SEARCH

SEARCH BY CITATION

Keywords:

  • malformations;
  • genetic heterogeneity;
  • medical diagnosis;
  • pleiotropism

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. PERSPECTIVES AND FUTURE DIRECTIONS
  7. REFERENCES

Limb development is clinically and biologically important. Polydactyly is common and caused by aberrant anterior–posterior patterning. Human disorders that include polydactyly are diverse. To facilitate an understanding of the biology of limb development, cataloging the genes that are mutated in patients with polydactyly would be useful. In 2002, I characterized human phenotypes that included polydactyly. Subsequently, many advances have occurred with refinement of clinical entities and identification of numerous genes. Here, I update human polydactyly entities by phenotype and mutated gene. This survey demonstrates phenotypes with overlapping manifestations, genetic heterogeneity, and distinct phenotypes generated from mutations in single genes. Among 310 clinical entities, 80 are associated with mutations in 99 genes. These results show that knowledge of limb patterning genetics is improving rapidly. Soon, we will have a comprehensive toolkit of genes important for limb development, which will lead to regenerative therapies for limb anomalies. Developmental Dynamics 240:931–942, 2011. © 2011 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. PERSPECTIVES AND FUTURE DIRECTIONS
  7. REFERENCES

Polydactyly results from defective patterning of the anterior–posterior axis of the developing limb and can occur as a simple or isolated malformation or as part of a pleiotropic developmental anomaly syndrome. Polydactyly is an important manifestation in clinical medicine not only because it has cosmetic and functional implications and often necessitates surgical treatment, but because it can serve as an immediately recognizable indicator that the patient, particularly a newborn, has a multiple congenital anomaly syndrome (pleiotropic developmental anomaly syndrome). For these reasons, it is important for the clinician to be prepared to evaluate the patient for polydactyly and consider the myriad syndromes that may be associated with this anomaly. In addition to the clinical manifestations, polydactyly can be considered by the scientist as an experiment of nature, in that the manifestation of this phenotype points to an underlying mutated gene or teratogen that is associated with a perturbation of the molecular machinery of limb development (Towers and Tickle,2009). The elucidation of the etiologies of polydactyly can serve as a tool to facilitate the dissection of the genetic and signaling pathways that underlie polydactyly, specifically, that of anterior–posterior specification of the limb (for recent review, see Butterfield et al.,2010).

For both clinical utility and to facilitate basic science investigations, it is useful to catalog the phenotypes that include polydactyly as a manifestation. In 2002, I cataloged the entities recognized by geneticists in the clinic that included polydactyly as a manifestation (Biesecker,2002). At that time, 97 clinical entities were listed, of which 37 were associated with mutations in one or more genes. Since that time, much progress has been made in molecular genetics of disease and it seemed appropriate to update this review. (These data were presented at the 6th International Limb Development and Regeneration Conference in Williamsburg, Virginia, USA, July 16, 2010.) In this article, I will delineate the disorders that have been associated with polydactyly, delineate the genes that are altered in these disorders, and speculate on the future implications of these data.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. PERSPECTIVES AND FUTURE DIRECTIONS
  7. REFERENCES

To tabulate syndromic and isolated polydactyly clinical entities, I drew from several sources. First, Mendelian Inheritance in Man (Anon,2010) was searched using the term “*polydactyl*”. Second, the Winter-Baraitser Dysmorphology Database (WBDD) was interrogated using the following terms: “postaxial polydactyly of fingers,” “postaxial polydactyly of toes,” “preaxial polydactyly of fingers,” “preaxial polydactyly of toes,” “meso-axial polydactyly of fingers,” “mesoaxial polydactyly of toes,” “palmar polydactyly,” “polydactyly/bifid thumb,” “polydactyly bifid hallux,” “mirror image polydactyly of fingers,” and “mirror image polydactyly of toes.” Third, the tabular listings of polydactyly in the appendix of Smith's Recognizable Patterns of Malformation (Jones and Smith,2006) and Tables 21-1 to 21-7 of the chapter “Hands and Feet” in Human Malformations and Related Anomalies (Stevenson,2006), were reviewed. For other online sources, see Mendelian Inheritance in Man at http://www.ncbi.nlm.nih.gov/omim.

