By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
During vertebrate embryogenesis, blood vessels and nerves have been observed to develop within close proximity to one another, often exhibiting congruent branching patterns. This is evident within developing limbs, where both deep and cutaneous neurovascular congruencies have been observed (Martin and Lewis,1989; Taylor et al.,2001; Mukouyama et al.,2002; Bates et al.,2002,2003). Within areas such as the skin and the musculature, major branches from both systems undergo extensive ramifications. Both systems also avoid similar regions, such as chondrogenic mesenchyme and the avascular zone at the periphery of the limb bud (Martin and Lewis,1989). Recent findings point to a similarity extending beyond anatomical pattern to that of shared molecular guidances molecules (Kitsukawa et al.,1995; Eberhart et al.,2000; Oosthuyse et el.,2001; Mukouyama et al.,2002,2005; Bates et al.,2003; Vieira et al.,2007; Zacchigna et al.,2008). Studies have documented the neurovascular pattern of the quail and chick forelimb, and numerous studies have detailed the motor and sensory pattern of the chick hindlimb, however the neurovascular pattern of the quail hindlimb has yet to documented (Landmesser,1978; Drushel et al.,1985; Tyrrell et al.,1990; Brand-Saberi et al.,1995; Wang and Scott,1999,2000; Taylor et al.,2001; Bates et al., 2002).
An understanding of the neurovascular anatomy of the Japanese quail (Coturnix coturnix japonica) hindlimb is of value to future studies of the molecular regulation of nerve and blood vessel development. The quail hindlimb is a highly amenable structure for the study of the mechanisms governing neurovascular patterning and the avian embryo has been the choice system for study of limb vascular development for close to 100 years, with the dominant focus being on the wing (Evans, 1912; Caplan,1985; Drushel et al.,1985; Sabin,2002 (1917 reprint)). The chick hindlimb has been used as the primary system for the study of sensory and motor neuron development, whereas the chick and quail do share many anatomical similarities during development, some differences in timing do exist (Landmesser,1978; Hollyday,1980a; Tosney and Landmesser,1985; Tanaka and Landmesser,1986a,b). The quail embryo develops more rapidly than the chick with a total incubation period of ∼18 days versus the ∼21 days of the chick. The major difference between the innervation of the wing versus the leg is that eight rather than five dermamyotomes contribute to muscles of the hindlimb (Tosney,1988). It has been reported that the spatial organization and motoneuron innervation pattern of hindlimb muscles are homologous between the chick and quail, however, quail limbs are much smaller than that of the chick (Tanaka and Landmesser,1986a,b).
Because peripheral nerves and blood vessels often follow parallel routes during their development, shared guidance mechanisms are suggested (Carmeliet,2003). Recent studies have used the chick and quail hindlimb as a model system for experiments aiming to determine the effect of various key signaling molecules on motor neuron and blood vessel formation (Eberhart et al.,2000,2002; Bates et al.,2003). Motoneuron growth into the developing limb is a highly specific process with axons using tip growth and branching out into the muscle mass region corresponding to the adult projection pattern (Lance-Jones and Landmesser,1980a,b). The proper selection of motoneuron pathways into the limb has been shown to be directed by specific local, as well as distal, cues within the developing limb bud (Lance-Jones and Landmesser,1980a,b,1981; Tosney and Landmesser,1984; Honig et al.,1998). Evidence suggests that sensory neurons respond to both the distal molecular cues as well as to the local position of motor neurons for proper pathway guidance into the chick hindlimb (Honig et al.,1986;1998). To better understand the molecular mechanisms regulating crosstalk between blood vessels and nerves, an understanding of normal neurovascular anatomy must exist. The present study uses whole mount immunohistochemistry and fluorescent lectin injection, along with laser scanning confocal microscopy and epiflourescent microscopy, to describe the E2.75 – E6.0 neurovascular anatomy of the Japanese quail hindlimb. It is the aim of this study to provide a model for further investigation of the development of major blood vessels and nerves in the Japanese quail hindlimb.
