can be enrolled in the ongoing clinical trial. However, because of the limits inherent to animal models, outcomes of this patient population will be carefully compared to the other enrolled subjects." (PMID:34823997, Mol Genet Metab 2021;134:309-316)

The concern was specific and well-founded — CARKL "influences macrophage polarization," and the therapy's mechanism of action depends on transplanted HSPCs differentiating into tissue macrophages that transfer cystinosin-bearing lysosomes via tunneling nanotubes.

Relevant registered trials (for the cystinosis cross-reference, not for this disease): - NCT03897361 — Stem Cell Gene Therapy for Cystinosis (CTNS-RD-04; autologous CD34+ HSPC, lentiviral CTNS; Phase 1/2; Completed; UC San Diego) - NCT05146830 — Long-Term Follow-Up of CTNS-RD-04 Recipients (observational, enrolling by invitation) - NCT06910813 — DFT383 in Pediatric Participants With Nephropathic Cystinosis (Phase I/II, recruiting; Novartis)

12.4 Treatment outcomes, adverse events, algorithms

Not applicable — no disease-directed treatment exists, therefore no response rates, no FAERS signal, no algorithm, no combination or genotype-guided strategy.

13. Prevention

• Primary prevention: not applicable to a constitutional genetic disorder. The only lever is reproductive: genetic counselling for consanguineous couples with an affected child. Given that both causal variants are ClinVar VUS/"Affects" and the disorder may be benign, counselling should explicitly frame recurrence risk as risk of the biochemical trait, with clinical consequences unknown. Offering prenatal diagnosis or PGT for a possibly benign biochemical trait raises a real ethical question and should not be presented as routine.
• Secondary prevention: no screening programme; no evidence that presymptomatic detection changes any outcome. Contrast with cystinosis, where DBS sedoheptulose screening would enable early cysteamine (PMID:21195649).
• Tertiary prevention: no complications are known to be preventable, because none are established as disease-related.
• Immunization, behavioural intervention, prophylaxis, public health/environmental intervention: all not applicable.
• Risk stratification: the only actionable stratifier is consanguinity + an affected proband.

14. Other Species / Natural Disease

14.1 Taxonomy and orthologs

The enzyme belongs to the FGGY carbohydrate kinase family, which is deeply conserved across bacteria, plants, and animals — and the substrate sedoheptulose is a Calvin-cycle-adjacent plant metabolite (sedoheptulose-1,7-bisphosphatase), which is why plants are a dietary source.

14.2 Natural disease in other species

None known. An OMIA search for SHPK returns no phene records in any species — there is no naturally occurring animal model, no companion-animal or livestock disease, and no wildlife counterpart. No veterinary relevance.

14.3 Comparative biology and transmission

• Comparative pathology: the mouse knockout replicates the human biochemical phenotype exactly (urinary sedoheptulose ↑, erythritol ↑, hepatic S7P ↓) and shows no clinical or histological phenotype — the strongest available cross-species evidence that the metabolite abnormality is not intrinsically pathogenic.
• Evolutionary conservation: the CARKL-as-metabolic-rheostat function is conserved from echinoderms to mammals — sea cucumber AjCARKL overexpression suppressed G6PD, ROS production, and phagocytosis in coelomocytes and drove M2-like polarization in mouse macrophages (PMID:32283109). This makes the absence of an immune phenotype in human/mouse deficiency more striking, not less.
• Zoonotic potential / cross-species susceptibility: not applicable (non-infectious genetic disease).

15. Model Organisms

15.1 Mouse — Shpk knockout (the key model)

Origin: Goodman et al. 2021 (Cherqui lab, UC San Diego), created specifically to test whether SHPK loss compromises cystinosis HSPC gene therapy (PMID:34823997).

Phenotype recapitulation:

"Shpk-/- mice also recapitulated the urinary excretion of sedoheptulose and erythritol found in cystinosis patients homozygous for the 57-kb deletion." (PMID:34823997)

IMPC broad phenotyping: 20 of 24 physiological systems tested; 0 significant phenotypes; 4 systems not yet evaluated (IMPC Data Release 24.0).

Model limitations: (a) it does not reproduce any human clinical feature — which may reflect genuine benignity rather than model failure; (b) mouse diet differs in sedoheptulose content from human diet, potentially altering substrate load; (c) IMPC coverage is incomplete (4 systems untested), and no aged cohort, immune-challenge, or metabolic-stress paradigm has been reported. A HUMAN_MODEL_MISMATCH discussion is arguably warranted here in the inverse of the usual direction: the model is healthier than the patients, and the open question is whether the patients' illness had anything to do with the gene.

Research applications: PPP flux in vivo; macrophage/HSPC biology in a CARKL-null background; the cystinosis gene-therapy eligibility question (already answered).

15.2 Cellular and in vitro models

No iPSC, organoid, zebrafish, Drosophila, C. elegans, or yeast model of SHPK deficiency has been reported.

15.3 Model databases

MGI (MGI:1921887) · IMPC (mousephenotype.org, Shpk) · MMRRC (043666-UCD) · Alliance of Genome Resources · Cellosaurus (for the GBM lines).