These lists were merged and duplicate entries were deleted, although, as will be noted in the discussion, this can be challenging. In contrast to the prior analysis, I included entities where only a single family or case was described. Some of these disorders must be considered as candidate polydactyly syndromes as it is possible in some cases that the association of polydactyly in these traits was coincidental. Many of these ultra rare or unique cases emanated from the Winter-Baraitser Dysmorphology Database. If a common malformation syndrome had a single case with polydactyly, that entity was deleted from this analysis, as it seemed more likely that it was coincidental, rather than causal. For rare malformation syndromes, it can be difficult to determine if a particular manifestation is uncommon or coincidental. In these cases, the entity was generally included in the listing.

It can also be challenging to determine the boundaries of the descriptor of polydactyly itself. For example, broad thumbs or great toes are potentially mechanistically related to anterior–posterior patterning. There is a large degree of interindividual variation in the width of these digits, arguing against inclusion of this finding as part of the spectrum of polydactyly. In contrast, in disorders with preaxial polydactyly (e.g., Greig cephalopolysyndactyly syndrome) a full spectrum of abnormalities are observed, including complete digit duplication, partial duplication, U shaped or forked distal phalanges, and broad digits, arguing for inclusion. In this review, I generally required frank polydactyly to be reported in the entity for it to be included. Next, entries that described polydactyly only in model organisms and entities that were associated with segmental aneusomy (and not with a specific gene) or teratogens were deleted. This is not because these entities are unimportant for understanding limb development. Instead, this was done because the purpose of this review was to review the genes that contribute to polydactyly and it can be difficult to identify the gene products that are pathogenetically important in many teratogens and segmental aneusomy syndromes. Entries that separately described a gene and a disorder (e.g., GLI3 and Pallister-Hall syndrome) were reduced to a single clinical entry. In recent years, there has been evolution of the MIM database with regard to clinical and gene entries. For some disorders, MIM has consolidated the clinical data into a single entry (e.g., Bardet-Biedl syndrome MIM 209900) with additional entries for the genes, and the gene entries have little or no clinical data. For other phenotypes, such as asphyxiating thoracic dystrophy (Jeune) and Meckel syndromes, there are multiple entries with genetic and clinical data.

The general approach in this review was to treat clinically distinct phenotypes as single entities, independent of genotype (Robin and Biesecker,2001). As is well known to clinical geneticists, it can be difficult to determine the boundaries that distinguish some clinical entities from overlapping disorders. For example, it can be challenging to clinically distinguish a mildly affected case of Jeune syndrome (MIM 208500) from Ellis-Van Creveld syndrome (MIM 225500). In addition, current standards of clinical data reporting and limitations of commonly used terminology limit the ability to finely parse the data in the clinical literature (Biesecker,2005; Allanson et al.,2009). For this reason, it can be difficult to accurately determine the presence or absence of some malformations in clinical reports, much less to know the precise form of that manifestation—for example, whether a case of polydactyly was postaxial or central. As well, it can be difficult to distinguish entities that have both syndromic and nonsyndromic manifestations, such as Greig Cephalopolysyndactyly syndrome (MIM 175700) and Crossed Polydactyly (MIM 174700), both of which can be caused by mutations in GLI3 (Biesecker,2006). In these cases, I generally deferred to the curators of MIM, WBDD, and the Stevenson et al textbook regarding the delineation of these entities. With respect to genes, this is generally straightforward to tabulate, with the exception of regulatory elements. Three entries were associated with such elements of the SHH and SOX9 genes.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. PERSPECTIVES AND FUTURE DIRECTIONS
  7. REFERENCES

This resulting list comprised 310 entries of disorders that include polydactyly, including syndromic (290 entries) and nonsyndromic (20 entries) forms (Table 1). This list is significantly larger than that from the 2002 review, primarily owing to the addition of the WBDD entities, improved curation, as well as general advances in medical knowledge. These 310 entities show a full range of Mendelian inheritance patterns, including autosomal dominant (73), autosomal recessive (103), X-linked (12), more than one inheritance pattern (9), but many had an unknown mode of inheritance (113).