MATERIALS AND METHODS
Microinjection of Quail Embryos
Japanese quail embryos (Coturnix coturnix japonica) were obtained from Cornell University Poultry Science Department (Ithaca, NY) and incubated in a forced draft, humidified incubator at 38°C to between stages 17 and 29 (Hamburger and Hamilton,1992). Embryos were removed from the eggshell into a glass petridish containing 38°C phosphate-buffered saline (PBS). Extraembryonic membranes were removed and the chest wall was carefully incised using Dumont forceps to expose the heart. Warm rhodamine Lens culinaris agglutinin solution (R-LCA, 5 mg/mL; Vector Laboratories, Burlingame, CA) was slowly injected into the heart at a concentration of 1:30 in PBS through glass micropipettes (50 μL; Drummond Scientific, Broomall, PA) hand-pulled over a flame to a diameter of 5–20 μm. Micropipettes were connected through flexible rubber tubing to a mouthpiece. Approximately 1–5 μL of R-LCA solution was injected into the vascular system of each embryo, depending on developmental stage. R-LCA was used in place of FITC-Dextran based on the enhanced binding properties to the endothelial cell lumenal surface of R-LCA (Jilani et al.,2003).
Following injection, the R-LCA was allowed to circulate throughout the embryo for ∼2 min while submerged in 38°C PBS. A complete transverse incision through the embryo was then made between the forelimb bud and the hindlimb bud. The remaining caudal portion of the injected embryo was then eviscerated and the most distal portion of the tail was removed to aide in mounting and visualization of the ventral hindlimb region. India ink (Pelikan No. 17 Black) injection was also used to elucidate details of the vascular anatomy, although the data are not shown here. India Ink was diluted in a 1:1 ratio with PBS, centrifuged at 4,000 rpm for 1 min and the supernatant was collected for microinjection into the heart.
Following R-LCA injection and evisceration, hindlimb segments were fixed overnight in 10% formalin at 4°C. Hindlimb segments were then washed 3 × 30 min in PBS, permeabilized in ice-cold methanol (30 min, 2 hr, 30 min), and rehydrated through graded ethanol (10 min × 90, 70, 50, 30%). Rehydration was followed by 2 × 15 min in PBT (PBS with 0.1% Triton X-100) and a 30 min rinse in 0.1% NaBH4 (Fisher Scientific, Pittsburgh, PA) at RT to quench endogenous fluorescence. Hindlimb segments were blocked in PBT containing 3% bovine serum albumin (BSA; Fraction V powder; Sigma, St. Louis, MO) for 30 min. The TUJ1 (anti-neuronal class III β-tubulin, mouse IgG2a; Covance, Berkeley, CA) antibody was then added at a dilution of 1/250 in 3% BSA/PBT and left overnight at 4°C on an orbital shaker. The hindlimb segments were then washed 2 × 30 min in PBT followed again by blocking in 3% BSA/PBT for 30 min after which a cy2 AffiniPure F(ab′)2 fragment goat anti-mouse (H + L; Jackson ImmunoResearch Laboratories, West Grove, PA) was added at 1:100 and allowed to incubate overnight at 4°C on an orbital shaker. The secondary antibody was then removed and the hindlimb segments were washed for 6 hr in PBS followed by graded dehydration through ethanol and a 2 min clearance in toluene. Hindlimb segments were then mounted on glass slides in Entellan (cat. no. 14800, EM Sciences, Fort Washington, PA). A minimum of 20 embryos at each HH stage between 17 and 30 were analyzed for this study and the vascular and nervous anatomy was consistent among all embryos imaged.
Embryonic whole mounts were imaged with a 1024ES laser scanning confocal microscopy system (BioRad, Hercules, CA) mounted on a Nikon Eclipse E600 fluorescent microscope (Nikon, Melville, NY). Images were collected as a z-series consisting of 30–60 sections at 5–10 μm intervals. A one-dimensional image of the captured z-series was produced using LaserSharp 2000™ (Zeiss) and Confocal Assistant software (BioRad) and formatted for publication with Adobe Photoshop 7.0. Epifluorescent images were captured using an Olympus C-5050 ZOOM digital camera mounted on a Leitz fluorescent microscope.
Embryonic Day 2.75 (HH18)
Primary vascularization of the hindlimb begins at HH stage 18 (65–69 hr) when dorsal segmental vessels located at, and posterior to, the twenty-seventh interspace, along with capillaries present in the hindlimb bud, anastomose with lateral offshoots from the dorsal aorta (DA; Fig. 1A). By Day 2.75 (HH18), the dorsal aortae are just beginning to fuse at the level of the hindlimb bud (Fig. 1A). Intersomitic arteries (ISAs) branch off the DA to form a plexus lateral to the neural tube, known as the perineural plexus. Motor neurite outgrowth from the ventral neural tube, which proceeds in a rostro-caudal graded manner (Spence and Poole,1994), and has yet to occur at the lumbosacral (LS) level by E2.75 (HH18). Likewise, dorsal root ganglia (DRG) have yet to form at this level, although initial neural crest cell migration into the rostral sclerotome of the anterior most somite is evident by HH17-18 (32 somite stage; Spence and Poole,1994).
