Morphological and molecular characteristics of living human fetuses ...

Human Reproduction Update 1997, Vol. 3, No. 1 pp. 59–78

European Society for Human Reproduction and Embryology

Morphological and molecular characteristics of living human fetuses between Carnegie stages 7 and 23: localization of inhibin mRNA α and βa subunits by in-situ hybridization L.M.Harkness1 and D.T.Baird Department of Obstetrics and Gynaecology, University of Edinburgh, 37 Chalmers Street, Edinburgh EH3 9EW, UK

TABLE OF CONTENTS Introduction Materials and methods Results Discussion References

Introduction 59 60 69 72 78

Localization of the mRNA α and βa subunits of inhibin has previously been reported in the human gonads during the second trimester. Adrenal inhibin has also been reported in the second trimester for the α, βa and βb subunits. Investigations showing localization by in-situ hybridization during the first trimester have not been reported. The results have shown hybridization of the α and βa subunits, throughout the period of development studied, in a variety of tissues including the dorsal and thoracic aortas and pericardium stages 13–22 (βa subunit); liver stages 19–21 (βa) and stages 21–22 (α); mesonephos stages 21 and 22 (βa); gonad stages (α and βa); adrenal stages 19–22 (α); surface ectoderm stages 16–22 (βa); mesenchyme stages 16–22 (βa); amnion stages 13–16 (βa); yolk sac stage 12 (α and βa); cartilage stages 19–22 (βa); and nasal proliferation stages 21 and 22 (βa). When compared with distribution of the protein subunits it was noted that more immunostaining activity was found, suggesting that that probes were not sufficiently sensitive enough to detect all levels of mRNA expressed. It can be surmised, therefore, that the lack of visual hybridization of the mRNA cannot preclude the possibility that it is not being translated within the tissue even though hybridization was not apparent. Key words: activin/embryo/inhibin/in-situ hybridization 1To

whom correspondence should be addressed

Localization of the inhibin α and βa subunits by mRNA in-situ hybridization has previously been reported during gonadal development in the human fetus. The α subunit was highly expressed at the beginning of the second trimester and decreased slightly during the second trimester; the βa subunit was expressed constantly throughout the second trimester (Erämaa et al., 1992). In the human fetus, the adrenals were found to contain α subunit mRNA, and in the testes the βb subunit was detected by Northern blot analysis; and by reverse transcription–polymerase chain reaction (RT–PCR) all three (α, βa and βb) subunits could be found in the fetal adrenals in the second trimester of gestation (Voutilainen et al., 1991). Investigations into developmental expression of inhibin in other mammalian embryos/fetus have been carried out in the rat, mouse, chick, ovine and bovine fetus. In the rat, βa subunit mRNA expression was found in the heart (12 days); dermal layer of the skin (13 days); whisker follicles, skeletal structure of the snout, limbs and intervertebral disks (14 days); striatum of the brain (16 days); hair bulbs (17 days); cerebral cortex and teeth (18 days); and in the tendons and gonads just prior to birth. βb subunit mRNA expression was noted in the cells surrounding the forebrain ventricle and in the gonads (14 days); and in the salivary glands (17 days); α subunit mRNA expression was only detectable in the gonads from 14 days post-coitum (Roberts et al., 1991). In 1994, Roberts and Barth reported the expression of inhibin mRNA in specific tissue by in-situ hybridization, after a re-analysis of tissue previously reported for inhibin expression (Roberts et al., 1991). Lu et al. (1993) reported developmental expression of inhibin βa and βb subunits in preimplantation mouse embryos; α was found neither in the ovary, oviduct, unfertilized oocyte nor embryo at early

