PAX7 is required for patterning the esophageal musculature

Background The mammalian esophageal musculature is unique in that it makes a transition from smooth to skeletal muscle, with most of this process occurring after birth. In order to better understand the mechanisms that control esophageal musculature development, we investigated the roles in this process of the paired box transcription factor, PAX7, a principal regulator of skeletal myogenic progenitor cells. Previous studies showed that Pax7 is important for determining the esophageal muscle composition. Results We characterized the postnatal development of the esophageal musculature in Pax7−/− mice by analyzing morphology, muscle composition, and the expression of markers of myogenesis, cell proliferation, and apoptosis. Pax7−/− mice displayed megaesophagus with a severe defect in the postnatal developmental process whereby esophageal smooth muscle is replaced by skeletal muscle. Pax7−/− esophagi have substantially reduced skeletal muscle, most likely due to diminished proliferation and premature differentiation of skeletal muscle precursor cells. This impaired the proximal-to-distal progression of skeletal myogenesis and indirectly affected the patterning of the smooth muscle-containing portion of the esophageal musculature. Conclusions Postnatal patterning of the esophageal musculature appears to require robust, PAX7-dependent cell proliferation to drive the proximal-to-distal progression of skeletal myogenesis. This process in turn influences distal smooth muscle morphogenesis and development of the mature pattern of the esophageal musculature. Electronic supplementary material The online version of this article (doi:10.1186/s13395-015-0068-0) contains supplementary material, which is available to authorized users.


Background
The musculature of the esophagus controls passage of food into the stomach by waves of peristaltic contractions. The lower esophageal sphincter (LES) is a bundle of smooth muscles at the distal end of the esophagus, where it meets the stomach. During swallowing, tonic smooth muscles of the LES relax briefly in order to allow passage of food into the stomach [1]. Esophageal muscles are adversely affected in a number of human disorders, including myotonic dystrophy, oculopharyngeal muscular dystrophy, and the inflammatory myopathies; this may lead to dysphagia, regurgitation, choking while eating, and other symptoms. Disorders of LES function include gastroesophageal reflux disease (GERD) and achalasia, the latter characterized by impaired relaxation of the LES and perturbed peristalsis, often resulting in megaesophagus. The primary cause of achalasia is defective signaling between the NO-producing inhibitory myenteric neurons and neighboring smooth muscle cells (SMCs) of the LES [2,3]. Although studies have identified several mutant mouse lines that display these esophageal defects, their etiologies are poorly understood.
The mammalian esophageal musculature is unique in that it makes a transition from smooth to skeletal muscle, with most of this process occurring after birth. The esophagus is ensheathed by the muscularis externa (ME) [1,4,5]. Skeletal muscle comprises the proximal portion of the ME and is critical for swallowing and proximal waves of peristalsis. Smooth muscle surrounds the distal portion, including the esophagogastric junction, which harbors the LES. During mammalian development, the ME initially comprises only smooth muscle.
In the mouse, skeletal muscle precursors are first detected in the proximal ME at embryonic day (E) 13 [6]. Over the next 3 weeks of life, smooth muscle is replaced by skeletal muscle in a proximal-to-distal manner, and the adult ME pattern is nearly completed by postnatal day (P) 14 [7][8][9]. Human esophageal myogenesis is similar, although smooth muscle is maintained in a more proximal position than in the mouse, but is still restricted to the distal one-third of the esophagus [10]. Interestingly, the amount and distribution of skeletal muscle varies among individuals [11,12]. The developmental and cell biological mechanisms that underlie the smooth to skeletal muscle replacement process have been controversial, with multiple mechanisms proposed [7][8][9][13][14][15] (reviewed in ref. [13]). Rishniw et al. [9] were the first to suggest that some sort of distal compaction of smooth muscle cells was important in the replacement of smooth muscle with skeletal muscle. However, the mechanisms for this process remained unclear.