Embryonic Day 3 (HH20)
By E3 (HH20) the dorsal aortae have become completely fused at the level of the hindlimb (Fig. 1B). The lateral offshoots from the dorsal aorta into the hindlimb capillary plexus have begun to regress (Fig. 1B). The ischiadic artery (IA) has formed at approximately the level of somites 31–33 (LS5-6) from anastomoses between several lateral offshoots of the DA and the capillary plexus within the hindlimb bud (Fig. 1B). The lateral offshoots from the DA and the ISAs still provide some arterial circulation to the hindlimb bud at this stage of development. The more anterior crural artery (CrA) is not present at E3 (HH20). Venous circulation within the E3 (HH20) hindlimb is provided by the marginal vein (MV), which is located at the periphery of the limb bud just below the avascular zone and overlying ectoderm. Although present, the MV is not distinctly visible in either Fig. 1A,B, but can be seen clearly in Fig. 3A, and is illustrated in Fig. 4A. The posterior cardinal vein (PCV) is present at the level of the hindlimb by E3 (HH20; Fig. 1B), and is located dorsal and lateral to the dorsal aorta. The ISAs drain into the intersomitic veins, which in turn drain into the PCV, along with the MV.
E3 (HH20) motor neurites exit the ventral neural tube in a rostro-caudal manner (Spence and Poole,1994), passing ventrolaterally through the anterior portion of the adjacent sclerotome (Fig. 2A). A pattern of congruency may be argued between the intersomitic vessels and the ventrolaterally migrating motor neurites, given that the two develop in such close proximity to one another. However, because motor and sensory neurites will only grow through the anterior portion of the somite (Keynes and Stern,1984), it seems unlikely that intersomitic vessels, which are present prior to motor and sensory nerve ingrowth, directly influence the position of developing nerve pathways at this stage of hindlimb development.
Embryonic Day 3.75 (HH23)
By E3.75 (HH23), the IA is observed coursing laterally through the sciatic foramen of the pelvic girdle precursor (PGP) into the base of the hindlimb bud. Beyond the base of the hindlimb the IA becomes the sciatic artery (ScA), a rather large and centrally located vessel. The ScA is the primary artery of the hindlimb bud at this stage. The MV system provides venous return to the PCV.
By E3.75 (HH23) motor and sensory neurites have begun to fill the plexus region in a position suitable to invade the limb by HH stage 24–25 (Fig. 2B). In agreement with the finding of others, we found that once axons reach the plexus region, there is a delay in neurite outgrowth into the hindlimb that lasts for ∼24 hr (Tosney and Landmesser,1985; Wang and Scott,1999,2000). During this lag period, there is a great deal of axon sorting in the plexus region and growth cones continue to be projected from the spinal cord through the lateral mesoderm to the plexus region (Lance-Jones and Landmesser,1981). It is during stage 24 and early 25 that two distinct plexus regions are formed along the anterior–posterior (A–P) axis at the base of the hindlimb situated between the myotome, PGP, endoderm, and PCV (Fig. 2B; Tosney and Landmesser,1985).
Embryonic Day 4.0 (HH24)
At E4.0 (HH24), the ScA remains the central artery within the hindlimb, and is positioned ventral to the femur anlage. The CrA is visible by E4.0 as an assemblage of capillaries lateral to the aorta at the somitic level 29–31 (LS 2–4). The CrA enters the hindlimb through the anterior-most foramen in the pelvic anlagen along with the crural nerve bundle (also referred to as the femoral nerve) and provides blood supply to the anterior thigh region. The MV returns blood to the PCV and remains the primary venous system for the E4.0 hindlimb.
The present study found that eight spinal segments, as opposed to seven as reported by Tanaka and Landmesser (1986a) and Tyrrell et al. (1990), contribute to the innervation of the quail hindlimb. During E4.0 (HH24) LS spinal nerves 1–8 continue to extend axons in distinct bundles towards the plexus region at the base of the hindlimb (Fig. 2B). The crural and sciatic plexuses are not yet fully populated, as much axon sorting within the plexus region still must take place (Lance-Jones and Landmesser,1981). However, a small number of neurites from both plexuses are beginning to project ventrolaterally into the pre-muscle mesoderm of the hindlimb (Fig. 2B).
