RAB23-related Carpenter Syndrome (CRPT1): Comprehensive Disease Characterization Report

Summary

RAB23-related Carpenter syndrome (Carpenter syndrome type 1, CRPT1; historically acrocephalopolysyndactyly type II; OMIM #201000; ORPHA:65759; MONDO:0008544) is an ultra-rare autosomal-recessive multiple-congenital-malformation disorder with an estimated prevalence of roughly 1 in 1,000,000 births. It is caused by biallelic loss-of-function variants in RAB23 (gene OMIM *606144; HGNC:9776; NCBI Gene 51715; UniProt Q9ULC3; chromosome 6p11.2), which encodes a small RAB-family GTPase that functions as a negative regulator of Hedgehog (HH) signaling and a regulator of ciliary membrane trafficking. When RAB23 function is lost, ciliary Smoothened turnover is impaired and downstream GLI-mediated transcription and FGF10–ERK signaling are de-repressed, and primary-cilium formation is perturbed in a cell-type–dependent manner. The convergent developmental consequence is the near-universal clinical dyad of multisuture craniosynostosis and preaxial polysyndactyly, accompanied by frequent obesity, congenital heart disease, cryptorchidism/hypogenitalism, umbilical hernia, and variable (~75%) intellectual impairment.

The molecular pathology is a classic recessive loss-of-function paradigm: most pathogenic alleles are truncating and subject to nonsense-mediated decay (NMD), and even in-frame or missense lesions (e.g., Y79del, disrupting the switch-II region; or C-terminal frameshifts that abolish the prenylatable cysteine) converge on loss of RAB23 activity. A recurrent L145X founder mutation in patients of northern European descent illustrates the population genetics of the disorder. A clinically overlapping but genetically distinct subtype, CRPT2 (OMIM #614976), is caused by biallelic MEGF8 variants and is distinguished by frequent left–right patterning defects and predominantly single-midline-suture synostosis.

Management is entirely symptomatic and reconstructive. Early cranial-vault expansion / fronto-orbital advancement (ideally within 6–12 months of life, and urgently in the setting of raised intracranial pressure) is the cornerstone, complemented by cardiac surgery, hand/foot reconstruction, orchidopexy, and multidisciplinary developmental support. No disease-modifying pharmacotherapy or gene therapy exists. Prognosis is variable and multisystem; intellectual outcome correlates with the presence of cerebral malformations and untreated raised intracranial pressure rather than being invariable, and affected individuals can survive to adulthood and pregnancy.


Key Findings

Finding 1 — RAB23 biallelic loss-of-function is the cause of CRPT1

Carpenter syndrome type 1 is caused by biallelic loss-of-function mutations in RAB23. The gene was identified by homozygosity mapping across 15 independent families, which linked disease to chromosome 6p12.1–q12 and identified five distinct RAB23 mutations (four truncating and one missense). RAB23 encodes a member of the RAB guanosine-triphosphatase (GTPase) family of vesicle-transport proteins and functions as a negative regulator of Hedgehog signaling. The loss-of-function mechanism is reinforced at the transcript level: truncating mutations produce mRNAs that are degraded by nonsense-mediated decay (NMD), an important contributor to pathogenesis. [human clinical / in vitro]

Ontology anchors: gene RAB23 (HGNC:9776); GO:0007224 (smoothened signaling pathway); GO:0045879 (negative regulation of smoothened signaling pathway).

Finding 2 — Core phenotype: multisuture craniosynostosis + polysyndactyly, with frequent obesity, cardiac defects, and cryptorchidism

Multisuture craniosynostosis and polysyndactyly are present in essentially all molecularly confirmed patients described to date, and abnormal external genitalia (cryptorchidism) are universal in affected boys. The cardinal clinical picture — historically termed acrocephalopolysyndactyly — comprises craniosynostosis, short fingers, soft-tissue syndactyly, preaxial polydactyly, congenital heart disease, hypogenitalism, obesity, and umbilical hernia. As many as three-fourths of patients have some degree of intellectual impairment. No genotype–phenotype correlations are apparent. [human clinical]

Suggested HPO terms: HP:0001363 (Craniosynostosis); HP:0004440 (Coronal craniosynostosis); HP:0100259 (Polysyndactyly); HP:0100258 (Preaxial polydactyly); HP:0001159 (Syndactyly); HP:0001513 (Obesity); HP:0001627 (Abnormal heart morphology); HP:0000028 (Cryptorchidism); HP:0001537 (Umbilical hernia); HP:0001249 (Intellectual disability).