Our previous study provided insight into mechanisms that control morphogenesis of the esophageal ME [13]. We showed that mice lacking the multifunctional cell surface receptor Cdo (also called Cdon) have a defect in the postnatal developmental process whereby esophageal smooth muscle is replaced by skeletal muscle. Analyses of various skeletal myogenic markers revealed that proliferative skeletal muscle progenitor cells migrated in a transition zone (TZ) along the length of the esophagus in a proximal-to-distal manner, leaving differentiated myofibers in its wake. Distal to the TZ, smooth muscle fascicles underwent a morphogenetic process whereby they changed their orientation relative to each other and to the lumen. In a distal-to-proximal manner, smooth muscle fascicles altered their orientation from parallel to the lumen to nearly perpendicular to the lumen. Upon completion of the process, smooth muscle occupied only the most distal part of the esophagus and the LES. Cdo −/− mice had no defect in skeletal myogenesis or in the numbers of various skeletal muscle progenitor cell types in the TZ. Rather, they were specifically defective in SMC fascicular reorientation, leading to ectopic, proximal fascicles displaying an inappropriate orientation and an aberrantly proximal skeletalsmooth muscle boundary [13]. Importantly, Cdo −/− mice also displayed megaesophagus and achalasia, with a smooth muscle cell-intrinsic defect in NO-mediated LES relaxation. Taken together, our previous findings identified a mechanism of ME maturation and provided a link between esophageal morphogenesis and motility disorders. Although the ME maturation defect in Cdo −/− mice is almost certainly smooth muscle cell-autonomous, a role for TZ-based skeletal myogenesis in promoting the fascicular reorientation process seems likely.
Multiple mechanisms have also been proposed for the developmental origin of esophageal skeletal muscles (ESMs). Minchin et al. reported that, similar to trunk and limb muscles, ESMs arise from Pax3/Pax7-expressing cells that originate in somites [16]. However, Gopalakrishnan et al. have recently demonstrated that ESMs are derived from the cardiopharyngeal mesoderm, and their development is dependent on expression of Tbx1 and Isl1, genes important for specific muscles of the head but not the trunk [17]. These factors lie upstream of the myogenic bHLH factors, MyoD and myogenin, which promote commitment and differentiation of cells in the skeletal muscle lineage [18]. Consistent with a cranial mesoderm origin for ESMs, Pax3 was not expressed during development of these muscles and was dispensable for their development in mice [17].
In contrast to Pax3, Pax7 is expressed in skeletal muscle precursors of the esophageal TZ [13,[19][20][21][22]. Importantly, however, Pax7 is never expressed in the SMC lineage as shown by lineage tracing studies, immunochemistry, and in situ hybridization [13,[19][20][21][22]. Worl et al. used electron microscopy to show that Pax7 −/− mice had a reduced number of skeletal muscle precursor cells in the esophageal ME, defective esophageal skeletal myogenesis, and an aberrantly proximal skeletal-smooth muscle boundary [14]. As Pax7 is not expressed in the smooth muscle lineage, such mice offer an opportunity to ask whether TZ-based skeletal myogenesis might affect smooth muscle patterning in the esophagus. We report here that Pax7 −/− mice have a severe defect in the postnatal developmental process whereby esophageal smooth muscle is replaced by skeletal muscle. Our results suggest that this defect is due to reduced cell proliferation in the TZ, probably due to precocious differentiation of skeletal muscle precursor cells, leading to a deficit in skeletal muscle. This defect in Pax7 −/− mice is associated with both megaesophagus and defective maturation of esophageal smooth muscle patterning. Our findings add to the previous model for esophageal musculature development. PAX7-dependent cell proliferation in the TZ is important for the proximal-to-distal progression of esophageal skeletal myogenesis. Furthermore, loss of PAX7 leads to a non-autonomous defect that influences smooth muscle fascicle reorientation and development of a mature ME.

Mice
Pax7 +/− mice were originally obtained from Peter Gruss [20] and are from the colony maintained by the Rudnicki laboratory. These mice are on a 129Sv background. It has previously been reported that on a C57BL/6J background, Pax7 +/− esophagi have a mild defect in development of the esophageal musculature [14]. On the 129Sv background, we have not observed differences in esophageal development between Pax7 +/+ and Pax7 +/− mice, and at P21 they were indistinguishable in esophageal length, location of the skeletal-smooth muscle boundary, and luminal diameter (Additional file 1: Figure S1). Therefore, we used both Pax7 +/+ and Pax7 +/− animals as controls. All animal procedures were conducted in accordance with institutional guidelines for the care and use of laboratory animals as approved by the Institutional Animal Care and Use Committees (IACUC) of the Icahn School of Medicine at Mount Sinai and according to the Canadian Council on Animal Care guidelines and the University of Ottawa Animal Care Committee protocols.