Embryonic Day 4.5 (HH25)
By E4.5 (HH25) the ScA has developed to a point dorsal to the knee joint distal to which it becomes the posterior tibial artery (PTA; Fig. 3A). As the PTA develops distally it is positioned directly between the tibial and fibular anlagen. Near the distal end of the tibial and fibular anlagen the PTA is beginning to bifurcate into a larger dorsal artery and a smaller ventral artery that will both supply the autopodal region of the hindlimb (Fig. 3A). The anterior and posterior MVs located just below the peripheral avascular zone continue to provide venous circulation to the PCV (Fig. 3A).
By E4.5 (HH25) the crural and sciatic plexuses are structurally well defined. LS 1–4 contribute to the crural plexus and LS 4–8 contribute to the larger sciatic plexus. Distinct projections from both plexuses into the hindlimb are evident at this stage. From the crural plexus a dorsal and a ventral nerve branch are projected into the pre-muscle mesoderm of the anterior thigh (Fig. 3B,C). Likewise, two distinct axonal branches have emerged from the sciatic plexus, a dorsal and a ventral branch. The dorsal branch, the sciatic nerve bundle, is broad and more extensively populated than the ventral branch, also referred to as the tibial nerve (Fig. 3B,C).
Embryonic Day 4.75 (HH26)
By E4.75 (HH26) the ScA and CA and PTA remain the prominent blood vessels in the hindlimb (Fig. 3D). The digital arteries are forming anastomoses with the dorsal and ventral branches of the PTA and are beginning to regress from the developing avascular spaces that will become the metatarsals. The once dense capillary network of the pre-muscle mesenchyme is being actively remodeled as muscle tissue is being formed.
By E4.75 (HH26) the dorsal and ventral nerves branches that emerged from the sciatic plexus have extended distally to approximately the mid-tibial and fibular region (Fig. 3D). Above the level of the knee, the dorsal branch is referred to as the sciatic nerve bundle. At the level of the knee joint the sciatic nerve bundle bifurcates into the medial and lateral branches of the peroneal nerve (Fig. 3D), which will innervate the muscles of the lateral and dorsal compartments of the shank as well as the muscles of the foot. The ventral bundle, or tibial nerve, is only faintly visible in Fig. 3D. The tibial nerve bifurcates into a dorsal and a ventral branch at approximately the same level within the hindlimb as the bifurcation of the sciatic nerve bundle. The dorsal and ventral branches of the crural nerve continue to project towards their respective muscle targets. The lateral femoral cutaneous nerve is also formed by E4.75 (HH26; Fig. 3D).
Embryonic Day 5.5 (HH28)
By E5.5 (HH28) the anterior tibial artery has formed just distal to the knee joint as a ventro–distal offshoot of the PTA (Fig. 3E and Fig. 5). The anterior tibial artery supplies the ventrolateral portion of the distal hindlimb. The digital arteries, which first formed anastomoses with the PTA, have now formed proximo–ventral offshoots that have assembled into a loose arteriolar network near the ventral surface of the shank. The crural artery (CA), which is largely responsible for vascularization of the anterior thigh musculature, has bifurcated at approximately the mid-femur level into dorsal and ventral branches (Fig. 3E). The dorsal branch extends to the level of the dorsum of the knee where it joins the PTA. The ventral branch of the CA supplies the ventral thigh musculature.
By E5.5 (HH28), the lateral branch of the peroneal nerve has extended to the proximal metatarsal border where it branches medially while defasciculating to form the digital nerves (Fig. 3E). The medial branch of the peroneal nerve has extended medially and distally near to the distal end of the tibia. The dorsal and ventral branches of the tibial nerve have extended distally just proximal to the distal end of the tibia. The dorsal branch of the tibial nerve has begun sprouting four axonal branches laterally into the musculature of the medial shank.
Embryonic Day 6.0 (HH29)
By E6.0 (HH29) the digital arteries have fully developed around each metatarsal and anastomosed with the MV at the distal periphery of the hindlimb (Fig. 3F). As the pre-osseous regions of the hindlimb continue to grow, so do the avascular areas they represent. Anterior tibial artery has formed anastomoses with the proximal digital arterial network (Fig. 3F).