Finding 3 — Recurrent L145X founder mutation; MEGF8 defines the distinct CRPT2 subtype

Among reported patients, 10 individuals were homozygous for the same nonsense mutation, L145X, on a common haplotype, indicating a founder effect in patients of northern European descent. Separately, a clinically overlapping but genetically distinct disorder — CRPT2 — is caused by biallelic MEGF8 variants and is frequently associated with abnormal left-right patterning (situs inversus, dextrocardia, transposition of the great arteries). Laterality defects occur in nearly half of MEGF8 cases but are rare in RAB23 cases, providing a clinically useful discriminator. [human clinical]

Additional recurrent/founder-type alleles reported include c.82C>T p.(Arg28*) (first molecularly confirmed continental-African case, Tanzania; PMID: 33368989) and c.86dupA in a Comorian family (PMID: 20358613).

Finding 4 — RAB23 coordinates suture osteogenesis by repressing FGF-ERK and GLI1; loss causes a context-dependent ciliopathy

In mouse calvarial models, RAB23 is active in osteoblasts at the osteogenic front and regulates both Hedgehog and FGF pathways, repressing FGF10–pERK1/2 and GLI1 during early osteogenesis. Across three independent vertebrate systems — Rab23 conditional-knockout mice, Carpenter-syndrome patient-derived iPSCs, and zebrafish morphants — RAB23 loss recapitulates CS/ciliopathy features and consistently perturbs primary-cilium formation, but in a cell-type–dependent (context-dependent) manner (affecting chondrocytes, mouse embryonic fibroblasts, neural progenitors, and neocortical neurons differently). [model organism / iPSC]

Suggested GO terms: GO:0060348 (bone development); GO:0001503 (ossification); GO:0060271 (cilium assembly); GO:0070848 (response to growth factor).

Finding 5 — RAB23 controls ciliary Smoothened turnover and Kif17 trafficking

Mechanistically, depletion of Rab23 or expression of dominant-negative Rab23 decreases the ciliary steady-state level specifically of Smoothened (but not of control ciliary proteins EB1 or Kim1), implicating RAB23 in protein turnover within the cilium. RAB23 also exists in a complex with the kinesin-2 motor Kif17 and importin β2, and ciliary localization of Kif17 is disrupted in Rab23-depleted cells. RAB23 is enriched at the primary cilium. Together these establish the physical basis by which RAB23 loss dysregulates the ciliary Hedgehog signal-transduction apparatus. [in vitro]

Suggested GO terms: GO:0005929 (cilium); GO:0060170 (ciliary membrane); GO:0042073 (intraciliary transport).

Finding 6 — Structural basis of loss of function: switch-II disruption and loss of prenylation

High-resolution crystal structures of human RAB23 (wild-type and the Y79del clinical mutant, bound to GDP and to the non-hydrolyzable GTP analog GMPPNP) demonstrate that the Y79 deletion causes structural distortions in the switch-II region relative to wild type, potentially disrupting binding to interacting partners and thereby producing loss of function. Clinical point mutations M12K, C85R, and Y79del all fall within the GTPase domain. A second, orthogonal loss-of-function mechanism arises from truncating frameshift variants (e.g., p.Val161Leufs) that remove the C-terminal prenylatable cysteine, so that the truncated protein fails to undergo the lipid modification required to associate with target membranes. [computational/structural / in vitro]

Suggested GO terms: GO:0003924 (GTPase activity); GO:0005525 (GTP binding); GO:0018344 (protein geranylgeranylation).