Summary of what could not be retrieved

Stated explicitly so gaps are not mistaken for negatives:

1. OMIM full-text entries (#617213 and *605060) — omim.org returned HTTP 403 to automated fetch. The OMIM clinical-synopsis content is indirectly captured via HPOA (2 terms) and the ClinVar OMIM submissions; the OMIM narrative "Clinical Features"/"Molecular Genetics" prose was not read.
2. gnomAD gene-level constraint for SHPK (pLI, LOEUF, obs/exp pLoF, homozygous pLoF count) — browser is client-rendered; API requires POST. Variant-level frequencies were obtained via ClinVar and dbSNP.
3. Full text of Wamelink et al. 2015 (paywalled, not in PMC) — so the index patients' exact metabolite concentrations, fibroblast enzyme activities, imaging details, and the reasoning behind "extensive functional and clinical workup" were not read directly. This is the single highest-value document to obtain before finalising a KB entry, since it contains the per-patient quantitative data.
4. Human Protein Atlas SHPK page — the fetched URL resolved to a different gene (RIDA); tissue expression here is sourced from UniProt and NCBI Gene instead.

Suggested dismech curation posture

• Curate as a Disease entry with a deliberately thin pathophysiology graph terminating at the two laboratory phenotypes; do not draw causal edges to the clinical features.
• Record a KNOWLEDGE_GAP discussion attached to the disease: "Is isolated SHPK deficiency a disease at all, or a benign biochemical trait? Two discordant cases, LoF alleles at population frequencies exceeding recessive-disease expectation, and a phenotypically normal knockout mouse." With proposed_experiments: systematic urine-metabolite screening of gnomAD-identified SHPK LoF homozygotes; deep phenotyping of the two index families' unresolved syndromes.
• Prefer supports: PARTIAL over SUPPORT for any evidence item linking a clinical phenotype to the genotype.
• Omit frequency: on all clinical phenotypes (see §3.2).
• Cross-reference the cystinosis entry for the contiguous-gene form, and note the TRPV1 co-deletion as a competing explanation for phenotypes in deletion patients.
• No conforms_to module fits; do not force one.

Sources

• First two unrelated cases of isolated sedoheptulokinase deficiency: A benign disorder? — PubMed (PMID:25647543)
• Springer Nature Link — J Inherit Metab Dis 2015, 10.1007/s10545-014-9809-1
• Sedoheptulokinase deficiency due to a 57-kb deletion in cystinosis patients… elucidation of the CARKL gene (PMID:18186520)
• Characterization of mammalian sedoheptulokinase and mechanism of formation of erythritol (PMID:18775706)
• Elevated concentrations of sedoheptulose in bloodspots of patients with cystinosis (PMID:21195649)
• The 57 kb deletion in cystinosis patients extends into TRPV1 (PMID:21546516)
• TRPV1 dysfunction in cystinosis patients harboring the homozygous 57 kb deletion (PMID:27734949)
• The genomic region encompassing CTNS… discovery of a novel gene within the common cystinosis-causing deletion (PMID:10673275)
• The promoter of CTNS… shares sequences with the promoter of an adjacent gene, CARKL (PMID:11505338)
• FISH diagnosis of the common 57-kb deletion in CTNS causing cystinosis (PMID:15365816)
• The sedoheptulose kinase CARKL directs macrophage polarization through control of glucose metabolism (PMID:22682222)
• Sedoheptulose kinase regulates cellular carbohydrate metabolism by sedoheptulose 7-phosphate supply (PMID:23514175)
• Metabolic reprograming in macrophage polarization (PMID:25228902)
• Time and Demand are Two Critical Dimensions of Immunometabolism (PMID:25904920)
• Sedoheptulose kinase bridges the pentose phosphate pathway and immune responses in sea cucumber (PMID:32283109)
• Selenium-dependent metabolic reprogramming during inflammation and resolution (PMID:33581115)
• IRAK4 inhibitor mitigates joint inflammation by rebalancing metabolism malfunction in RA (PMID:34732329)
• Deficiency of the sedoheptulose kinase (Shpk) does not alter the ability of hematopoietic stem cells to rescue cystinosis (PMID:34823997)
• Sedoheptulose Kinase SHPK Expression in Glioblastoma (PMID:35682658) and its Correction (PMID:39941164)
• The sedoheptulose kinase CARKL controls T-cell cytokine outputs and migration (PMID:39669692)
• Metabolic reprogramming and signalling networks in microglial activation (PMID:42031319)
• Untargeted metabolomics… non-oxidative branch of the pentose phosphate pathway (PMID:32828637)
• OMIM #617213 — Sedoheptulokinase deficiency (403 to automated retrieval)
• OMIM *605060 — Sedoheptulokinase; SHPK (403 to automated retrieval)
• Orphanet: Isolated sedoheptulokinase deficiency (ORPHA:440713) / Orphadata cross-referencing API
• GARD — Isolated sedoheptulokinase deficiency
• NORD/MONDO — isolated sedoheptulokinase deficiency
• HPO / ontology.jax.org annotations for OMIM:617213 and ORPHA:440713
• NCBI Gene — SHPK (GeneID 23729) · GTR — SHPK
• UniProt Q9UHJ6 — SHPK_HUMAN
• ClinVar VCV000372202 — SHPK c.355C>T (p.Arg119Ter) · ClinVar VCV000372203 — SHPK c.211G>T (p.Glu71Ter)
• dbSNP rs144071313
• MGI:1921887 — mouse Shpk · IMPC — Shpk · MMRRC:043666-UCD
• GeneReviews — Cystinosis
• Disorders of the Pentose Phosphate Pathway and Polyol Metabolism (Springer)
• ClinicalTrials.gov API v2 — NCT03897361, NCT05146830, NCT06910813
• EBI OLS4 (ChEBI, GO term verification)
• OMIA — Online Mendelian Inheritance in Animals (no SHPK phene)