By E6.0 (HH29), the medial branch of the peroneal nerve has extended to proximal metatarsal border where it joins with axons of the lateral peroneal nerve (LPN) and has begun to innervate the first and second digits. The LPN has begun innervating the third and fourth digits. The four lateral nerve projections off the dorsal tibial nerve have projected to just below the dermis of the medial shank (Fig. 3F).
Proper development of the vascular and nervous systems within the limb depends on a sequential series of events that decide whether a forming blood vessel or growing axon will be attracted to, or repulsed from, distinct areas (Carmeliet,2003). The segregation of presumptive endothelial cells (PECs; angioblasts) from the mesoderm and the initial assembly of these PECs into a primitive vascular plexus occur through a process termed vasculogenesis (Coffin and Poole,1988; Risau and Flamme,1995). In the limb, the final vessel branching pattern is determined by a combination of angiogenesis and arteriogenesis (Brand-Saberi et al.,1995; Mukouyama et al.,2002). During limb development, blood vessels and axons follow the guidance of dynamic and spatially restricted molecular expresson patterns during embryogenesis (Iwamasa et al.,1999). Both systems invade the limb bud by similar mechanisms involving a growth cone in developing nerves, and a growth cone-like, specialized endothelial cell (termed “tip cell”) in developing blood vessels. These leading edge structures function similarly by protruding filopodia and exhibiting elevated protease activity at the leading tips (Martin and Lewis,1989; reviewed in Carmeliet,2003). In addition, peripheral nerves and blood vessels often follow parallel routes during their development, which suggests a potential shared guidance mechanism (Adams, 2002). Previous studies have focused on the quail forelimb and the chick hindlimb for studying vascular and nervous development, (Tosney and Landmesser,1985; Bates et al.,2002,2003), and heretofore, no studies have examined the combined nervous and vascular anatomy of the embryonic quail hindlimb. In this study, we have analyzed the pattern of neurovascular development in the Japanese quail hindlimb between E2.75 (HH18) and E6.0 (HH29) using whole-mount immunohistochemistry, intravital labeling, and laser scanning confocal microscopy as well as epifluorescent microscopy.
Our findings suggest that the only area where A–P neurovascular congruency was evident was between the LS neurites and the intersomitic vessels as they developed towards the plexus region of the hindlimb (Figs. 2B, 3C). Lateral to the PGP however, the large blood vessels and nerves of the hindlimb did not exhibit an intimate spatial relationship to one another in the A-P plane. Patterns of cutaneous blood vessels and nerves were not analyzed in the current investigation. In their study of neurovascular congruency in the Japanese quail forelimb, Bates et al. (2002) note that the spatial relationship between blood vessels and nerves tends not to be one-on-one but rather much more subtle. The permissive corridors within the vertebrate limb are defined by restrictive tissues, such as cartilage (bone anlagen) and the peripheral avascular zone. In the quail forelimb, Bates et al. (2002) identified four distinct capillary layers in the dorsal–ventral (D–V) plane that are formed prior to axonal ingrowth. Not surprisingly, blood vessels and nerves were found to be invariably restricted to the same D–V planes (Bates et al.,2002). With regard to the A-P and proximal–distal (P–D) axes, Bates et al. (2002) proposed a hierarchy of neurovascular congruence as D–V > A–P > P–D, with the greatest degree of congruence near the geometric center of the limb. On the basis of the findings of the present study however, a specific relationship between blood vessels and peripheral nerves resulting in strict spatial congruency in the quail hindlimb is not confirmed.
Neurovascular congruence in the chick hindlimb was once considered to be largely coincidental (Tosney and Landmesser,1985).With regard to the avian hindlimb, once the crural and sciatic nerve trunks and arteries both pass through the same foramen in the PGP, their previously congruent pathways diverge. It has been clear for at least two decades that nerve bundles in the limb avoid areas of presumptive cartilage, such as the PGP. There is even evidence to suggest a predetermination among certain lateral motor column (LMC) neurons of the LS region with regard to individual muscle targets (Lance-Jones and Dias,1991). With regard to the large nerves and blood vessels deep within the hindlimb, it is less frequent to observe strict patterns of neurovascular congruence. Where neurovascular congruence has been more frequently observed is in the more superficial layers of the hindlimb, such as the skin (Martin and Lewis,1989; Mukouyama et al.,2002). Bates et al., (2003) notes that cutaneous neurovascular congruence is only seen along the D–V axis, which is seemingly contrary to imaging data obtained from mouse skin (Mukouyama et al.,2002).