Preparation of frozen sections
Dissected esophagi were prepared for histology by directly freezing in OCT. For longitudinal sections, incisions were made along the entire esophagus in order to clear out the ingesta, and the sheet of dissected esophagus was laid flat prior to freezing. Frozen tissue blocks were sectioned at 10 μm and placed on Superfrost Plus slides (Thermo Fisher Scientific).

Microscopy
Microscopy was performed at the Mount Sinai Microscopy Shared Resource Facility using Axioplan2 and Axioplan2 IE microscopes (Carl Zeiss) equipped with ×10, ×20, ×40, and ×100 objective lenses that had numerical apertures of 0.5, 0.8, 0.75, and 1.4, respectively. Images were acquired at room temperature with a camera (Axio Cam MRm; Carl Zeiss). Mosaic images were compiled using the stitching feature in the AxioVision software. The maximum luminal diameter and length of esophageal skeletal/smooth muscle were measured by using ImageJ software (National Institutes of Health).

Statistical analysis
Data were expressed as mean ± standard deviation. Differences were tested by two-tailed t test. Values with P < 0.05 were considered statistically significant.
No other gross abnormalities were observed in the GI tracts of Pax7 −/− mice. P21 Pax7 +/− esophagi resembled the wild type in esophageal length, location of the skeletal-smooth muscle boundary, and luminal diameter (Additional file 1: Figure S1); therefore, Pax7 +/+ and Pax7 +/− esophagi were used interchangeably as controls. Immunofluorescence analysis (IFA) of longitudinal sections of P21 esophagi with markers of smooth muscle (α-smooth muscle actin (αSMA)) and skeletal muscle (sarcomeric actin (SA)) revealed that Pax7 −/− mice have a mispatterned ME, with the skeletal-smooth muscle boundary occurring at an abnormally proximal position (Fig. 1c, d). Most smooth muscle in the esophagi of Pax7 +/− mice was found in a short, broad segment at the esophagogastric junction, where SMCs were bundled into fascicles that were stacked side by side with an orientation perpendicular to the lumen (Fig. 1i). In contrast, distal to the aberrantly proximal boundary in Pax7 −/− esophagi, there was a long, thin, ectopic extension of smooth muscle that connected to the LES (Fig. 1d, j). The fascicles in this ectopic region of smooth muscle in Pax7 −/− esophagi were parallel to the lumen, indicating they had failed to undergo the normal reorientation process. A few SMCs were dispersed within the skeletal muscle at the skeletal-smooth muscle boundary in both Pax7 +/− and Pax7 −/− esophagi (Fig. 1g, h).
Despite the mispatterned and thinner ME distal to the aberrantly proximal boundary in Pax7 −/− esophagi, the morphology of the smooth muscle fascicles appeared normal in the Pax7 −/− LES (Fig. 1k, l). The Pax7 −/− LES was Longitudinal sections of P21 Pax7 +/− and Pax7 −/− esophagi were stained with antibodies to α-smooth muscle actin (αSMA; red) and sarcomeric actin (SA; green) for smooth and skeletal muscles, respectively. c, d Pax7 −/− esophagi have an aberrantly proximally located skeletal-smooth muscle boundary (arrows). Note that the Pax7 +/+ esophagus is cut off at the top end, making it appear shorter than the Pax7 −/− esophagus, the full length of which is shown. In fact, at P21, Pax7 −/− esophagi are shorter than Pax7 +/+ esophagi (see Fig. 3e). e, f The proximal ME of Pax7 −/− esophagi has skeletal muscle of relatively normal appearance. g, h Smooth muscle cells are dispersed with skeletal myofibers around the skeletal-smooth muscle boundary (Boundary). Note that the ME of the Pax7 −/− esophagus is thinner than the control at the skeletal-smooth muscle boundary. i, j The distal Pax7 −/− esophagus has a long, thin extension of smooth muscle instead of the short, broad segment of smooth muscle found in controls. e-j The thin αSMA + layer adjacent to the ME is the muscularis mucosa (arrowheads). k, l Longitudinal sections through the LES of P21 Pax7 +/+ and Pax7 −/− esophagi mice were stained with antibodies to αSMA (green) and nNOS (red) to label inhibitory intramural neurons. m Quantification of the relative numbers of nNOS + cells to αSMA staining in the ME of the LES. Values are means ± SD, n = 5. Bars: (c, d) 1 mm; (e-l) 0.2 mm somewhat thinner than that of the control mice (control = 135.8 ± 9.6 mm (n = 5); mutant = 92.0 ± 12.9 mm (n = 6); p < 0.01). This may be due to a combination of defective development of the ME and the overall growth retardation displayed by these mice [20]. The ME of the proximal portion of Pax7 −/− esophagi was similar to that of Pax7 +/− esophagi, although the skeletal myofibers in mutants appeared to have a greater amount of between-fiber space than in controls (Fig. 1f).
