Postaxial limb hypoplasia (PALH): the classification, clinical features, and related developmental biology

Abstract

Postaxial limb hypoplasia (PALH) is a group of nonhereditary diseases with congenital lower limb deficiency affecting the fibular ray, including fibular hemimelia (FH), proximal femoral focal deficiency (PFFD), and tarsal coalition (TC). The etiology and the developmental biology of the anomaly are still not fully understood. Here, we review the previous classification systems, present the clinical features, and discuss the developmental biology of PALH.

Introduction

Postaxial limb hypoplasia (PALH) is a group of nonhereditary diseases with congenital lower limb deficiency. It affects the ipsilateral hip, femur, knee, tibia/fibula, ankle, and foot. PALH includes fibular hemimelia (FH), proximal femoral focal deficiency (PFFD), and tarsal coalition (TC).¹ FH has been known for over 300 years and was first described by Gollier in 1698.² FH is characterized by the partial or total absence of the fibula. Patients with FH often require surgical reconstruction.³ However, if the ankle joint cannot be formed in FH patients, the only treatment option is amputation.⁴ PFFD affects the pelvis, particularly the hip bone, and the proximal femur.⁵ The disorder may affect one side or both, with the hip becoming deformed and the leg shortened. TC is an abnormal bridge of tissue that connects two separate tarsal bones.⁶ The onset of symptoms related to TC usually occurs at about 9–17 years of age, with a peak incidence occurring at 10–14 years of age. When the bridging link becomes bone, a limitation of motion results, and the onset of pain is initiated. The bones of the tarsus are the most posterior bones of the foot. These bones form the two major foot joints: the subtalar and midtarsal joints. These joints allow complex motions to occur in the feet, which are necessary for activities such as traveling over uneven terrain and creating a gait for the function of the knees, hips, and back. The basic information concerning the three types of PALH is summarized in Table 1. The etiology of PALH appears to be related to developmental defects that occur during the embryonic period, but no genetic factor has been identified.

Table 1.

The classification, etiology, and clinical features of the three types of PALH disorders.

	Commonly used classification system	Etiology	Clinical features
FH	The Achterman and Kalamchi classification	Unknown	Mild hypoplasia to complete absence of the fibula, with variable shortening of the tibia
PFFD	The Aiken classification	Unknown	Variable degrees of shortening or absence of the femoral head, with associated dysplasia of the acetabulum and femoral shaft
TC	The Downey classification	Unknown	The fusion of two or more bones of the hindfoot

FH, fibular hemimelia; PFFD, proximal femoral focal deficiency; TC, tarsal coalition.

Fibular hemimelia

FH is a congenital deficiency or absence of the fibula. It may be associated with anatomical abnormalities in the lower limbs.⁷ It is rare but still one of the most common congenital deficiencies of the long bones, with an estimated incidence of 7.4–49 per 1 million live births.^2,8–10

More than 40 years ago, Thompson et al.¹¹ used the term “congenital absence of the fibula” to describe the anomaly that bears its name, which emphasizes only one aspect of a complex of congenital malformations. However, FH often involves femoral and tibial hypoplasia, hypoplasia of the lateral femoral condyle and anteromedial bowing of the tibia, valgus or varus deformity of the ankle and deficiency of the lateral ray of the foot, ball-and-socket ankle joint deformity, and tarsal coalition of the foot.^4,12–14 The hypoplasia of the lateral femoral condyle and anteromedial bowing of the tibia could lead to genu valgum or knock-knee, and valgus or varus deformity of the ankle and deficiency of the lateral ray of the foot may preclude plantigrade walking. The term “postaxial hypoplasia” may be more appropriate in describing this disease as compared with traditional terms FH or “fibular hypoplasia,” since it mainly affects lower extremities.¹

Most reported cases of FH are sporadic. The virtual lack of inheritable recurrence suggests that FH may be caused by mosaic somatic mutations. The presence of fibular hypoplasia is sufficient to define FH as a developmental defect. However, the etiology and the developmental biology of the anomaly are still not fully understood.

