Differences in Formation of Prepuce and Urethral Groove During Penile Devel-Opment Between Guinea Pigs and Mice Are Controlled by Differential Expression of Shh, Fgf10 and Fgfr2

1. Introduction

Mammalian penis develops from the bisexual precursor named genital tubercle (GT) under the influence of androgens[1-4]. Compared with mice and rats, humans and guinea pigs form fully opened urethral groove at the end of the bisexual stage, and the process of tubular urethral formation follows distal-opening-proximal-closing, that is, the “Double Zipper” model[1,3]. In humans, penile urethra develops via an “Opening Zipper”, by canalization of the solid urethral plate to form the open urethral groove before 9.5 weeks of gestation, then androgen-induced proximal closure of the urethral groove starts tubular urethral formation in the penile shaft before 10.5 weeks [1]. Human preputial development is initiated at 10-11 weeks of gestation, which is almost the same time when tubular urethra starts to form in the proximal region. Using guinea pig model, we found cell proliferation in the outer layers (including basal layer) and programmed cell death in the inner layers of urethral epithelium play key roles during dorsal-to-ventral displacement and final opening of the urethral canal to form urethral groove, and this processes have no difference between males and females[5]. In mice, the urethral epithelium forms urethral plate, a urethral opening can be observed in the proximal region at E13.5, and from the urethral opening to distal tip, the ventral part of the urethral epithelium has never formed an opened urethral groove during fetal development and thus there’s no distal-opening-proximal-closing process during tubular urethral formation[6]. The preputial swellings appear as secondary outgrowths on the lateral edges of the tubercle at E13.5 and continue to grow laterally and ventrally to form the prepuce[6]. At E15.5, the preputial swelling covers proximal region of the glans before sexual differentiation[6]. Clearly, human preputial development is initiated at the same time of androgen-dependent-proximal-closing of urethral groove initiation, while in mice, the preputial development is initiated relatively earlier, begins at bisexual stage of the external genital development; unfortunately the mechanism causing the difference is totally unknown.

Theories on development of human prepuce fall into two opposing ideas. One idea is that development of the human prepuce involves formation and distal extension of the preputial folds to eventually completely cover the glans. This idea is exemplified by Hunter who described “folds of ectodermal tissue that appear to flow over the dorsum of the glans as the beginning of the prepuce then extends distally to cover the glans”[7]. An alternate theory is that a dorsal skin fold forms preputial fold and extends distally to cover the glans, the same time epithelial ingrowth occurs to form the preputial lamina. This hypothesis was proposed by Glenister in 1956[8]. The two theories were illustrated recently by Cunha et al.[9]. Liu et al[10] found that the prepuce initially forms on the dorsal aspect of the glans at approximately 12 weeks of gestation, after sequential proximal to distal remodeling of the ventral urethral plate along the ventral aspect of glans, the prepuce of epidermal origin fuses in the ventral midline. Cunha et al[9] proposed a novel morphogenetic mechanism of formation of the preputial lamina, namely the splitting of the thick epidermis of the glans into the preputial lamina and the epidermis via the intrusion of mesenchyme, and they found the process begins on the proximal aspect of the glans and extends distally. The cellular and molecular mechanisms of preputial development are unknown.