60

L.M.Harkness and D.T.Baird

preimplantation stages. In the chick, immunoreactive inhibin was measured in the gonadal tissue of both male and female embryos by the Monash assay from homogenized tissue (Rombauts et al., 1992). In the ovine fetus, α subunit mRNA expression was measured, by in-situ hybridization, in the seminiferous tubules of the testis, with increasing concentrations from day 70 to term (Thomas et al., 1995). In the ovary, inhibin α subunit mRNA expression was localized in some follicles at term but only in small amounts (Englehardt et al., 1995) no expression of the βa subunit mRNA was found in either the testis or ovary. In the bovine fetus bioactive and immunoactive inhibin concentrations were measured in the developing gonads, and in addition these tissues were evaluated by Northern blot analysis and in-situ hybridization; both α and βa subunit mRNA expression was detectable throughout gestation but no βb subunit mRNA expression could be detected (Tourney et al., 1990). The aim of this study was to investigate the localization and level of expression of the inhibin α and βa mRNA subunits during the human embryonic developmental period from stage 12 (25–27 days post-ovulatory) to stage 22 (~56 days post-ovulatory). Six embryos, in excellent condition, were selected, one from each of the stages 12, 13, 16, 19, 21 and 22.

Figure 1. Quantification of inhibin α and βa subunit mRNA in trophoblast controls.

Human embryos and trophoblast tissue were obtained from first trimester pregnancy terminations, using a combination of RU 486 and prostaglandins. The tissues were processed within 1 h of removal from the cervical os or collection from the bedpan and placed in fresh 4% paraformaldehyde in phosphate-buffered saline (PBS), pH 7.4, overnight. The following day the tissue was dehydrated through a graded series of alcohols and then embedded in paraffin wax. Serial sections of a 5 µm thickness were mounted onto glass slides dipped in 2% 3-aminopropyltriethoxysilane (TESPA; Sigma, Poole, Dorset, UK) in acetone, twice; washed in acetone; washed in sterile water and left to air dry, and dried at 50C overnight.

and ultra pure water used to make up the final volume. The mixture was then incubated overnight at 37C. The following day an equal volume of Tris-saturated phenol solution (Amresco, Solon, OH, USA) was added and the plasmid preparation centrifuged at 11 600 g for 3 min. The bottom layer was removed and an equal volume of phenol:chloroform:isoamyl alcohol, 25:24:1 (Amresco) added to the tube, centrifuged at 11 600 g for 3 min. Again the bottom layer was removed before an equal volume of choloroform:isoamyl alcohol (49:1) was added, and centrifuged at 11 600 g for 3 min. The bottom layer was removed and 400 µl of absolute ethanol and 40 µl of 10.5 M ammonium acetate was added to the tube before incubation overnight at –20C. The following day the tube was removed from the freezer and brought to room temperature before being spun at 11 600 g for 15 min, the supernatant carefully removed and 400 µl of 75% ethanol added. This was then spun at 11 600 g for 3 min and the supernatant removed; the tube was left until the ethanol had evaporated and then TE buffer added so that the template was at a concentration of 1 µg/µl. The template was then stored at –20C until required.

Preparation and labelling of cDNA probes

Checking the DNA template

DNA probes for the α and βa subunits of inhibin were kindly provided by R.G.Forage, Sydney, Australia (Forage et al., 1986) and the plasmid preparations were carried out as previously reported by Sambrook et al. (1989). An aliquot of plasmid (15–20 µg) was incubated with transcription buffer, appropriate to the enzyme (Promega, Southampton, UK); the restriction enzyme, to cut for either the sense or antisense template (final concentration 1:20);

Molecules of double stranded linear DNA (template) become orientated according to the charge in an electrical field and therefore migrate through gel matrices at rates inversely proportional to log10 of the number of base pairs. The location of the DNA in the gel can be directly determined by staining with ethidium bromide which binds to DNA by intercalating between bases, and fluoresces under UV light, and can then photographed for a permanent record.

Materials and methods Tissue preparation

Localization of inhibin mRNA α and βa subunits by in-situ hybridization

61

Figure 2. Dark-field (A, C, E, G) and bright-field (B, D, F, H) micrographs showing hybridization of the inhibin α and βa subunits. (A, B) The α antisense subunit demonstrating intense hybridization in the syncytio and cytotrophoblast cells. (C, D) α sense subunit demonstrating no hybridization beyond the background level of expression. (E, F) Intense hybridization of the βa antisense subunit in the syncytio and cytotrophoblast cells. (G, H) βa sense subunit showing no hybridization above the background level of expression. Bar = 200 µm.