Classification of FH

There are currently many different classification systems for FH. In 1952, Coventry and Johnson² gave the earliest classification of FH, describing the severity of the deficiency and the associated findings. The commonly used classification system to describe the limb hypoplasia was introduced by Achterman and Kalamchi in 1979.¹⁵ This classification divided FH into two types: type I FH described limbs in which a portion of the fibula was present, and type II FH described limbs where there was a complete absence of the fibula. The type I group was further subdivided according to the extent of the deficiency.¹⁵ In type IA, the proximal fibular epiphysis is distal to the level of the tibial growth plate and is often smaller than the normal side, while the distal fibular growth plate is proximal to the dome of the talus. In type IB, there is a partial absence of the fibula: proximally the fibula is absent for 30–50% of its length, while distally it is present but does not support the ankle.¹⁵

A new classification system was recently suggested by Stanitski et al.¹⁶ on the basis of the morphology of the fibula and ankle, hindfoot coalition, and foot ray deficits. In this classification system, three types of fibula have been suggested: (1) nearly normal fibula; (2) small or miniature fibula, regardless of the position of the fibula in the limb; and (3) complete absence of the fibula. The three types of tibiotalar joint and distal tibial epiphyseal morphology are categorized: H = horizontal; V = valgus (triangular distal tibial epiphysis); and S = spherical (ball-and-socket). The lowercase “c” is used to describe the tarsal coalition, and the number of foot rays (medial to lateral) is numerically quantified (1–5).¹⁶ This classification system reflects the broad expression of this disorder and the unpredictable interrelationships between the foot and ankle.¹⁶ This classification includes additional data on the status of the foot and ankle and can help in clinical decisions about limb salvage or ablation.

Most recently, the Paley classification¹⁷ became the first classification of FH to be designed with reconstructive surgery options in mind. The Paley classification divided FH into three types: type 1––stable ankle; type 2––dynamic valgus; and type 3––fixed equino-valgus. Type 3 can be subdivided into three groups: type 3A––the fixed equino-valgus deformity is due to a malorientation of the ankle joint; type 3B––there is a malunited subtalar coalition; and type 3C––co-existence of both distal tibial malorientation and malunited subtalar coalition. Each type has a different surgical treatment, so it is a reconstructive surgery classification.

Clinical features of FH

We retrospectively reviewed the records and radiographs of patients with FH treated in Shanghai Sixth People’s Hospital, China between 2014 and 2016. A total of 15 patients (seven females and eight males) with 15 affected limbs (six left and nine right legs affected) were identified. The clinical features of each patient are described in Table 2. There were no major malformations in organ systems other than limbs in these patients. Ages ranged from 3 to 47 years, with an average age of 13.9 years at their first visit. The radiographs were evaluated, along with the clinical data, to determine the presence of the following: lateral femoral condylar hypoplasia, ball-and-socket ankle joint malformation, tarsal coalition, and forefoot ray deletion. Using the Achterman and Kalamchi classification system, six patients (Patient No. 1–6) had type IA (Fig. 1A), one (No. 7) type IB (Fig. 1B), and eight (No. 8–15) type II FH (Fig. 1C). Almost all patients had hypoplasia of the lateral femoral condyle (Fig. 1, white arrows) and tarsal coalition (Fig. 2A) with different severity. Four patients (No. 8–10 and 13) with type II FH had anteromedial bowing of the tibia (Fig. 2B). The ball-and-socket ankle joint (Fig. 2C) was observed in eight patients older than 11 years of age, while, in seven patients below 6 years of age, the ankle was poorly developed (Fig. 2B). There were four patients (No. 8–10 and 12) with anteromedial bowing of the tibia. Four limbs (No. 1–3 and 5) had five rays, but patient No. 5 had hypoplasia of the fifth ray. One limb (No. 10) had four rays, eight (No. 4, 6–9, 12–14) had three rays, and two (No. 11 and 15) had two rays. Most limbs with type IA FH had five-ray feet, which were only detected in type IA FH patients. Stanitski et al.¹³ reported that three-ray feet occurred in none of the type IA or type IB fibulae but 100% of type II fibulae. However, one patient (No. 4) with type IA FH had a three-ray foot in our clinics.

Table 2.

The clinical features of Chinese patients with fibular hemimelia.