Finding 7 — Ultra-rare autosomal recessive disorder; model organisms

Carpenter syndrome has an estimated prevalence of ~1 in a million births. Disease models that recapitulate CS features include Rab23 conditional-knockout mice, CS patient-derived iPSCs, and zebrafish morphants. The spontaneous mouse mutant "open brain" (opb) carries a homozygous Rab23 mutation and shows embryonic lethality with open neural-tube defects — a notable species difference, since human RAB23-null homozygosity is not lethal, indicating a divergent early-developmental requirement. RAB23 also regulates Nodal expression in the left lateral plate mesoderm and Kupffer's vesicle, contributing to left–right patterning. [human clinical / model organism]

Finding 8 — Management is multidisciplinary and reconstructive

Management is centered on early surgical release of the fused sutures with fronto-orbital advancement, clearly indicated particularly in cases of elevated intracranial pressure. Early correction of craniofacial deformity is usually safe within 6 to 12 months of life; operative planning uses 3D CT because venous drainage abnormalities and ectatic emissary veins can cause significant intraoperative bleeding. Advanced techniques include cranial-vault remodeling and monobloc distraction osteogenesis. Cardiac defects (e.g., Tetralogy of Fallot) require surgical correction, and treated patients can survive to adulthood and successful pregnancy. No CS-specific pharmacotherapy or gene therapy exists; care is supportive and reconstructive. [human clinical]

Suggested NCIT terms: cranial-vault remodeling / fronto-orbital advancement; distraction osteogenesis (NCIT:C92968); cardiac surgical correction; orchidopexy; rehabilitation therapy.

Finding 9 — Lifelong multisystem morbidity; intellectual outcome is variable, not invariable

Although up to three-fourths of patients show some degree of intellectual impairment, mental retardation is not an invariable feature; the most severe developmental delay is associated with cerebral malformations demonstrable on MRI/CT, so neuroradiologic examination can help predict intellectual outcome. Characteristic craniofacial features include marked absence/underdevelopment of the anterior cranial fossa with bulging of the middle cranial fossa, and there is no correlation between the degree of craniofacial dysmorphology and brain dysmorphology. Congenital and progressive residual cardiac defects contribute to morbidity, and an atypical case associated chronic kidney disease with CS. The phenotypic spectrum has been expanded to include overgrowth with advanced bone age, epileptogenic EEG changes, and autistic features. [human clinical]

Finding 10 — Identifiers and nosology

RAB23-related Carpenter syndrome = CRPT1 / acrocephalopolysyndactyly type II, OMIM #201000, caused by RAB23 (gene OMIM 606144; HGNC:9776; NCBI Gene 51715; UniProt Q9ULC3; chromosome 6p11.2). A second locus, MEGF8 (CRPT2, OMIM #614976), causes a subtype with substantial clinical overlap but frequent laterality defects and typically single-midline-suture synostosis, whereas RAB23-CRPT1 shows multi-suture craniosynostosis. Orphanet ORPHA:65759; MONDO:0008544. [human clinical]*


Mechanistic Model / Interpretation

The unifying interpretation is that RAB23 is a ciliary "brake" on morphogen signaling. Under normal conditions, RAB23 at the primary cilium promotes turnover of Smoothened and correct trafficking of ciliary motors (Kif17), thereby keeping Hedgehog/GLI output — and, in cranial osteoblasts, FGF10–ERK output — appropriately low. Removing this brake (through NMD-mediated protein loss, switch-II GTPase-cycle disruption, or loss of prenylation-dependent membrane targeting) de-represses these pathways. Because different tissues rely on cilium-dependent signaling to different degrees, RAB23 loss manifests as a context-dependent ciliopathy: strongest and most consistent in the developing skull (multisuture synostosis) and limb (polysyndactyly), with variable CNS, cardiac, and metabolic consequences.

Ordered causal chain (initiating lesion → clinical manifestation):