Megaesophagus is sometimes associated with loss of nitrergic neurons in the myenteric plexus of the LES [3]. We therefore quantified the number of neurons in the LES from P21 Pax7 +/+ and Pax7 −/− mice. IFA of LES sections with an antibody to nNOS revealed that the number and pattern of nitrergic neurons were similar between control and mutant mice (Fig. 1k-m). Taken together, P21 Pax7 −/− mice showed megaesophagus with an aberrantly proximal skeletal-smooth muscle boundary. The morphological defects of the P21 Pax7 −/− ME were most apparent in the region between the skeletalsmooth muscle boundary and LES, but were not as severe in the proximal region and in the LES itself.
Defective ME patterning in Pax7 −/− esophagi To determine when these defects in Pax7 −/− mice arose, cross sections were taken from P0-P21 at proximal, midlevel, and distal regions. Expression of αSMA and SA was assessed. The proximal and distal ME were composed of skeletal and smooth muscles, respectively, and the cross sections of proximal and distal Pax7 −/− esophagi showed no obvious defects at P0, P7, and P21 ( Fig. 2a-f, m-r, s, u). Mid-level cross sections revealed that esophageal dilation became recognizable in Pax7 −/− esophagi at P7 (Fig. 2g-j, t), and this distention was more severe at P21 (Fig. 2k, l, t). H&E-stained cross sections revealed that the P21 Pax7 −/− ME was thinner than that of controls in the mid-level region (Fig. 2v-x). In contrast to the control mid-level ME, which was composed of skeletal muscle at P7 and P21, the Pax7 −/− mid-level ME displayed only smooth muscle at these time points (Fig. 2i-l).
These findings are consistent with the aberrant skeletal-smooth muscle boundary seen in P21 Pax7 −/− esophagi (Fig. 1d), and they suggested a defect in the process whereby smooth muscle is replaced by skeletal muscle in these animals. We therefore analyzed the proximal-to-distal progression of skeletal myogenesis during ME development. Expression of αSMA and SA was assessed in longitudinal sections taken at P0, P7, and P21 (Figs. 1c, d and 3a-d). The distal-most skeletal muscle cells in these sections were embedded in smooth muscle. Progression of ME development was monitored by measuring the distance from the distal-most SA + cell to the LES. At P0, this distance was about half of the length in Pax7 +/− esophagi, whereas it constituted nearly 80 % of the total length in Pax7 −/− esophagi (Fig. 3a, b, e). The distance from the distal-most SA + cell to the LES diminished in Pax7 +/− esophagi from P0 to P21, and the distal-most SA + cell was found just proximal to the LES at P21 (Figs. 1c and 3a, c, e). In contrast, the distal-most SA + cell in Pax7 −/− animals was at a significantly more proximal location than in Pax7 +/− animals as early as P0 and maintained through P21 (Figs. 1d and 3b, d, e). The distance from the distal-most SA + cell to the LES slightly decreased in Pax7 −/− esophagi by P21; however, the distance still constituted~65 % of the total length (Fig. 3e). The total length of Pax7 −/− esophagi is~20 % shorter than that of control animals by P21 (Fig. 3e) [14]. This is likely due to overall growth retardation in these mice [20], but cannot alone account for the ME patterning defects.

Pax7 −/− esophagi have reduced number of cells undergoing myogenic differentiation
To analyze the Pax7 expression pattern in developing esophageal ME, longitudinal sections of Pax7 +/− esophagi were taken at P7 and P21 and stained with an antibody for Pax7. At P7, most Pax7 + cells were found in the TZ, and their number decreased at more proximal locations (Fig. 4a, c, e, g). Pax7 + cells diminished from the entire esophagi at P21, when proximal-to-distal progression of skeletal muscle was complete (Fig. 4b, d, f, g). At this stage, relatively few Pax7 + cells were present, as these probably mainly represent the population of satellite cells seen in the adult esophagus [13,24].