62


boric acid; 0.02 M EDTA). The gel solution was dissolved by microwaving and the temperature returned to 50–60C before 0.5 ng/ml ethidium bromide was added. The gel was poured into the mould or sealed plate and the comb set into position, 0.5–1 mm above the mould; the final depth of the gel being 3–5 mm depth. After the gel had set completely (30–45 min) the comb was carefully removed and the mould loaded into the electrophoresis tank. Running buffer (TBE) was poured into the tank until it covered the gel to a depth of ~1 mm. The template (1 µg) was mixed with 1 µl of DNA loading dye (0.25% Bromophenol Blue, 30% Ficoll, in water) and ultra pure water to take the total volume to 10 µl. A 1 kb DNA ladder (Gibco, Paisley, UK) was used as a standard to check the size of the DNA template; a 1 µl aliquot of this was mixed with 8 µl water and 1 µl DNA loading dye. The DNA migrates from negative (black) to positive (red) and a voltage of 75 V was applied. The gel was run for 1–2 h or until the dye front could be seen to have migrated approximately two thirds of the way through the gel, the power supply was then removed and the gel photographed in a dark room over a fluorescent light box. Preparation of probes

From a full-length human α inhibin probe, a fragment of 360 bp between PstI and EcoRI (nucleotides 752–1112) was subcloned into pGEM3Zf(+) (Promega) and the templates cut with EcoRI (using SP6 polymerase to make sense probes) and HindIII (using T7 polymerase to make antisense probes). From the full-length human inhibin βa cDNA a fragment of 618 bp (between HindIII, base 312 and PstI, base 930) was subcloned into pBluescript II SK+ (Promega). Digestion with BssHII isolated the whole inset plus the multiple cloning site and the T7 and T3 promoters. The template was used to make sense RNA using T7 polymerase (748 bp), or antisense RNA using T3 polymerase (745 bp). Labelling of probes

Figure 3. (a) Quantification of inhibin βa subunit mRNA in the dorsal and thoracic aortas. (b) Quantification of inhibin α and βa subunit mRNA in the pericardium.

The cut template was checked by gel electrophoresis using a 0.8% agarose gel in TBE buffer (0.1 M Tris; 0.1 M

A 1 µg aliquot of template was added to a solution of 5× transcription buffer (provided with the enzymes ordered from either Promega or Stratagene, Cambridge, UK) with 1 µl each of 10 mM riboprobe (r)CTP, 10 mM rATP, 10 mM rGTP (all Stratagene) and 1 M dithiothreitol (DTT; Sigma); sterile water (making the volume of template plus water to 8µl); [35S]-rUTP, 12 µl (1 mCi/100 µl; Amersham, Little Chalfont, Bucks, UK) all at room temperature. A 1.2 µl aliquot of RNase inhibitor (rRNasin, 40 000 IU/ml; Promega) and a 0.8 µl aliquot of the relevant polymerase enzyme (T7 RNA, 15 000 IU/ml; T3 RNA, 20 000 IU/ml; SP6 RNA, 20 000 IU/ml; Promega) was added on ice, bringing the total volume to 31 µl, before the


63

Figure 4. Dark-field (A, C, D, F, G, I, J, L) and bright-field (B, E, H, K) micrographs demonstrating hybridization of the inhibin βa sense and antisense subunits in the developing cardiovascular system. Bar = 200 µm. (A, B, C) Hybridization in the cells surrounding the dorsal aortas (da) in a stage 13 embryo; (A) weak hybridization of the βa antisense subunit; (B) bright-field; and (C) βa sense subunit. (D, E, F) Hybridization of the cells of the umbilical artery (ua) in a stage 13 embryo; (D) hybridization of the βa antisense subunit, (E) bright-field, and (F) βa sense subunit. In stage 21, hybridization is demonstrated in the thoracic aorta (ta) and the pericardium (p); (G) βa antisense subunit hybridization in the thoracic aorta, H) bright-field, and I) βa sense subunit in the thoracic aorta; (J) intense hybridization of the βa antisense subunit in the pericardium, (K) bright-field and (L) Non-specific hybridization of the βa sense subunit and also showing blood cells as false positives.