  1. Biallelic RAB23 loss-of-function variant (truncating → NMD; or switch-II/prenylation-disrupting) results in absent/non-functional RAB23 GTPase protein. (Demonstrated.)
  2. Loss of RAB23 at the primary cilium leads to failure of normal ciliary protein turnover — specifically dysregulated ciliary Smoothened and disrupted Kif17 motor trafficking. (Demonstrated in vitro.)
  3. Dysregulated ciliary Smoothened results in de-repression of Hedgehog (GLI-mediated) signaling; in parallel, RAB23 loss de-represses FGF10–pERK1/2 signaling at the cranial osteogenic front. (Demonstrated in calvarial models.)
  4. Branch A (skull): De-repressed HH/GLI1 + FGF-ERK in cranial osteoblasts leads to premature/accelerated osteogenic differentiation and multisuture craniosynostosis. (Demonstrated.)
  5. Branch B (limb): Altered HH gradient in the limb bud results in preaxial polydactyly/polysyndactyly. (Inferred from HH-pathway biology and model phenotypes.)
  6. Branch C (CNS/cardiac/other): Context-dependent perturbation of primary-cilium formation and Nodal regulation leads to cerebral malformations, variable intellectual impairment, cardiac defects, and (rarely) laterality anomalies. (Partly demonstrated, partly inferred.)
 RAB23 biallelic LOF (NMD / switch-II / no prenylation)
                 │
                 ▼
   Loss of ciliary RAB23 function
        (dysregulated Smoothened turnover, ↓Kif17 trafficking)
                 │
        ┌────────┴─────────┐
        ▼                  ▼
 De-repressed HH/GLI1   De-repressed FGF10–pERK1/2
        │                  │
        └────────┬─────────┘
                 ▼
   Aberrant osteogenic & patterning programs
        │           │            │
        ▼           ▼            ▼
  Multisuture   Preaxial     Cilium-dependent
 craniosynostosis polysyndactyly CNS/cardiac defects

This model explains the near-universal core dyad, the variable expressivity, the correlation of cognitive outcome with cerebral malformation, and the absence of genotype–phenotype correlation (all pathogenic alleles converge on the same loss-of-function endpoint). It also clarifies why the closely related CRPT2 (MEGF8) shares the craniosynostosis/limb phenotype yet reaches it through a distinct, largely Hedgehog-independent BMPR1A–BMP-SMAD route and adds laterality defects (PMID: 42399640).


Section-by-Section Report

1. Disease Information

Overview. RAB23-related Carpenter syndrome is a rare autosomal-recessive syndromic craniosynostosis (an "acrocephalopolysyndactyly") defined by the co-occurrence of multisuture craniosynostosis and polysyndactyly with a constellation of additional malformations (obesity, congenital heart disease, hypogenitalism, umbilical hernia, and frequently intellectual impairment).

Key identifiers.

Resource Identifier
OMIM (disease, CRPT1) #201000
OMIM (gene) *606144 (RAB23)
Orphanet ORPHA:65759
MONDO MONDO:0008544
HGNC HGNC:9776 (RAB23)
NCBI Gene 51715
Ensembl ENSG00000112210
UniProt Q9ULC3
Cytoband 6p11.2
ICD-10 Q87.0 / Q75.x
SNOMED CT 21086008 (Acrocephalopolysyndactyly)
MeSH Acrocephalopolysyndactyly

Synonyms: Carpenter syndrome; Carpenter syndrome type 1 (CRPT1); acrocephalopolysyndactyly type II (ACPS II); ACPS2.

Data source type: Information is derived from aggregated disease-level resources (OMIM, Orphanet) and from individual patient reports/case series in the primary literature; the disorder is too rare for EHR-scale cohorts.

2. Etiology

Causal factor: purely genetic — biallelic loss-of-function variants in RAB23 (Finding 1). There is no known environmental, infectious, or toxic contribution to CRPT1.

Genetic risk factors: The disease requires two pathogenic RAB23 alleles; heterozygous carriers are unaffected. A founder L145X allele elevates carrier frequency in populations of northern European descent (Finding 3); other recurrent alleles (p.Arg28*, c.86dupA) occur in specific pedigrees/populations. Consanguinity substantially raises risk owing to autosomal-recessive inheritance (multiple reported families are consanguineous). Genetic heterogeneity exists at the disease level: MEGF8 causes CRPT2.

Environmental / lifestyle / protective factors / gene–environment interactions: None established. No environmental risk factors, protective genetic or environmental factors, or gene–environment interactions have been demonstrated for this Mendelian disorder. (Not applicable / not available.)

3. Phenotypes

Onset is congenital; features are structural. Frequencies are qualitative given small cohorts.