We hypothesized that impaired proximal-to-distal progression of skeletal myogenesis in Pax7 −/− esophagi resulted from a decrease in the number of cells undergoing myogenic differentiation. We therefore assessed the expression of MyoD (a marker of determined myoblasts) and myogenin (a marker of differentiating myoblasts) during ME development. Longitudinal sections of P7 Pax7 +/− and Pax7 −/− esophagi were stained with antibodies for MyoD and myogenin. Consistent with our previous study [13], most MyoD + and myogenin + cells were found in the TZ and their numbers diminished at more proximal locations in control animals. Pax7 −/− mice had~30 % the number of MyoD + cells and~50 % the number of myogenin + cells as control mice (Fig. 5a-f).
Reduction of the number of cells expressing these markers of myogenesis may result from a decrease in the muscle progenitor cell population, either by alteration in cell proliferation or cell survival. To investigate cell proliferation in the developing ME, longitudinal sections were taken from P7 Pax7 +/− and Pax7 −/− esophagi and stained with antibodies for Ki67 and phospho-histone H3. Most Ki67 + and phospho-histone H3 + cells were found in the TZ in Pax7 +/− esophagi (Fig. 5g, k, l, p). Much lower numbers of Ki67 + and phospho-histone H3 + cells were found in the Pax7 −/− TZ (Fig. 5h, k, m, p). Consistent with the largely postmitotic nature of ME smooth muscle cells [13], fewer Ki67 + and phospho-histone H3 + cells were present in the distal ME below the TZ in both Pax7 +/− and Pax7 −/− esophagi (Fig. 5i, j, k, n, o, p). Therefore, cell proliferation in the TZ is substantially reduced in Pax7 −/− mice.
To confirm that cell proliferation was reduced in skeletal muscle precursor cells per se, we stained longitudinal P7 sections with antibodies for MyoD or another lineage marker, M-cadherin, plus Ki67. Pax7 −/− mice had~20 % the number of MyoD + /Ki67 + cells in the TZ as control mice (Fig. 5q-w). Pax7 −/− mice had only~13 % the number of M-cadherin + cells as controls (Fig. 5x-dd). Furthermore,~45 % of control M-cadherin + cells were also positive for Ki67, whereas none of the small number of M-cadherin + cells in the Pax7 −/− TZ were also positive for Ki67. These results indicate that cell proliferation of Fig. 2 Defects in ME patterning in Pax7 −/− esophagi. a-r Cross sections of P0, P7, and P21 Pax7 +/− and Pax7 −/− esophagi were stained with antibodies to αSMA, SA, and DAPI. j, l The P7 and P21 Pax7 −/− esophagi are dilated, and skeletal muscle is replaced by smooth muscle at the mid-level region (Mid). s-u Luminal diameter at the mid region of Pax7 −/− esophagi is larger than that of Pax7 +/− esophagi at P7 and P21. Values are means ± SD, n = 3-11. v, w Cross sections of P7 Pax7 +/− and Pax7 −/− esophagi at the mid region were stained with H&E. Ep epithelium, MM muscularis mucosa, ME muscularis externa. The black horizontal bars indicate the width of the ME; the asterisks indicate the lumens. x Quantification of the width of the ME, as shown in (v and w). Values are means ± SD, n = 3. Bars: (a-r) 0.2 mm, (v, w) 50 μm. **P < 0.01; ***P < 0.001 skeletal muscle precursors is strongly diminished in the Pax7 −/− esophagus.

Discussion
In this study, we investigated the role of Pax7 in esophageal musculature development. Pax7 −/− mice displayed megaesophagus with a severe defect in the postnatal developmental processes whereby esophageal smooth muscle is replaced by skeletal muscle and the ME acquires a mature pattern. The proximal-to-distal progression of skeletal myogenesis was impaired in Pax7 −/− esophagi, associated with a significant reduction in cell proliferation of skeletal muscle precursors in the TZ. Additionally, our findings revealed that Pax7 is important for proper postnatal patterning of the entire ME, including the distal smooth muscle layers.