The development of external genitalia is controlled by local developmental genes, such as fibroblast growth factors (Fgfs), Sonic hedgehog (Shh), Bone morphogenetic protein (Bmp), Wnt and several Hox genes[11-13]. Several FGF genes, including FGF8, FGF10, and fibroblast growth factor receptor 2 (FGFR2) were found to be expressed in foreskin of children with hypospadias[14]. Mutation in Fgf10 or Fgfr2 induces genital malformation in mice[15,16]. It has been shown that BMP4 is required for initiation of GT outgrowth[17]. Additionally, knockout of an upstream regulator of Bmp4, Isl1, in murine shows abrogated genital outgrowth[18]. Several WNT ligands are expressed in the developing mouse GT, canonical WNT signaling is required in normal murine GT outgrowth[19,20]. WNT5A is one of the main WNT ligands regulating urethral tube formation as well as external genitalia outgrowth in mice[21]. Human autosomal dominant mutations in WNT5A are a cause of Robinow Syndrome which shows micropenis and hypoplastic scrotum in males, and hypoplasia of the clitoris and labia in females[22]. Extensive studies of Hoxd13 and Hoxa13 have demonstrated their essential role in development of the GT, double mutants exhibit agenesis of the GT and heterozygosity for Hoxa13 or Hoxd13 causes patterning defects of the phallus[23]. Human mutations in the HOXA13 gene are responsible for the range of phenotypes observed in Hand–Foot–Genital Syndrome[24]. Homozygous mutations of HOXD13 in men exhibit hypospadias[25]. Sonic hedgehog (SHH) is one of the most intensively studied genes in external genital development. It is a major morphogenic regulator of the outgrowth of the GT mainly through regulation of cell cycle progression[6,26]. A significant decrease in mRNA expressions of SHH and PTCH1 genes was found in boys with proximal hypospadias compared with boys without hypospadias[27]. Shh expression has been detected in the urethral plate in developing GT of guinea pigs. Compared to mice, Shh expression domain on the ventral side of developing GT in guinea pigs extends out to the ventral surface epithelium [5].

Here, we report the results of experiments aimed at identifying differentially expressed genes in a guinea pig model that more closely resembles the human penile development and tested the function of the selected genes in preputial development and urethral groove formation using organ culture. Since a detailed interpretation of guinea pig penile development at the stage of distal-opening-proximal-closing of urethra is not available, we first present a detailed embryological study of genital development between E27 to E33. Preputial development is initiated before sexual differentiation occurs in mice but after sexual differentiation starts in guinea pigs; we compare key genes’ expression patterns and levels between guinea pigs and mice; finally, we demonstrate that Shh and Fgf10 play critical roles in preputial development and urethral groove formation during distal-opening-proximal-closing of penile development.

2. Materials and Methods

Animals and Treatments

Sexually mature Hartley guinea pigs were purchased from Elm Hill Labs, ICR mice were purchased from Envigo RMS Inc (Indianapolis, USA). Guinea pigs and mice were housed in a pathogen-free barrier facility on 12-hour light/dark cycles with access to food and water ad libitum, and all experiments were conducted in accordance with the “Guide for the Care and Use of Laboratory Animals”. The experimental protocols (Guinea pigs: 20-014, mice: 23-011) were approved by the Institutional Animal Care and Use Committee of Southern Illinois University Carbondale. Time-mating of guinea pigs was achieved based on previously established method[3]. A minimum of 2 litters of embryos were collected at each stage, and GTs were dissected under a stereoscope. Sex of the embryos at E30 and later stages was identified by inspection of gonadal morphology under the stereoscope, and sex of the embryos at earlier stages was identified by genotyping using chromosomes Y (Sry) and X (Dystrophin) specific genes[28]. To detect cell proliferation in guinea pigs, 5-bromo-2′-deoxyuridine (BrdU, 50 mg/kg) was injected intraperitoneally and embryos were collected 4 h later for immunohistochemistry.

Organ Culture and In Vitro Cyclopamine, BGJ-398, Shh and Fgf-10 Administration

Mouse and guinea pig embryonic GT (at E14.5 and E27, respectively) culture setup followed the previously described method[29]. Methyltestosterone (MT) was dissolved in ethanol (0.05 M stock, filter sterilized) and then diluted into the medium to 10nM. Cyclopamine (Cat#: A8340), BGJ398 (Cat#: A3014) and mouse Shh protein (Cat# P1230) were purchased from ApexBio (Houston, USA). The Cyclopamine stock solution (10mM in DMSO) was diluted to final concentration (200nM) with culture medium before use. BGJ398 stock solution (5mg/ml in DMSO) was diluted to 1.4nM to inhibit Fgf-10/Fgfr2 using culture medium. Mouse Fgf-10 protein was purchased from R&D systems (Cat# 6224-FG-025/CF). Shh and Fgf-10 proteins were added into culture medium with the final concentration of 100ng/ml and 50ng/ml, respectively. The GTs were cultured for 48 h and then processed for morphology analysis.