Patient No.	Sex	Age	Side	Achterman and Kalamchi classification system	Hypoplasia of the lateral femoral condyle	Anteromedial bowing of the tibia	Ankle morphology	Tarsal coalition	Number of foot rays
1	Male	4	Left	Type IA	Yes	No	Dysplasia of the ankle	Yes	Five
2	Male	11	Left	Type IA	Yes	No	Balland-socket type	Yes	Five
3	Female	21	Right	Type IA	Yes	No	Balland-socket type	Yes	Five
4	Male	23	Right	Type IA	Yes	No	Balland-socket type	Yes	Three
5	Female	23	Left	Type IA	Yes	No	Balland-socket type	Yes	Five but hypoplasia of the fifth ray
6	Male	19	Right	Type IA	Yes	No	Balland-socket type	Yes	Three
7	Female	23	Right	Type IB	Yes	No	Balland-socket type	Yes	Three
8	Female	3	Right	Type II	Yes	Yes	Dysplasia of the ankle	Not fully developed	Three
9	Male	3	Right	Type II	Yes	Yes	Dysplasia of the ankle	Not fully developed	Three
10	Female	4	Left	Type II	Yes	Yes	Dysplasia of the ankle	Yes	Four
11	Male	6	Left	Type II	Yes	No	Dysplasia of the ankle	Yes	Two
12	Male	6	Right	Type II	Yes	Yes	Dysplasia of the ankle	Yes	Three
13	Male	12	Left	Type II	Yes	No	Balland-socket type	Yes	Three
14	Female	47	Right	Type II	Yes	No	Balland-socket type	Yes	Three
15	Female	3	Right	Type II	Yes	No	Dysplasia of the ankle	Not fully developed	Two

Figure 1.

According to the Achterman and Kalamchi classification, (A) represents type IA FH, (B) represents type IB FH, and (C) represents type II FH. Note that all three patients had hypoplasia of the lateral femoral condyle (white arrows) with different degrees of severity.

Figure 2.

(A) Tarsal coalition. (B) Anteromedial bowing of the tibia. Also note that the ankle was poorly developed. (C) Ball-and-socket ankle joint.

Proximal femoral focal deficiency

Proximal femoral focal deficiency (PFFD) is a musculoskeletal malformation in the proximal portion of the femur that is a result of early disturbance of growing mesenchymal homeostasis.^17–19 The first PFFD case, reported in 1974 by Kelly, delineated a deformity of the right leg of an infant.²⁰ The classification of PFFD was described by Torode and Gillespie in 1991.²¹ In the following decades, continuous case reports were published illustrating that the epidemiology of PFFD ranges from 1 in 50,000 live births (1995) to 0.11–0.2/10,000 live births (2015).^17,18 Specifically, the ratio of the incidence of PFFD is greater in males than females (1.5:1); furthermore, PFFD more often presents unilaterally than bilaterally.^17,22 The clinical diagnosis of PFFD has depended on imaging techniques available at the time; these techniques have varied to a great extent, from radiography in the 19th century to emerging ultrasonography in the 20th century. By this progress, the categorizations of PFFD are divided subjectively.^19,23 The Aiken classification system of PFFD has been well accepted. This classification system separates PFFD on the basis of the presence of anatomic features of the femoral head, acetabulum, and femoral segment.^17,18,24 According to the radiographic appearance on plain X-ray, the Aiken classification delineated PFFD into four types (A–D). Type A presents as a short femur with a femoral head and acetabulum; there is a possibility of subtrochanteric and pseudoarthrosis. Type B refers to a short femur with bony tuft; the femoral head is present, and there is a moderately dysplastic acetabulum. Type C is described as a short and tapered femur with an absent/very small femoral head, and the acetabulum is severely dysplastic. The most severe situation is type D, where the femoral head and acetabulum is missing, and the femur is short and deformed.^17,18,24 Other classifications were devised with more groupings or with the aid of advanced diagnostic methods. For example, Amstutz categorized PFFD into five types by dividing type A of the Aiken classification into two subgroups on the basis of the presence of subtrochanteric pseudoarthrosis and coxa vara.¹⁸ Advanced diagnostic methods, such as sonographic technique, improve the detection of these features of PFFD during the prenatal examination.^18,22,25,26