Table 1. Clinical Entities That Manifest Polydactylya
inline imageinline imageinline image

Among the 310 entities, 80 have been associated with causative mutations in a total of 99 genes. This represents major progress, as the 2002 analysis identified only 39 entities associated with mutations. As before, these disorders illustrate another prominent feature, which is that of genocopies and pleiotropism. First, there are several entities that can be caused by mutations in more than one gene (genocopies). Among the entries with identified mutations, Fanconi anemia can be caused by homozygous or compound heterozygous mutations in one of 14 genes and Bardet-Beidl syndrome can be caused by mutations in at least 12 genes. These two disorders alone account for 26 of the 99 genes known to cause polydactyly when mutated. The remaining 308 entries are associated with mutations in 73 genes (note that MKKS is counted in both categories because distinct mutations in that gene can cause either McKusick-Kaufman syndrome or Bardet-Biedl syndrome). These 73 genes are associated with 95 distinct clinical entities, which is an expression of the concept of pleiotropy. A major culprit in this pleiotropy is the GLI3 transcription factor gene, which can be attributed to five phenotypes (Kang et al.,1997; Radhakrishna et al.,1997,1999; Vortkamp et al.,1991; Elson et al.,2002). In the 2002 review, there were three genes associated with two phenotypes each: MKKS, EVC, and DHCR7 (Cormier-Daire et al.,1996; Wassif et al.,1998; Katsanis et al.,2000; Ruiz-Perez et al.,2000; Slavotinek et al.,2000; Stone et al.,2000). At the time of the present analysis, the Rutledge lethal multiple congenital anomaly syndrome, associated with DHCR7 mutations, has been collapsed into Smith-Lemli-Opitz syndrome, because it was determined that the two, previously considered distinct entities, actually formed a continuous spectrum of phenotype. As of the present review, there are 15 sets of multiple phenotypic entities that have been shown to be caused by mutations in a single gene (Table 2).

Table 2. Genes That, When Mutated, Can Cause More Than One Clinically Distinct Phenotype in Humansa
GeneClinical entity
  • a

    Note that these phenotypes can be caused by mutations in more than one of the genes listed in the left column. **Note that these mutations are in the SHH ZRS (ZPA regulatory sequence) and not in the SHH gene proper.

CD96C syndrome
C-like syndrome
CXORF5Joubert syndrome*
Oral-facial-digital syndrome type I
DYNC2H1Asphyxiating thoracic dystrophy type 3
Short-Rib polydactyly syndrome
EVCEllis-Van Creveld syndrome
Weyers acrofacial dysostosis
FBLN1De Smet complex synpolydactyly
Synpolydactyly*
FGFR2Apert syndrome
Lacrimoauriculodentodigital syndrome
FLNBAtelosteogenesis type III
Spondylocarpotarsal dysostosis
GDF5Brachydactyly type C
Chondrodysplasia, Grebe type
GLI3Greig cephalopolysyndactyly syndrome
Pallister-Hall syndrome
Acrocallosal syndrome
Polydactyly, postaxial
Polydactyly, preaxial*
INPP5BJoubert syndrome*
Mental retardation, truncal obesity, retinal dystrophy, and micropenis
MKKSBardet-Biedl syndrome
McKusick-Kaufman syndrome
ROR2Brachydactyly type B
Robinow syndrome
RPGRIP1LJoubert syndrome*
Meckel Syndrome
SHH**Polydactyly, preaxial*
Synpolydactyly*
WNT7AFibular a/hypoplasia, femoral bowing and poly-, syn-, and oligodactyly
Ulna and fibula, absence of, with severe limb deficiency

These results illustrate several important principles of clinical medicine, medical diagnosis, and the taxonomy of disease. The boundaries that delineate and demarcate disorders from other disorders can be difficult to determine. Essentially all human pleiotropic developmental anomaly syndromes overlap to a small, or sometimes large, extent with one or more other syndromes. It is not always possible to define unambiguous phenotypic categories; therefore, there is debate about the boundaries of clinical entities. As noted above, it can also be difficult to rigorously distinguish syndromic from nonsyndromic phenotypes.