Histology and Immunohistochemistry

Guinea pig and mouse embryos at different stages were harvested and rinsed with PBS. paraffin sections of GT were prepared and stained with hematoxylin and eosin as we described previously[3]. Immunohistochemistry was performed using anti-BrdU (G3G4, DSHB, Cat# AB 2314035, RRID: AB_2618097) according to procedures modified from previously established immunofluorescence protocol[5]. BrdU antibody was detected using ABC Kit (Vector Laboratories, PK-6100) according to the manufacturers’ operation manual. The Fgf-10 immunofluorescence in guinea pig and mouse GTs was performed following established method in our laboratory[3,5,29] using FGF-10 (H121) polyclonal antibody purchased from Santa Cruz (Cat# SC-7917, RRID: AB_2262731). Sample size, n =3 litters, slides obtained from 3 males (one from each litter) were selected for immunostaining.

In Situ Hybridization

Three embryos of different stages were selected for in situ hybridization analysis. In situ hybridization was performed as described [30] with some modifications [5]. To make the RNA probes, the cDNA of developmental genes was PCR-amplified from embryonic guinea pig cDNA using the primers designed with PrimerQuest software from IDT-DNA (https://www.idtdna.com/pages/tools/primerquest) and all primers were listed in Table S1, and then cloned to pGEM^®-T Easy Vector (Promega, Cat# A1360) according to operation manual except for Fgf10. Guinea pig Fgf10 IMAGE clone was purchased from GenScript (https://www.genscript.com/, Clone ID: ODo12835). Cloned DNAs were then amplified using M13 primers and reverse transcription reaction was performed using DIG RNA Labeling Kit (SP6/T7: Thermo Scientific, #EP0131/#EP0111) from Roche (Cat#: 11175025910) to make antisense RNA probes. Guinea pig Fgf10 RNA probe was synthesized using T3 RNA polymerase (Thermo Scientific, #EP0101).

Quantitative Gene Expression Analysis Using RT-QPCR

Total RNA was extracted from male GT of E12.5 and E13.5 mice, E23 and E26.5 guinea pigs using the TRIzol method according to the operation manual (Invitrogen, Cat: 10296010). RNA quality was assessed following published method[31]. The cDNA was synthesized from 500 ng total RNA using iScript Reverse Transcription Supermix (Bio-rad, Hercules, CA, USA). Primers of all guinea pig genes were designed using PrimerQuest Tool (Integrated DNA Technologies, Inc) to amplify cDNAs of around 90-150 bp sequences, and all exhibited similar amplification efficiency (r ≥ 97) as assessed by the amplification of control cDNA dilution series. Primers for mouse genes were designed and validated by OriGene. Primer sequences were summarized in Table S2. Quantitative PCR was performed using a CFX96 Real-Time PCR Detection System (Bio-rad) with iQ SYBR Green Supermix (Bio-rad) as the detector. The Real-Time QPCR was programed as 3 min for 95 °C followed by 40 repetitive cycles of melting (94 °C), annealing and extension (60 °C) for 10 and 20 sec, respectively. The cycle threshold (Ct) values were used to calculate the relative steady-state levels of specific mRNA in the samples. After amplification, the specificity of the PCR was determined by both melt-curve analysis and gel electrophoresis to verify that only a single product of the correct size was present. Data were normalized against a housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (Gapdh) using ∆∆Ct method[32].

Statistical Analysis

Cell proliferation and RT-qPCR data were subjected to statistical analysis using SPSS 22.0 software. Quantitative data were presented as mean ± standard error (mean ± SE). Paired t-test were used for comparisons. Statistical significance in the results was noted: *P≤0.05, ** P≤0.01, *** P≤0.001.