To date, treatments for PFFD have included two methods: surgical and nonsurgical. Nonsurgical treatments include prosthesis fitting around the plantar-flexed foot as assistance.¹⁷ Surgical approaches vary depending on the individual case, including Van Ness rotationplasty, Torode-Gillespie rotationplasty, the knee fusion technique of King, and Syme’s amputation technique.¹⁹ The etiology of PFFD remains unclear. However, some hypotheses have been proposed, such as abnormal neural crest development, defects in chondrocyte maturation, teratogen exposure, such as thalidomide and misoprostol, poorly controlled diabetes, viral infection, radiation exposure, and focal ischemia.^26–29 Epps²⁶ suggested that sclerotome subtraction as a result of injury to the neural crest could be one theory related to limb deformities. The neural crest is responsible for the formation of the precursor of the peripheral sensory nerves for the fourth and fifth lumbar vertebrae. By analyzing a fetus at 21 weeks with unilateral PFFD, Boden et al.²⁷ found that PFFD may be related to unsuccessful proximal growth plate migration, the immature arrangement of proliferative and hypertrophic chondrocytes, abnormal septal architecture truncating the immature hypertrophic zone, and irregular vascular invasion of primary trabecular. Hamanishi³⁰ studied 70 patients with 91 congenital short femora and illustrated that either different etiological factors or different timing of drug insult to the fetus in the thalidomide group resulted in the defects. The thalidomide group had femur–tibia–radius anomalies, while the non-thalidomide group had femur–fibula–ulna anomalies. It was suggested that, in the non-thalidomide group, a multifactorial genetic background is related to hypoplasia of the femur. Marchese et al.²⁹ reported a 5-month-old female orphan with an absence of the proximal femur, evident fibular hypoplasia, and abnormal metatarsals and phalanges and suggested that misoprostol elicited these defects through vascular disruption by uterine contraction during the first trimester of gestation. Lin²² and Mailath-Pokorny²⁵ reported a case in which a 30-year-old woman with over 19 weeks’ gestation had a fetus with an isolated unilateral short femur on ultrasound examination. The fetus presented with a normal karyotype via amniocentesis, suggesting the possibility that epigenetic modification may be involved in the development of PFFD.

Tarsal coalition

TC refers to a pedal anomaly in which fusion of two or more bones of the hindfoot occurs, resulting in absent or restricted movement between bones.^6,31–36 Buffon is considered to be the first to describe TC, in 1769.^31,32,34 Later, Coventry³⁷ published an initial study in 1950 that illustrated the idea of hindfoot abnormality with TC. The reported epidemiology of TC varied from the 19th century to the 20th century. Before the millennium, the incidence of TC is suggested to have been 1–2%,^31,32 while, in 2003, Solomon et al.³⁸ concluded that the incidence of TC could be as high as 12.7% after studying 100 cadaveric feet. The most common coalitions are the talocalcaneal coalition and calcaneal navicular fusion; the coalitions are 50% bilateral and have a slight male preponderance.^6,34 Unlike PFFD patients who come to the clinic for limb correction, TC patients usually present with symptoms, such as vague aching hindfoot pain, subtalar stiffness, peroneal spasm, and sinus tarsi syndrome.^6,32,34 These symptoms are elicited by interruption and restriction of rotation and gliding of subtalar joints, thus leading to a compensation of involved hindfoot and ligaments, which can ultimately result in a peroneal spastic flatfoot or traction spur.^31,34 Initial clinical diagnosis of TC primarily depends on radiographic analysis and nuclear scintigraphy (bone scintigraphy).^33,34 With the advent of three-dimensional imaging techniques (CT and MRI), clear stereoscopic images can be taken to provide better characterization and facilitate diagnosis. Radiographic features, such as talar beaking, C sign, and drunken waiter sign, support the clinical diagnosis of TC.^6,31,32,34

The classification of TC is mainly based on the inheritance of the disease and anatomic defects. Specifically, Tachdjian and Buckholtz divided TC by anatomic occurrence of the coalitions and concluded that medial talocalcaneal coalitions are the most common form of TC, while the second most common are calcaneal navicular coalitions.^6,31 Jacobs, Schlefman, and Sakellariou classified TC on the basis of the type of connective tissue involved with uniting the involved osseous segments.^6,31,32 For example, when the union of the coalition between two or more tarsal bones is attributed to cartilaginous tissue, it is termed cartilaginous/synchondrosis. Perlman and Wertheimer classified TC based on etiology as congenital or acquired disease.³¹ Cowell and Harris classified TC as complete, incomplete, or rudimentary coalition on the basis of the motion at the joint.⁶ Downey and Lawrence categorized TC by taking the status of osseous morphological changes related to the middle talocalcaneal joint as extra-articular– or intra-articular–type coalitions.^31,35