Vascular Development of the Japanese Quail Hindlimb
The mature quail hindlimb is vascularized by two main vessels, the anterior positioned CA and the posterior positioned IA. The IA is much larger than the CA and is considered the main artery of the avian hindlimb. The crural and ischiadic arteries extend into the hindlimb as direct extensions of the DA (Romanoff, 1960). The IA in the quail hindlimb is evident by E3 (HH20; Fig. 1B). The CA is evident by E4 (HH24; Fig. 3A). In both quail and chick embryos, the umbilical arteries (UAs) shift away from their ventral position, and the proximal portion of the IA serves as the point of origin for the UAs (Evans, 1912). In the chick embryo the two dorsal aortae become fused at the level of the posterior hindlimb (somite level 27) by the end of E4.0 (HH24), and are completely fused in the caudal region by E5. In the Japanese quail embryo however, development proceeds at a slightly more rapid pace. At the level of the hindlimb in the quail, the dorsal aortae have fused into a single vessel by E3.0 (HH20), located at the midline of the embryo, just ventral to the notochord (Fig. 1B).
Initial development of the venous system in the hindlimbs of humans, other mammals, and in the chick involves small dorsal and ventral vessels providing drainage of the capillary plexus directly into the PCV (dorsal) and umbilical vein (ventral). Soon a distinct border vein (i.e., MV) develops with a prominent caudal (fibular) portion and a less prominent cranial (tibial) portion. Within the E3.0 (HH 20) hindlimb bud, venous circulation is provided by the MV at the limb periphery (Arey, 1962; Evans, 1912). At stage 23 (E3.5–4.0) the tibial MV still persists and drains into the umbilical vein and the fibular MV empties into the PCV (Evans, 1912; Romanoff, 1960). Evans (1912) and Arey (1962) note that the tibial portion of the MV will undergo near complete atrophy while the fibular portion of MV persists and becomes a permanent vessel. During E2.75 (HH18) – E6.0 (HH29), the two main drainage portals for the hindlimb are the umbilical veins and the PCVs.
An avascular zone of mesoderm separates the MV from the overlying ectoderm (Feinberg and Noden,1991). The MV is continuously being remodeled in a distal direction and the initial formation of the MV is dependent on the presence of the apical ectodermal ridge (Feinberg and Noden,1991). Romanoff (1960) reported that by E3 (HH20) of chick development, the PCV remains anterior to the dorsal segmental vessels and the hindlimb capillary plexus. However, a significant finding from the present study is that in the Japanese quail embryo the PCV has developed caudally to the level of the hindlimb plexus (somitic level 28–36) by E3.0 (HH20; Fig. 1B).
Nervous Development of the Japanese Quail Hindlimb
At the level of the hindlimb, motor axons of the LMC exit the neural tube beginning at E2.75 (HH18) through the ventral roots and extend ventrolaterally through the anterior portion of the adjacent sclerotome into the somatopleure at the base of the hindlimb (Hollyday1980a; Keynes and Stern,1984; Tosney,1988; Tanaka et al.,1989; Wang and Scott,2000; Turney et al.,2003). Neural crest cells are present with the dorsal portion of the anterior sclerotome at E2.75 (HH18; Spence and Poole,1994). Cell migration, into and through the somite, has been shown to be controlled by chemorepulsive signals from within the somite (reviewed in Carmeliet,2003; Carmeliet and Tessier-Lavigne,2005). Within half a stage, motor neurites that have exited the neural tube join with the, now formed, DRG near the ventrolateral somitic border (Tosney and Landmesser,1985). Our imaging data suggest that the quail LS plexus receives innervation from eight LS spinal levels, as opposed to the seven reported by others (Fig. 3B,C; Tanaka and Landmesser,1986b; Tyrrell et al.,1990). Within the LMC itself, cell bodies in the medial portion of the LMC provide innervation to the ventral portion of the hindlimb while cell bodies in the lateral portion of the LMC supply the dorsal portion of the hindlimb (Tosney and Landmesser,1985; Tyrrell et al.,1990; Tosney et al., 1995; Turney et al.,2003). By E3.5 (HH21), some motor neurites have reached the hindlimb plexus region where they undergo extensive axon sorting with axons from neighboring spinal cord segments (Tosney and Landmesser,1985; Tanaka et al.,1989; Wang and Scott,1999). Axons of the DRG arise through the ventrolateral migration of cells from the neural crest and come to occupy a position corresponding to the rostral half of the particular somitic level (Teillet et al.,1987). Sensory axons extend out from the DRG by HH21 and reach the plexus region by HH22 (Wang and Scott,2000). Like motor axons, sensory axons from the DRG also wait for ∼24 hr (stages 21–24) at the plexus region and emerge by late stage 24 (E4.0), and neither is dependent on the presence or absence of the other for the proper timing and pattern of innervation (Wang and Scott,2000). This waiting period is the same for both the quail and the chick.