3. Results

3.1. Guinea Pigs Show Delayed Preputial Formation During the Glans and Preputial Development Compared with Mice

The early development of glans penis before sexual differentiation is similar between mice and guinea pigs[3,6]. The difference can be observed at a comparable stage of around E13.5 in mice and around E26.5 in guinea pigs. In guinea pigs, the opened urethral groove is first observed at E28[3]. Based on the limb and external genital development, the stage of E27 in guinea pigs matches E13.75 in mice (Figure 1A, I). At this stage, the shape of the urethral plate from distal (reverse triangle) to proximal (triangle) is also similar between guinea pigs and mice except that guinea pigs have a larger urethral canal (Figure 1B-H, J-O). The development of preputial swellings is initiated at E13.5 and preputial glands can be observed at around E14.5 in mice[6]. The epidermal epithelium folds to form preputial lamina on either lateral side of proximal GT of E13.75 mice, which subsequently separates the glans and preputial mesenchyme (Figure 1L, M).

In guinea pigs, no preputial swelling was observed, and no preputial lamina and preputial glands were found in all transverse sections from proximal to distal GT at E27 (Figure 1A-H). We found mouse preputial swelling like structure in E29 guinea pigs (Figure 2A) and preputial development is initiated around the same time as the initiation of sexual differentiation (Figure 2A-G and ref [3]). The epidermal epithelium folds to form preputial lamina from both lateral sides in the proximal region and separate the glans and preputial mesenchyme in E29 GT (Figure 2A-H). Compared sections in Figure 2G and Figure 2H, we can see that the epithelial ingrowth exists during preputial lamina formation in guinea pigs. The opened urethral groove emerges at E28 and sexual differentiation is initiated at E28 to E29 in guinea pigs[3]. In mice, sexual differentiation of external genitalia is observed around E16, and penile masculinization stage of E16.5 mice is comparable to that of E29 guinea pigs (Figure 2A-O). We can clearly see that the closing of urethral tube is initiated in proximal region of E29 GT (Figure 2F-H), then progressively extends distally and reaches at the distal tip of glans penis at E33 (Figure 3D-K).

As Figure 2 shows, we can see the fully opened urethral groove in middle region of glans penis (Figure 2E), not fully opened urethra epithelium with urethral canal in more distal region (close to the distal tip) (the opening zipper, Figure 2C, D), and closed tubular urethra in proximal region (the closing zipper, Figure 2F, G) in E29 guinea pigs. In E16.5 mice, the tubular urethra has formed in proximal region (Figure 2O), indicating this is a comparable penile masculinization stage to E29 guinea pig. In the middle region of E16.5 mouse penis, the urethral epithelium in glans region forms urethral plate without urethral canal, the medial boundary of epidermal epithelium derived preputial epithelium from either side, together with the ventral surface epithelium of GT, form a shallow groove (Figure 2L, M), and a urethral groove like structure is only found in distal region without preputial covering (Figure 2J). The prepuce has covered more than half of the glans and reached to distal region in E16.5 mice (Figure 2K-O), but only a very small portion of proximal region in E29 guinea pigs (Figure 2G, H). Interestingly, we found that in mid-distal region of E33 guinea pig penis, the epidermal epithelium has increased layers and the thick epidermis is split and remodeled by an intrusion of preputial mesenchyme to form the preputial lamina and a thin surface epidermis (Figure 3F-N). When we compare the proximal (Figure 3I-K) with distal (Figure 3D-H) region, the thicker epithelium in distal region is apparently noticeable.