Phenotype Type Frequency HPO term
Multisuture craniosynostosis (often bicoronal + sagittal + metopic; cloverleaf/turricephaly) Physical/skeletal Near-universal (~100%) HP:0001363 / HP:0002676
Polysyndactyly (preaxial polydactyly, cutaneous syndactyly, brachydactyly) Physical/skeletal Near-universal (~100%) HP:0100259 / HP:0100258 / HP:0001159
Cryptorchidism / abnormal genitalia (males) Physical Universal in boys HP:0000028
Obesity Physical/metabolic Frequent HP:0001513
Congenital heart disease (ASD, VSD, PDA, ToF, TGA) Clinical sign Frequent (~30–50%) HP:0001627
Intellectual disability / developmental delay Behavioral/cognitive ~75% (variable) HP:0001249
Umbilical hernia Physical Frequent HP:0001537
Characteristic facies (flat nasal bridge, hypertelorism, low-set ears) Physical Frequent HP:0000316 / HP:0005280
Genu valgum / short stature / skeletal dysplasia Physical/skeletal Variable HP:0002857 / HP:0004322
Hydrocephalus / cerebral malformations Clinical sign Occasional HP:0000238 / HP:0002011
Corneal/ophthalmic anomalies Clinical sign Occasional HP:0007957
Atypical: overgrowth/advanced bone age, seizures, autistic features Various Rare HP:0001548 / HP:0001250 / HP:0000729

Onset: congenital (some prenatally detectable). Severity/progression: structural anomalies are static in origin but craniosynostosis can drive progressive raised intracranial pressure; cardiac lesions may progress. Variable expressivity, including intrafamilial (PMID: 20358613). Quality of life: substantial and lifelong (reconstructive-surgery burden, motor/orthopedic complications, cardiac limitation, cognitive outcome); no CS-specific validated QoL instrument data exist.

4. Genetic / Molecular Information

Causal gene: RAB23 (6p11.2), ~237-aa small GTPase, 6 coding exons. Variant spectrum: predominantly truncating (nonsense, frameshift, splice-site) with occasional missense/in-frame deletions; ≥12 distinct mutations across dozens of families. Classification: biallelic pathogenic/likely-pathogenic per ACMG/AMP (PVS1 for null alleles; segregation; functional evidence). Functional consequence: loss of function via (i) NMD of truncating transcripts, (ii) switch-II structural disruption (Y79del), and (iii) loss of C-terminal prenylation/membrane targeting (Findings 1, 6). Allele frequency: pathogenic alleles are extremely rare in gnomAD. Origin: germline. Modifier genes / epigenetics / large chromosomal abnormalities: none established for CRPT1 (karyotype typically normal).

Representative variants: L145X (founder, N. European); p.Arg28* (Tanzania); c.86dupA (Comoros); c.481G>C p.Val161Leufs*16 (exon-6 skipping, prenylation loss); M12K, C85R, Y79del (GTPase-domain).

5. Environmental Information

Not applicable. CRPT1 is a monogenic disorder with no established environmental, lifestyle, or infectious contributors. (Obesity, once present, is a genetically driven feature that may be modifiable by diet/lifestyle as supportive care, but is not an environmental cause.)

6. Mechanism / Pathophysiology

See the "Mechanistic Model / Interpretation" section above for the full ordered causal chain and diagram.

7. Anatomical Structures Affected

8. Temporal Development

9. Inheritance and Population

10. Diagnostics

11. Outcome / Prognosis

12. Treatment

No disease-modifying, pharmacologic, gene, cell, or RNA therapy exists. Management is symptomatic, reconstructive, and multidisciplinary.

13. Prevention

14. Other Species / Natural Disease

15. Model Organisms

Model Type Key features / recapitulation Reference
"Open brain" (opb) mouse Spontaneous mammalian mutant Open neural-tube defect, embryonic lethal (more severe than human); established Rab23 as Shh antagonist PMID: 29727300
Rab23 conditional-KO mouse Genetic (conditional) Best mammalian model; skeletal/chondrocyte/neural CS features; context-dependent cilia defects PMID: 40825043
Calvarial/osteoblast explant Ex vivo RAB23 represses FGF10-pERK1/2 & GLI1 in osteogenesis PMID: 32662771
CS patient-derived iPSCs In vitro human Perturbed cilium formation, context-dependent PMID: 40825043
Zebrafish morphants Vertebrate Ciliopathy/patterning defects; Nodal/laterality PMID: 40825043
MDCK/knockdown cells; recombinant RAB23 In vitro / structural Ciliary Smoothened/Kif17 trafficking; crystal structures (WT, Y79del) PMID: 20375059, PMID: 26136363, PMID: 39615683