The treatment of TC is more focused on symptomatic relief and is dependent on the degenerative extent.^6,31,32,34 Patients often complain of pain, talar joint sprain, or severe peroneal spasm. The standard nonsurgical treatments are assistant therapies (orthosis, non-steroidal anti-inflammatory medication, peroneal nerve block, cast/immobilization, physical therapy, etc.). Surgical treatments, including resection and arthrodesis, are performed when non-operative treatments are not sufficient.^6,31,34 Zaw et al.³⁴ pointed out a salvage surgery (triple arthrodesis) for the treatment of patients with TC and persistent pain. Regarding the etiology of TC, acquired TC in the adult population may be ascribed to trauma and posttraumatic complications (infection, neoplasia).^31,32,34 The possible etiologies of congenital TC have also been proposed, such as unsuccessful mesenchyme differentiation, delayed ossified accessory ossicles, and autosomal dominant inheritance.^6,31,32,34 Jack and Leboucq both illustrated the idea that TC may be related to the failure of segmentation and differentiation of primitive/embryonic mesenchyme.^6,31,32,34 The delayed ossified accessory ossicles were proposed by Pfitzner in 1896 with the observation that accessory ossicles presented at the location of tarsal bone development.^32,34 Interestingly, Harris found a talocalcaneal coalition in the fetus and confirmed Jack and Leboucq’s hypothesis to some extent; while denying Pfitzner’s hypothesis.^32,34 By analyzing three generations of the calcaneal navicular and talocalcaneal coalitions in the same family, Wray and Herndon suggested an autosomal dominant trait with reduced penetrance.^6,34 Leonard³⁹ concluded that tarsal coalition is a unifactorial disorder of autosomal dominant inheritance with full penetrance, after reviewing 98 first-degree relatives of 31 index patients with confirmed tarsal coalitions. Plotkin⁴⁰ considered the inheritance of calcaneal navicular coalition as part of a complex polygenic system responsible for overall limb development by studying monozygotic twins.

Development of lower limb

The etiology of PALH remains largely unknown. To determine the mechanisms of PALH defects, it is necessary to understand the developmental biology of the limb, in particular, the limb patterning and the genetic network during embryonic skeletal development.

The vertebrate limb develops from a small bud of undifferentiated mesoderm cells encased in ectoderm.⁴¹ The appearance of limb buds, which are small projections of fetal mesenchymal condensation on the ventrolateral aspect of the trunk, occurs during the 4th to 5th weeks of gestation.¹³ The development of the bud has intrinsic polarity and occurs along three different axes: proximal–distal (PD), anterior–posterior (AP), and dorsoventral (DV). The complex patterning process is regulated by three main signaling centers––the apical ectodermal ridge (AER), the zone of polarizing activity (ZPA), and the progress zone (PZ).^42,43 The AER is a structure formed from ectodermal cells at the distal end of the bud; the ZPA is located in the mesenchyme on the posterior rim of the bud; and the PZ is located adjacent to the AER.^42,43 Each signaling center controls patterning in one of the axes of the limb, and they are all essential for development since they provide the growth factors necessary for patterning the limb. The functions of these signaling centers are interdependent, and adequate interactions between them are essential for proper morphological limb development.

The PD axis

The AER is the specialized and thickened ectoderm rimming the distal edge of the limb bud and is required for PD elongation.⁴⁴ The AER is formed through a complex and not completely understood process that starts with the induction of the AER precursor cells, which are marked by their expression of fibroblast growth factor 8 (FGF8).⁴⁵ Limb bud outgrowth and the patterning of mesenchymal progenitors are coordinately regulated by epithelial–mesenchymal interactions between the AER and mesenchyme. Early findings showed that removal of the AER resulted in limb truncation.⁴⁶ Recently, the “two-signal model” was formulated, which proposes that two opposing signals, such as retinoic acid (RA) secreted from the flank mesenchyme and FGF secreted by the AER, pattern the skeletal elements along the PD axis.⁴⁷ This model indicates that AER FGFs play an instructive role in regulating the expression of specific genes and thereby identities along the PD axis.⁴⁷