The evaginating preputial mesenchyme can be observed from mid-distal boundary into the distal (Figure 3H, G and L-N). Similar processes can also be found in E16.5 mice (Figure 2L). In both E16.5 mice and E33 guinea pigs, we can see that the prepuce has fully covered the middle region of glans penis and preputial lamina has almost formed a circular (tubular) shape and separated the preputial and glans mesenchyme in the same region (Figure 2M, 3I, 3J). The only differences are the frenulum on the ventral side of E33 guinea pig penis (Figure 3A, I and J), and still unclosed urethra in middle glans region of E16.5 mouse penis (Figure 2M). Only fractional preputial lamina can be observed in proximal (Figure 2N and 3K) of glans penis at this stage, which suggests that the epidermal epithelium might evaginate towards proximal region of glans penis to form preputial lamina in both guinea pigs and mice.

The preputial glands in guinea pigs develop even later and no preputial glands have been found at any penile developmental stages before E30 (Figure 3A-H; ref[3]). The earliest stage we observed preputial glands in guinea pigs is E33, when the preputial glands are seen in proximal penis (Figure 3J, K). In addition, the epithelial tag can be distinguished in the distal-most region of E29 (Figure 2A) and E33 (Figure 3A-C) guinea pig penises, but not in mouse penises of comparable stages.

3.2. Differential Cell Proliferation in Epithelium and Mesenchyme Contributes to Preputial Development

In light of our findings that both epithelial and mesenchymal cells may evaginate to form preputial lamina and prepuce during penile development, we hypothesized that differential cell proliferation may exist and contribute to the epithelial ingrowth and mesenchymal evagination during preputial lamina formation in guinea pigs. To test this hypothesis, we performed BrdU labeling cell proliferation assay. Figure 4E is the transverse section of proximal (initiation of preputial development) GT of E29 guinea pigs, we found that there were more BrdU-positive cells (3.1 times, p=0.0026, n=3) in the epithelium than in the surrounded mesenchyme, and epithelium evaginated inside and split the mesenchyme (Figure 4E, F and I). In a slightly distal section (Figure 4C), even more BrdU-positive cells (5.6 times, p=0.0007, n=3) were detected in epithelium than in mesenchyme, the epithelium increased layers and a small portion of mesenchymal cells were separated from glans mesenchyme and formed the precursor cells of preputial mesenchyme (Figure 4C, D and H). In a more distal section (Figure 4A), the majority of BrdU-positive cells were restricted in preputial mesenchyme and epithelial layer adjacent to glans mesenchyme (Figure 4A, B and G). The results indicate that epithelial cell proliferation leads to the increase in epithelial layers and the evagination of preputial mesenchyme into thickened distal epithelium is due to mesenchymal cell proliferation. In distal region of E33 penis, the majority of BrdU-positive cells located in the basal epithelial layer adjacent to glans mesenchyme and developing preputial mesenchyme (Figure 4J and K). In proximal region of E33 penis, more BrdU-positive cells located in the basal layer of preputial lamina epithelium (Figure 4L). The results indicate that the fractional preputial lamina in proximal region of E33 penis (Figure 3K) are mainly derived from epithelial invasion.

3.3. The Expression Patterns of Key Developmental Genes in Guinea Pig Genital Tubercle

We next sought to identify the molecules that mediate GT development in guinea pigs and performed a comparison with previously identified important genes in mouse GT development. We carried out an in situ hybridization screening on guinea pig GTs of those developmentally important genes identified from mice, focusing on expression of Shh, Fgf, Bmp, Wnt and Hox genes. We first present key gene expression patterns of these pathways in E23-23.5 guinea pig GTs, which are comparable to E12.5 mouse GTs.

Previous work has reported the presence of Shh in developing guinea pig GT, and found the expression pattern was slightly different from that of mice at multiple stages[5]. Here we show at E23, Shh is expressed in urethral epithelium of guinea pig GT (Figure 5A), the pattern is similar to that of E12.5 mice[6]. The Shh receptor gene Ptch1 is expressed in a broad domain of the mesenchymal cells surrounding the Shh-expressing urethral epithelium at E23 in guinea pigs (Figure 5B), which is also similar to that of E12.5 mice[6]. In mouse GT, Shh is expressed in urethral epithelium from E11.5 to E14.5 and in preputial glands from E13.5 to E14.5[11,33]. To identify Shh expression at later stages in guinea pig GT, we performed Shh in situ hybridization during E27 to E33. Shh expression in urethral plate can be detected as late as E28 and no preputial expression of Shh has been observed at any stages before E28 (Figure 6A-C and ref[5]). The earliest stage that preputial Shh expression can be detected is around E32, when Shh is initially expressed in developing preputial glands as small dots in both male and female GT (Figure 6D, E).