Phenotype recapitulation & limitations: models reproduce craniofacial/skeletal defects, ciliary dysfunction, and Hedgehog/Nodal dysregulation, but no single model captures the full human multisystem spectrum; the mouse null's lethality and species-specific developmental requirements limit direct translation. Resources: MGI (mouse), ZFIN (zebrafish), IMPC/KOMP, Cellosaurus (iPSC lines), PDB (RAB23 structures).


Evidence Base

PMID Contribution Role
17503333 Gene discovery; RAB23 as HH negative regulator; L145X founder Foundational — supports F1, F3
21412941 NMD of truncating alleles; universal core features Supports F1, F2
8352858 Clinical spectrum; cerebral malformation predicts cognition Supports F2, F9
23063620 MEGF8/CRPT2 with laterality defects Supports F3, F10
32662771 RAB23 represses FGF10-pERK1/2 and GLI1 in osteogenesis Supports F4
40825043 Context-dependent ciliopathy across 3 models Supports F4, F7
20375059 RAB23 regulates ciliary Smoothened turnover Supports F5
26136363 RAB23–Kif17 ciliary trafficking Supports F5
39615683 Crystal structure; Y79del disrupts switch-II Supports F6
23599695 Loss of prenylatable cysteine → membrane-targeting failure Supports F6
39040725 Prevalence ~1/1,000,000; CKD association Supports F7, F9
29727300 Open brain mouse; species difference Supports F7
25162549 FOA indicated, esp. raised ICP Supports F8
34244844 Surgical timing 6–12 mo; bleeding risk Supports F8
34748996 Expanded phenotype/mutations Supports F9
38760421 CRPT1 multi-suture vs CRPT2 single-midline Supports F10
42399640 MEGF8 BMP-SMAD mechanism (contrast to RAB23-FGF-ERK) Mechanistic contrast
25168863 Prenatal findings; novel splice variant Supports Diagnostics
20358613 Comorian family; intrafamilial variability Supports Inheritance
33368989 First continental-African case (R28X) Supports Population
23706836 Adult survival/pregnancy; cardiac progression Supports Prognosis

Consistency: No contradictory findings were encountered. All ten confirmed findings are mutually reinforcing, spanning human genetics, structural biology, cell biology, and clinical management. The one apparent tension — mouse null lethality vs. viable human null — is explicitly reconciled as a species-specific developmental requirement.

Limitations and Knowledge Gaps

Proposed Follow-up Experiments / Actions

  1. Assemble an international CRPT1 registry to quantify phenotype frequencies, natural history, raised-ICP incidence, and neurodevelopmental outcomes with denominators.
  2. Systematic RAB23 variant curation (ClinVar/gnomAD reconciliation) with functional assays (GTPase cycling, ciliary Smoothened turnover, membrane association) to standardize ACMG classification and probe genotype–phenotype signals.
  3. Test pathway-targeted rescue in Rab23-CKO mice and patient iPSC-derived cranial mesenchyme: SMO/GLI antagonists and FGF-ERK inhibitors, benchmarking against BMP-receptor-inhibition rescue reported for MEGF8/CRPT2.
  4. Single-cell and spatial transcriptomics of cranial sutures in Rab23 models to map cell-type-specific de-repression of HH and FGF programs and identify the "context" determining the cilium phenotype.
  5. Prospective neuroimaging-outcome study to validate MRI/CT cerebral-malformation findings as a quantitative prognostic biomarker for intellectual outcome.
  6. Search for modifier loci/epigenetic marks underlying intrafamilial variability using trio/family designs.

Report compiled from 10 confirmed findings across 33 reviewed papers over 5 investigation iterations. Evidence types: human clinical (case series/reports), model organism (mouse, zebrafish), in vitro (iPSC, cell lines), and computational/structural (crystallography). Frequencies are approximate given the ultra-rare nature of the disorder; no large-scale omics, QoL, or survival datasets currently exist for CRPT1.