The AP axis

Patterning along the AP axis is controlled by sonic hedgehog (Shh) in the ZPA.^48,49 Grafting a ZPA to the anterior distal margin of a chicken limb bud causes mirror-image duplications of skeletal elements, and restriction of Shh expression in the ZPA can mimic the effects of ZPA grafts. These findings suggest that Shh mimics the function of ZPA and plays a key role in the AP patterning of the limb bud.^50,51 Shh acts as a morphogen diffusing from the ZPA, establishing a concentration gradient across the digit-forming field.^52,53 Experimental manipulation in chicks and genetic analysis in mice suggested that Shh participates in the early establishment of AP-axis polarity and controls the proliferative expansion of specified mesenchymal progenitors.^54,55 Deficiency of Shh results in ulnar ray deficiency in the upper limbs and fibular ray deficiency in the lower limbs.^56,57 Misexpression of Shh could lead to its ectopic activation at the anterior border of the bud, and this is considered to be at the root of many pre-axial polydactylies.

The expression of Shh needs to be tightly regulated at different levels, since alterations in the strength, duration, or location of Shh expression could have profound phenotypic consequences.⁵⁸ First, normal Shh expression in the ZPA is regulated by an enhancer designated zone of polarizing activity regulatory sequence (ZRS).^59,60 Recently, Williamson et al.⁶¹ showed that Shh and ZRS co-localize to the ZPA, and constrained chromatin configuration optimizes the opportunity for the active enhancer to locate and facilitate the expression of Shh. Point mutations in ZRS can induce anterior, ectopic Shh expression and can cause preaxial polydactyly, triphalangeal thumb, Werner mesomelic syndrome, or tibial hypoplasia.^60,62–67 Duplications and even triplications of the ZRS have been implicated in polysyndactyly disease, including triphalangeal thumb-polysyndactyly syndrome and Haas-type polysyndactyly syndrome.^66,68,69 Second, the activation and posterior restriction of Shh expression in the ZPA are maintained by mutual antagonism between the transcription factors HAND2 and GLI3.⁴¹ Initially, HAND2 and GLI3 are expressed throughout the presumptive limb field, and, during the onset of limb bud outgrowth, direct cross-regulation results in posterior restriction of HAND2 and anterior restriction of GLI3.⁷⁰ This mutually antagonistic interaction results in the activation and posterior restriction of Shh expression in the limb bud mesenchyme.⁷⁰ Recently, Deimling et al.⁷¹ identified a ~ 5-kb deletion within the Shh repressor GLI3 in two patients with bilateral tibial hemimelia. This deletion results in a truncated GLI3 protein that failed to repress anterior ectopic Shh expression.⁷¹ Third, in the established limb bud, the Shh/Gremlin1 (GREM1)/FGF feedback loop coordinates Shh signaling by the mesenchymal ZPA with FGF signaling by AER.⁷² Shh maintains FGF expression in the AER through the upregulation of bone morphogenetic protein (BMP)⁷³ and GREM1,^74,75 and FGFs produced by the AER maintain Shh expression in the ZPA.^76,77 Finally, Wnt7a secreted by the dorsal ectoderm is required for Shh expression,^78,79 which is discussed below.

The DV axis

The DV axis determines the development of dorsal (e.g., the nails and the thin hairy skin) and ventral (e.g., the thick, hairless skin of the palm and sole) structures in the hands and feet.⁵⁶ The development of ventral structures is regulated by engrailed-1 (En1) (present in the ventral ectoderm), and the development of dorsal structures is controlled by WNT7a and LMX1b (present in the dorsal ectoderm and mesoderm, respectively) (80, 81). The expression of WNT7a is limited to the dorsal ectoderm, and it is repressed by En1 in the ventral ectoderm. Inactivation of En1 in mice results in double-dorsal limbs, probably because of the successive misexpression of WNT7a in the ventral ectoderm.⁸² WNT7a has two functions: it defines the DV axis by inducing LMX1b via the noncanonical Ca2⁺-mediated pathway,^83,84 which is responsible for the development of dorsal structures. WNT7a also maintains Shh activity in the ZPA⁷⁹ via the canonical β-catenin pathway,⁸⁵ which is responsible for the development of the ulnar/fibular ray. Mutations in LMX1B result in nail–patella syndrome, a human condition characterized by loss of dorsal features,⁸⁶ indicating that LMX1B participates in DV patterning. Mice lacking WNT7a activity exhibit loss of dorsal features and lack posterior digits, demonstrating that WNT7a is required for both DV and AP patterning.⁷⁸