Hoxd13 is expressed in the distal mesenchyme of of E23.5 guinea pig GT (Figure 5C, D), and the expression pattern resembles that of E12.5-14.5 mouse GT[6,34]. Interestingly, the expression of Hoxd13 in GT is relatively weaker compared with the strong expression in hindlimbs in guinea pigs (Figure 5C, D).

Bmp4 is expressed strongly in distal tip mesenchyme of E23.5 guinea pig GT (Figure 5E, F), similar to that of E12.5 mouse GT[6], the expression on ventral side is localized to the mesenchyme surrounding urethral epithelium, which resembles the expression of Ptch1 in this region (compare Figure 5B with 5E), and Bmp4 expression on dorsal side resembles the expression of Hoxd13 in the same region (compare Figure 5D with 5F). Bmp7 and Bmp4 have similar expression patterns in E23.5 hindlimbs (Figure 4E, H), but their expression patterns in GT are quite different, Bmp7 is expressed in both urethral epithelium and mesenchyme of E23.5 GT (Figure 5G, H), the urethral expression is weak and only confined to partial of urethral epithelium near distal tip, and the mesenchymal expression is not like Bmp4 and Ptch1, which are expressed in cells adjacent to urethral epithelium, but further away from urethra, mainly on the side of distal tip (Figure 5G, H).

Fgf8 expression is detected in a subset of the Shh expression domain in guinea pig GT and restricted to the anterior region of urethral epithelium at E22 (Figure 5J), remains to the distal tip of the urethral plate when the tubercle grows out at E23 and can only be seen at the dorsal view (Figure 5K, L). The pattern of Fgf8 in guinea pig GT also resembles the expression in mice[6,35]. Similar to Hoxd13, compared with strong Fgf8 expression in limb buds, the signal is much weaker in developing GT.

Fgf10 is mainly restricted in the urethral plate in E23 guinea pig GT (Figure 5M, N), which is different from reported mesenchymal expression in mouse GT[11,15]. The expression of Fgf receptor Fgfr2 in E23 guinea pig GT is also in urethral epithelium (Figure 5O, P) and the expression pattern of Fgfr2 in guinea pig urethra is similar to that in mice at early stages[16], but lack of preputial expression before E28 compared with mice of similar developmental stage (E13.5-14.5) because of no preputial development until E29. As the urethral plate expression of Fgf10 is not the same as reported in mouse GT, we performed Fgf10 and Fgfr2 in situ hybridization on GT of E22, E23.5 and E26 guinea pigs (comparable to E11.5, E12.5 and E13.5 mice respectively according to limb and GT morphology). We found that both Fgf10 and Fgfr2 expression are mainly restricted in the urethral plate (Figure 6F-L). Interestingly, Fgf10 expression was also found in the dorsal and ventral region of labioscrotal swellings in E26 guinea pig GT and were confined to 4 small, round areas (Figure 6H, I). The Fgf10 expression in labioscrotal swellings has never been reported before and the function of Fgf10 in these domains is unidentified.

Wnt5a expression domain locates in the mesenchyme of the GT with the strongest expression at the distal tip, from both ventral and dorsal view, we can clearly see that Wnt5a expression mainly locates in the distal glans region (Figure 5Q, R), suggesting it may control outgrowth of the glans at early stage as reported in mouse GT[21,36].

Taken together, these results reveal dynamic patterns of key developmental genes expression in guinea pig external genitalia, and Fgf8, Fgfr2, Bmp4, Bmp7, Hoxd13 and Wnt5a show similar expression patterns during GT development between guinea pigs and mice. We identified that Shh and Fgf10 show differential expression patterns between the two species, which may contribute to the differential morphogenesis of external genitalia between guinea pigs and mice.