During E4.0 (HH24) and early E4.5 (HH25) two distinct plexus regions, crural and sciatic, are formed along the A–P axis at the base of the hindlimb situated between the myotome, PGP, endoderm, and PCV (Fig. 2B; Tosney and Landmesser,1985). Tosney and Landmesser (1985) note that, in the chick hindlimb, growth cones do not emerge from the plexus region until stage 24.5, ∼1 day after the first growth cones from the spinal cord reach the plexus region. Our data suggest that, in the quail hindlimb, the exit of growth cones from the plexus begins at a slightly earlier stage (E3.75 or HH23) than that of the chick (Fig. 2B). Hollyday (1980b) noted that, in the chick embryo, by HH23 – 24 motor neurites have reached the base of the limb, by HH25 have reached the future knee, and by HH28 – 29 (E7) have established an identifiable adult innervation pattern. It is during this early period of dorsal and ventral premuscle mass condensation and differentiation into their respective leg muscles that the ultimate nerve distribution pattern is formed (Hollyday,1980b).
The present study indicates that, in the Japanese quail embryo, the crural plexus receives innervation from LS 1, 2, 3, and a portion of LS 4 (Fig. 3B,C). In the embryonic chick hindlimb, the crural plexus is innervated by LS 1, 2, and 3. In the chick embryo, the sciatic (or ischiadic) plexus is innervated by part of LS 3 and LS 4–8 (Landmesser and Morris,1975). In the Japanese quail embryo, our data suggest that the sciatic plexus is innervated by LS 4–8. However, because the present study did not record muscle nerve stimulation back to specific spinal nerves, it is possible that LS 3 may contribute a small number of axons to the sciatic plexus.
Neurites from the crural plexus supply the preaxial thigh musculature exclusively (Landmesser and Morris,1975; Tosney and Landmesser,1984). Specific hindlimb muscles innervated by neurites from the crural plexus are the following: sartorius, femorotibialis, iliotibialis (anterior head), cutaneous, and adductor (Landmesser and Morris,1975; Lance-Jones,1988). The ventral crural nerve trunk or obturator nerve, which innervates the sartorious and femorotibialis muscles, was not able to be seen in any of the present study's images. The development of the obturator nerve has been well established by earlier work (Lance-Jones and Landmesser,1981) in the chick embryo, however, based upon our findings, further research is needed characterize the development of the obturator nerve in the quail embryo.
Neurites from the sciatic plexus innervate a portion of the postaxial thigh musculature. These neurites form the large sciatic nerve, and are responsible for the sole innervation of the lower leg and foot (Landmesser and Morris,1975; Tosney and Landmesser,1984). Specific hindlimb muscles innervated by neurites from the sciatic plexus are the following: iliotibialis (posterior head), peroneus, digit flexors/extensors, gastrocnemius, iliofibularis (biceps femoris), and flexorius (semimembranosus and semitendinosus; Landmesser and Morris;1975).
Our imaging data suggest that eight (LS 1–8), rather than seven, LS spinal nerve segments contribute to the innervation of the Japanese quail hindlimb. Motor neurites from LS 1–4 are seen innervating the crural plexus, and motor neurites from LS 4–8 are seen innervating the sciatic plexus. While axons of the LS plexus do pause for ∼24 hr before entering the limb, at least in the chick and quail, a small portion of these axons in the quail do begin exiting the crural and sciatic plexuses and entering the hindlimb by E3.75 (HH23), as opposed to ∼E4.5 (HH25) in the chick. The present study also demonstrates the presence of the PCV at the level of the quail hindlimb plexus by E3 (HH20), unlike the later formation of the PCV at the hindlimb level previously demonstrated in the chick (Romanoff, 1960).
The authors wish to thank Kit Hefner MS, CMI for the illustrations.