Potential signaling involvement in PALH

PALH is likely caused by a disturbance of the AP axis that determines the development of ulnoradial structures in the upper limbs and fibula tibial structures in the lower limbs. Shh is considered the essential controller of this axis. Indeed, as discussed previously, deficiency of Shh results in ulnar ray deficiency in the upper limbs and fibular ray deficiency in the lower limbs.^56,57 However, it is not known how Shh and the regulatory factors in the AP axis act in the pathogenesis of PALH. Moreover, it is important to understand that the three axes are not independent and that they interact with each other. Therefore, a review of syndromes with fibular aplasia/hypoplasia can provide additional clues to the pathogenesis of PALH.

Al-Awadi, Raas-Rothschild, Schinzel phocomelia (AARRS) and Fuhrmann syndromes were considered to be distinct limb-malformation disorders. Fuhrmann syndrome is characterized by bowing of the femurs, aplasia or hypoplasia of the fibula, and poly-, syn-, and oligodactyly.⁸⁷ In contrast, AARRS is much more severe and characterized by severe malformations of upper and lower limbs (including hypoplastic femora and absent ulnae and fibulae) with severely hypoplastic pelvis and abnormal genitalia.⁸⁸ Human genetic studies have suggested that the both diseases are caused by mutations in the same gene––WNT7A; a partial loss of WNT7a function caused Fuhrmann syndrome, whereas the more severe limb truncation phenotypes were observed in AARRS, which resulted from functionally null mutations of WNT7A.⁸⁹ All patients with WNT7A mutations have evidence of posterior hypoplasia in addition to the loss of dorsal features in the limbs, and this is explained by WNT7a effects on Shh via β-catenin signaling. The findings also implicate the importance of β-catenin signaling pathway in the pathogenesis of PALH.

Du Pan syndrome, characterized by a combination of brachydactyly and bilateral fibular hypo/aplasia, is the disease with growth differentiation factor 5 gene (GDF5) mutations.⁹⁰ GDF5 belongs to the superfamily of BMPs binding with high affinity to the BMP receptor 1B (BMPR-1B).^91,92 Affected individuals are homozygotes or compound heterozygotes for loss-of-function mutations in the highly conserved mature domain of GDF5, which results in an impaired BMP signaling through the BMPR-1B.⁹⁰ Szczaluba et al.⁹³ identified heterozygous mutations in GDF5 in a family with dominant Du Pan syndrome. A possible synergistic effect of the cis-acting mutations located in the active domain of the mature GDF5 protein is likely to be responsible for the clinical expression of the disorder. Recently, Stange et al.⁹⁴ identified and characterized a hypomorphic BMPR-1B missense mutation that causes Du Pan syndrome. Several BMPs could interact with Shh signaling and have a weak polarizing activity to enhance the polarizing effect of a brief Shh application.^52,95 These findings indicate that BMP signaling also plays a role in the pathogenesis of PALH.

Waardenburg anophthalmia syndrome is characterized by ocular anomalies ranging from mild microphthalmia to anophthalmia and limb defects.⁹⁶ The distal abnormalities of the limbs include the absence of the fifth metatarsals and toes, oligodactyly, syndactyly, synostosis of the fourth and fifth metacarpals, and hypoplasia of the fibula with short femur or tibia.⁹⁶ These defects could be caused by homozygous mutation in the SMOC1 gene.^97–99 Although the molecular mechanism for this is unknown, SMOC1 has been shown to function as a BMP antagonist.¹⁰⁰ These findings suggest that proper BMP antagonism is essential for postaxial development at the embryonic stage.

Summary

In summary, PALH, which includes FH, PFFD, and TC, is a group of nonhereditary diseases with congenital lower limb deficiency. The etiology of PALH remains unknown, although it appears to be related to developmental defects that occur during the embryonic period. We reviewed the previous classification systems, presented the clinical features, and discussed the potential signaling involvement in PALH.