3.4. Relative Expression Levels of Shh, Fgf10 and Fgfr2 in Developing Guinea Pig GT Are Reduced Compared with in Comparable Stage of Mouse GT

Compared with mice, the major difference in guinea pig GT development is that urethral plate opens to form urethral groove before sexual differentiation (Figure 1, Figure 2 and Figure 3, ref[3,5]). Shh, Fgfs, Hoxd13, Hoxa13, Bmp4 and Wnt pathways play key roles in glans and preputial morphogenesis[11-13]. In light of our findings about the differential expression of Hoxd13 and Fgf8 between GT and limb buds and between guinea pigs and mice, we hypothesized that the expression levels of these developmental genes in developing guinea pig GT may be relatively weaker than those in mouse GT. To verify this hypothesis, we tested and compared the relative expression levels of these genes between mouse and guinea pig GTs with comparable developmental stages. Developing limb buds were widely used to determine the developing stages in different species[37,38]. According to limb (Figure 7A-D) and external genital (Figure 7E-H) morphology, we found that E23 and E26.5 GTs of guinea pigs are comparable to E12.5 and E13.5 GTs of mice respectively. As we compare relative gene expression levels between two different species, selection of housekeeping gene is important. We compared Ct values of commonly used housekeeping genes Gapdh and Actb, and found the Ct values of Gapdh were closer to each other between mouse and guinea pig GTs and more stable when same amount of total RNA was applied (Table S3), thus, Gapdh was selected as the housekeeping gene to quantify the gene expression levels. Compared to 2 housekeeping genes, the Ct values for all the selected developmental genes, Shh, Fgf8, Fgf10, Fgfr2, Hoxd13, Hoxa13, Bmp4 and Ctannb1, of all the tested stages, were found to be higher. Compared with E12.5 and E13.5 mice, lower expression levels of all the 8 developmental genes were detected in GTs of E23 and E26.5 guinea pigs (p≤0.0061, n=5). Hoxa13, Bmp4 and Ctannb1 were about 2-fold lower in guinea pig GT compared with mice, and the expression of Shh, Fgf8, Fgf10, Fgfr2 and Hoxd13 were more than 4.5-fold lower compared with mice (Figure 7I, J). We also found Actb relative expression in developing GTs showed no significant difference (E23, 1.23fold, p=0.163, n=5; E26.5, 1.14fold, p=0.248, n=5) between guinea pigs and mice (Figure 7I, J). Compared with the expression in E12.5 mice, Shh, Fgf8, Fgf10, Fgfr2 and Hoxd13 in E23 guinea pig GT were 5.54-, 4.5-, 14.62-, 6.47- and 7.51-fold downregulated, respectively (Figure 7I), and likewise, all the 5 genes were dramatically upregulated in E13.5 mouse GT (Shh, 6.23-fold; Fgf8, 5.9-fold; Fgf10, 9.65-fold; Fgfr2, 7.24-fold; Hoxd13, 9.98-fold) compared with E26.5 guinea pigs (Figure 7J). Because these are relative gene expression levels in two species, we think that around 2-fold reduction of Hoxa13 (E23, 1.9-fold; E26.5, 2.8-fold), Bmp4 (E23, 2.1-fold; E26.5, 2.6-fold) and Ctnnb1 (E23, 2.5-fold; E26.5, 2.4-fold) may be due to systemic differences, the dramatical differences (more than 4.5-fold) in expression of Shh, Fgf8, Fgf10, Fgfr2 and Hoxd13 may play roles in patterning cellular processes and leading to differential morphological development such as urethral groove formation and preputial development between the two species.