In the 1980s, it was presumed that molecular diagnosis would solve these problems. In fact, an unsupported but commonly held aphorism in those times was “one gene—one disease.” A large body of work has overthrown that naive view, and we now recognize that the relationship of genotype and phenotype is much more complex and nuanced. An example of this is the situation with the McKusick-Kaufman and Bardet-Biedl syndromes. The McKusick-Kaufman syndrome comprises polydactyly, congenital heart disease, and hydrometrocolpos (obstruction secondary to imperforate uterovaginal plate) in females and hypospadias in males (Slavotinek and Biesecker,2000). This phenotype is almost entirely limited to the Old Order Amish with ancestors in Southeastern Pennsylvania and is caused by a homozygous double mutation in the MKKS gene (homozygous c.[726C>G + 250C>T], p.[Ala242Ser + His84Tyr]; Stone et al.,2000). In contrast, all other homozygous or compound heterozygous mutations in MKKS cause the Bardet-Biedl syndrome (polydactyly, pigmentary retinal dystrophy, obesity, intellectual disability, hypogonadism, and other anomalies; Katsanis et al.,2000; Slavotinek et al.,2000). Although these two phenotypes can overlap in childhood, the adult manifestations are readily distinguished and the consequences for the patients are markedly distinct. In addition, the phenotypes breed true—that is, families with true McKusick-Kaufman syndrome do not have members affected with Bardet-Biedl syndrome and vice versa. They are therefore considered to be distinct clinical entities.

To further complicate the situation, it is now known that Bardet-Biedl syndrome can be caused by mutations in at least 11 genes in addition to MKKS. This is an example of how mutations in one gene can cause more than one phenotype (as shown in Table 2) and mutations in more than one gene can apparently cause a single phenotype (e.g., Table 1 entries for Bardet-Biedl or Fanconi syndromes). This means that the genetic etiology alone cannot completely solve the diagnostic challenges that clinicians face, although they can be extremely helpful in confirming or eliminating diagnoses and narrowing the list of potential disorders that the clinician must consider.

The classes of genes products can be interpreted as a reflection of categories of genes that are important in mammalian development. Whereas transcription factors were the largest group in the prior analysis, the current gene list is dominated by genes that encode proteins involved in cell signaling (21 genes) and genes that encode for proteins in the basal body and cilium (13 genes). The other major categories of gene products implicated in phenotypes associated with polydactyly are the DNA repair genes (15 genes, primarily attributable to the Fanconi anemia phenotypes), followed by the transcription factors (16 genes). Much less commonly involved (fewer than five identified) genes encode structural proteins, catalytic proteins, immunoglobulin superfamily proteins, chaperones, and gap junctions. This significant shift in the distribution of identified genes since 2002 reflects the rapidly increasing emphasis on signaling pathways in development (Duboc and Logan,2009) and the key role of the cilium and basal body in the Sonic hedgehog-GLI3 transcription factor pathway (Goetz and Anderson,2010).

PERSPECTIVES AND FUTURE DIRECTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. PERSPECTIVES AND FUTURE DIRECTIONS
  7. REFERENCES

From the data reviewed above, it is clear that a great deal of progress has been made in elucidating the etiology of polydactyly since the previous review in 2002. Recent technologic developments, especially that of massively parallel sequencing (Shendure and Ji,2008), have shown remarkable initial promise to accelerate the process of gene identification in rare disorders. The coupling of massively parallel sequencing to various implementations of target selection (Gnirke et al.,2009) to enrich DNA samples for protein-coding exons has been shown to be effective for mutation detection. Several disorders have recently been elucidated by various implementations of this technology including Miller syndrome (Ng et al.,2010) and TARP syndrome (Johnston et al.,2010). These technologies accelerate human genetic discovery because they have the capability not only to increase sequence acquisition, but to potentially bypass meiotic mapping, which can be limiting in ultra rare disorders. As many of the disorders in Table 1 that have not yet been associated with mutations in a gene are known to occur in only a few or even in single families, this technologic advance holds out the promise of elucidating the etiology of most, if not nearly all of these disorders. Therefore, one can anticipate that human geneticists will, in the next several years, have a catalog of nearly all of the genes that are mutated in patients with polydactyly phenotypes, thus comprising a rich catalog of pathologic variations as well as a large list of genes important in anterior–posterior patterning in the limb. One can readily imagine that massively parallel sequencing of transcripts in developing limbs will delineate another set of genes, and that these sets can be analyzed and intersected to develop a rich, and perhaps nearly complete catalog of genes important for the patterning of the limb. With this knowledge, one can further predict that developmental biologist will have the resources to characterize the key pathways in the development of the limb in humans and in model systems, which can be used to develop regenerative or cell replacement therapies for limb anomalies.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS AND DISCUSSION
  6. PERSPECTIVES AND FUTURE DIRECTIONS
  7. REFERENCES