We previously proposed a model for postnatal development of the esophageal ME [13]. During esophageal ME maturation, skeletal muscle progenitors commit to the myogenic lineage and differentiate within a TZ that migrates along the length of the esophagus in a proximal-to-distal manner. Distal to the TZ, SMCs are largely non-proliferative and bundled into long, thin fascicles. SMC fascicles are initially arranged end-to-end and parallel to the lumen. During ME maturation, SMC fascicles rearrange their orientation such that they are ultimately arranged side by side and nearly perpendicular to the lumen; as a consequence, at the end of ME maturation, the fascicles occupy only the distal portion of the ME. Fascicular reorientation occurs in a distal-toproximal manner. We suggested that this process "clears Fig. 3 Pax7 −/− mice display defective proximal-to-distal progression of ME development. a-d Longitudinal sections of P0 and P7 Pax7 +/− and Pax7 −/− esophagi were stained with antibodies to αSMA, SA, and DAPI. Pax7 −/− esophagi have an aberrantly proximally located skeletal-smooth muscle boundary (arrows). The SA + tissue near the distal end of the Pax7 +/− esophagus in (a) is the diaphragm. e The distance between the distal-most SA + cell and the LES was measured and is represented by the red portion of the histogram bars. The distance decreases progressively with age in Pax7 +/− esophagi, but this fails to occur in Pax7 −/− esophagi. Values are means ± SD, n = 3-5. Bars: 1 mm. ***P < 0.001 a path" for the TZ, which migrates distally and produces the skeletal myofibers that constitute the majority of the length of the mature ME of the adult mouse [13].
Mice lacking the multifunctional cell surface receptor, Cdo, have a mispatterned ME, including an aberrantly proximal skeletal-smooth muscle boundary, and they develop megaesophagus. These mice are specifically defective in SMC fascicular reorientation, without obvious alterations in TZ-based skeletal myogenesis [13]. Cdo −/− mice had an ectopic, proximal region of smooth muscle in which the fascicles remained in the end-to-end, parallel-to-the-lumen orientation that normally characterizes an earlier developmental stage. This phenotype is very likely to be SMC-autonomous [13]. In contrast, the esophageal defects in Pax7 −/− mice are almost certainly autonomous to the skeletal muscle lineage. Pax7 is expressed in skeletal muscle precursor cells throughout the body, including the esophageal TZ, but it is never expressed in the SMC lineage [13,[19][20][21][22]. Examination of esophageal ME patterning in Pax7 −/− esophagi revealed that their musculature consisted mainly of smooth muscle, having skeletal muscle only at the very proximal end. This phenotype was more severe in Pax7 −/− esophagi than Cdo −/− esophagi. Smooth muscle occupied the distal~20 % of P14 Cdo −/− esophagi, whereas it occupied the distal~65 % of P21 Pax7 −/− esophagi. There was no difference in the position of the skeletal-smooth muscle boundary between control and Cdo −/− mice at P0, but Pax7 −/− esophagi already had an aberrantly proximal skeletal-smooth muscle boundary at P0. Moreover, in contrast to Cdo −/− mice, the numbers of MyoD + and myogenin + cells were reduced, as was cell proliferation, in the Fig. 4 Expression of Pax7 in the esophageal musculature at P7 and P21. a-f Longitudinal sections of P7 and P21 Pax7 +/− esophagi were stained with antibodies to Pax7, laminin, and DAPI. Proximal, mid-level (Mid), and TZ regions were analyzed. The greatest number of Pax7 + cells was observed in the TZ at P7, and the numbers diminished in mid and proximal regions. The number of Pax7 + cells is diminished at P21 when proximal-to-distal progression of ME development is complete. g Quantification of Pax7 + cells in proximal (prox), mid, and TZ regions measured as a percentage of total DAPI + cells in the ME. Bars: 0.2 mm Pax7 −/− TZ. Therefore, the aberrantly proximal skeletalsmooth muscle boundary in Pax7 −/− esophagi resulted from reduced production of ESM.
Distal to the aberrantly proximal skeletal-smooth muscle boundary, the Pax7 −/− smooth muscle layer was very thin, with the fascicles parallel to the lumen for nearly the entire length of the ME. The LES was normally patterned, consistent with it being in the region in which fascicles reorient earliest [13]. As Pax7 is never expressed in the SMC lineage, the mispatterning of ESM in Pax7 −/− esophagi is presumably a non-cell-autonomous effect. It seems likely that smooth muscle mispatterning occurs as a consequence of impaired proximal-to-distal progression of the Pax7 −/− TZ, but further work will be required to prove this. One physiological consequence of this combination of deficient skeletal myogenesis and SMC fascicle mispatterning is that Pax7 −/− mice have an enlarged luminal diameter specifically in the mid region of the esophagus (i.e., megaesophagus). We speculate that smooth muscles, in this abnormal location and fascicular arrangement, are not successful in maintaining the full integrity and function of the ME. This may occur for a variety of non-mutually exclusive reasons, including defective peristalsis, causing the slow passage of food and physical distension; suboptimal innervation; or some combination of these and additional processes.