3.5. Shh and Fgf10/Fgfr2 Play Key Roles in Preputial and Urethral Groove Formation

Shh is a chief morphogen which organizes the structure in the ventral midline of multiple organs[6,39,40], and deletion of Shh signaling genes at E13.5-15.5 led to an opened urethral plate (hypospadias) and reduction of glans and preputial development in mice [33,41]. Fgf10 and its receptor Fgfr2 also play important roles in urethral and preputial development[15,16,42]. Additionally, Shh and Fgf10/Fgfr2 signalings interact with the expression of other developmental genes, such as Fgf8, Hoxd13, Bmp4 and Wnt5a, coordinately regulate morphogenesis during GT development, e.g., Shh negatively regulates Bmp and Wnt signaling molecules, which will define the limits of the range of each domains[11,40]. Based on our findings that Shh expression levels reduces more than 5-fold in developing GT of guinea pigs than that of mice and the difference in Shh expression patterns between mouse and guinea pig GTs (Figure 6A-E and ref [5]), Fgf10 expression pattern shifts from in mesenchyme of mouse GT to mainly in urethral epithelium of guinea pig GT and expression levels in guinea pig GT reduces more than 14-fold at E23 (compared with E12.5 mouse GT) and more than 9-fold at E26.5 (compared with E13.5 mouse GT), we hypothesized that reduction of Shh and Fgf10/Fgfr2 signaling may induce urethral groove formation, and delay preputial development in guinea pigs. To test this hypothesis, we performed GT organ culture and tested the roles of Shh and Fgf-10 in urethral groove and preputial formation using Fgf-10 and Shh proteins, Fgf receptor inhibitor NVP-BGJ398, and hedgehog signal inhibitor cyclopamine. Male GTs of E14.5 mice and E27 guinea pigs were dissected and maintained in culture using media supplemented with or without androgen (10nM MT), Shh and Fgf10/Fgfr2 inhibitors (for mouse GT culture), and SHH and FGF-10 proteins (for guinea pig GT culture) to test the function of Shh and Fgf-10 in urethral groove and preputial development. We observed androgen’s effect on mouse GT culture first. After 48 hours of culture, both the control male mouse GTs with and without androgen developed preputial swellings and urethral plate (Figure 8A-C), but compared with the GTs with 10nM MT, there was a small hole remaining open at the middle of the urethral plate on ventral side of the GTs without MT (Figure 8B), thus we added 10nM MT in our male GT culture system when testing the effects of Shh, Fgf10/Fgfr2 and the inhibitors. We then performed the same stage of male mouse GT culture with administration of 1.4nM BGJ398, 200nM cyclopamine, or 1.4nM BGJ398 plus 200nM cyclopamine to the media from the beginning, all the three group of GTs formed smaller preputial swellings and widely opened urethral grooves compared with controls (Figure 8D, E). After 48 hours of culture with both 1.4nM BGJ398 and 200nM cyclopamine, the preputial swelling almost disappeared, the GT grew its shape into a cylinder with an opened urethral groove, in addition, the distal tip also malformed (Figure 8F). Next, we performed guinea pig GT culture (starting from E27) with or without 10 nM MT (Figure 8G-I). Because the guinea pig GT is very soft and bigger/heavier compared with the mouse GT around this stage, when being cultured with ventral side up position, the distal of the GT became flat and it seemed very hard for it to overcome the gravity to form a closed urethra using the same method as for mouse GT culture, we believe additional support material should be applied into our guinea pig GT culture system in the future. After the administration of 50ng/ml Fgf-10 protein, the guinea pig GT enlarged its size compared with the control and formed preputial swellings after 48 hours which have never been seen in controls (Figure 8J). After 48 hours of culture with 200ng/ml Shh protein, guinea pig GT also formed preputial swellings, but the elongation of the GT was not as obvious as Fgf-10-treated ones (Figure 8K). When male guinea pig GT was cultured with both 50ng/ml Fgf-10 and 200ng/ml Shh proteins, the preputial swellings became more obvious after 48 hours (Figure 8L). Our results suggest that Shh and Fgf10/Fgfr2 signaling play key roles in urethral groove and preputial formation during penile development.