Our findings support and add to the previous model for esophageal musculature development. Based on the fact that Pax7 −/− mice display a presumably non-SMCautonomous defect in fascicular reorientation, we propose the following hypothesis. Smooth muscle fascicular reorientation may be facilitated by forces exerted via the proximal-to-distal progression of the TZ. This progression may "push" the SMCs to a more distal region, where they are eventually exposed to the signals that trigger the rearrangement of SMCs relative to one another, resulting in reorientation of fascicles. Proliferation of skeletal muscle precursor cells was significantly reduced in the Pax7 −/− TZ during esophageal muscular development. This suggests that robust cell proliferation, which presumably participates in driving the proximalto-distal progression of the TZ, depends on PAX7. This is consistent with the known roles of PAX7 in maintaining proliferation and preventing precocious differentiation [24,25]. Therefore, it is most likely that skeletal muscle precursor cells differentiate prematurely during esophageal myogenesis in the Pax7 −/− TZ, resulting in reduced production of ESM.
That robust cell proliferation in the TZ may be critical for the proximal-to-distal progression of skeletal myogenesis is supported by a mathematical model of cell migration [26]. This model predicts that proliferation at the invading front is the key mechanism driving directed migration. These predictions were experimentally validated with the directional migration and colonization of the gut by vagal neural crest cells that establish the enteric nervous system [26]. This "frontal expansion" model, which emphasizes the importance of proliferation to a carrying capacity, is potentially relevant to a wider range of invasion systems such as angiogenesis, epidermal wound healing, and malignant invasion [26]. Therefore, this model may also apply to the proximal-to-distal progression of esophageal skeletal myogenesis.
Megaesophagus is observed as a feature of several animal models of Duchenne muscular dystrophy, including some strains of mdx mice, and some dystrophic dog and cat models [27][28][29][30]. It was proposed that megaesophagus in some of these animal models was caused by a hypertrophic diaphragm and ensuing esophageal constriction, a very different scenario than that seen in Pax7 −/− mice. There is, however, little in the literature to suggest that megaesophagus is common in patients with muscular dystrophy or other myopathies. Megaesophagus occurs more frequently in severe cases of achalasia, as a consequence of inability to fully relax the LES [2,3]. It will be of interest for future studies to use electrophysiology and manometry to test if Pax7 −/− mice have a nonautonomous defect in LES function.

Conclusions
Our analyses on Pax7 −/− esophageal musculature development suggest that ME patterning requires robust, PAX7-dependent cell proliferation in the TZ, which helps drive the proximal-to-distal progression of skeletal myogenesis. This process may in turn be necessary for distal smooth muscle fascicle reorientation and development of a mature ME.

Additional files
Additional file 1: Figure S1. Pax7 +/+ and Pax7 +/− esophagi have similar luminal diameter, esophageal length, and location of the skeletal-smooth muscle boundary. (A) Cross-sections of mid-region (Mid) esophagi from P21 mice of the indicated genotype were used to quantify luminal diameters. Values are means ± SD, n = 3. (B) Longitudinal sections of P21 Pax7 +/+ and Pax7 +/− esophagi were stained with antibodies to αSMA and SA and with DAPI. The length of the entire esophagus was measured, as was the location of the skeletal-smooth muscle boundary. Values are means ± SD, n = 3. Pax7 +/+ and Pax7 +/− esophagi were not different in these parameters.
Additional file 2: Figure S2. Lack of apoptosis in control or Pax7 −/− esophageal musculature at P7. (A-D and F-I) Longitudinal sections of P7 Pax7 +/− and Pax7 −/− esophagi were stained with antibodies to cleaved caspase 3 (C-caspase 3) (A-D) and DAPI, or for TUNEL + cells (F-I) and DAPI. The TZ and distal region were analyzed. Apoptotic cells were not observed in the esophageal ME. (E and J) Pax7 +/− thymuses were used as a positive control and displayed easily detectable apoptotic cells by both IFA for cleaved caspase 3 (E) and TUNEL (J). Bars: 0.2 mm.

Competing interests
The authors declare that they have no competing interests.