64


Figure 5. Dark-field (A, C, D, F, G, I, J, L) and bright-field (B, E, H, K) micrographs demonstrating in-situ hybridization of the inhibin βa subunit in the cardiovascular system of a stage 22 embryo. Bar = 200 µm. (A, B, C) Intense hybridization in the cells surrounding the atrio-ventricular canal (av); (A) βa antisense hybridization; (B) bright-field; and (C) βa sense hybridization. (D, E, F) Intense hybridization in the endocardial cushion (ec); (D) βa antisense hybridization; (E) bright-field; and (F) βa sense hybridization. (G, H, I) Intense hybridization in the parietal pericardium (pp); (G) βa antisense hybridization; (H) bright-field; and (I) βa sense hybridization. (J, K, L) Intense hybridization in the thoracic aorta (ta); (J) βa antisense hybridization; (K) bright-field; and (L) βa sense hybridization.


65

Figure 6. Quantification of inhibin α and βa subunit mRNA in the digestive system.

mixture was incubated at 37C for 25 min. A second 0.8 µl aliquot of the relevant polymerase enzyme was added to the solution and it was again incubated at 37C for a further 25 min. After incubation had been completed, 1 µl was removed and added to 20 µl of TE buffer (0.01 M Tris–HCl pH 8.0, 0.001 M EDTA) for the calculation of percentage incorporation of the labelled probe. To the rest of the probe mixture, 2 µl of yeast RNA (10 mg/ml; Sigma) and 1 µl of RNase-free DNase (10 IU/µl; Stratagene) was added and the probe incubated at 37C for a further 10 min. One µl of 200 mM EDTA, 67.5 µl of TE + 50 mM DTT, and 103.5 µl of saturated phenol (Amresco) was added to the probe, on ice, then vortexed before spinning at 11 600 g for 3 min. The bottom layer of the tube was discarded and another 103.5 µl of phenol extract added, vortexed and spun as before. The bottom layer was again removed before 103.5 µl of phenol:chloroform:isoamyl alcohol 25:24:1 (Amresco) was added, on ice, vortexed and spun as above. After the bottom layer had been discarded 200 µl of 6 M NH4 acetate (pH 5.2) with 1 µl tRNA (10 mg/ml; Sigma) and 400 µl of ice-cold absolute ethanol was added. The mixture was either precipitated at –20C overnight or at –70C for 2 h. Following precipitation, the probe mix was brought to room temperature and then spun at 11 600 g for 30 min, the top layer removed and, as the probe could not usually be seen, a β counter was used to keep a check on the probe in the tube. 300 µl of 80% ethanol in TE buffer with 50 mM DTT was added to the probe, spun at 13 000 r.p.m. for 10 min and the supernatant removed, this step was repeated, then 500 µl of absolute ethanol used to dehydrate the probe and spun at 11 600 g for 5 min. The ethanol was removed and the pellet (invisible) air-dried. The probe was

re-dissolved in 20 µl TE buffer/2.5 mM DTT, and stored until used at –70C.

Probe calculations

Two 10 µl aliquots of labelled probe (1 µl removed after the second incubation with the polymerase enzyme plus 20 µl TE buffer) were placed onto glass microfibre filters (Whatman, Maidstone, UK) and left to dry. One filter (unwashed prehybridization) was place in a scintillation vial and the other (washed post-hybridization) was washed three times with ice-cold 3% trichloroacetic acid (TCA) and then ice-cold absolute ethanol, left to dry and then also placed in a scintillation vial. Ecoscint (4 ml; National Diagnostics, Atlanta, GA, USA) was added and the vials counted on a β-counter (LKB, Wallac, 1216 Rackbeta II, Milton Keynes, UK), for 1 min on an [35S]-channel. The counts per min (c.p.m.) from the washed and unwashed filters were noted and a percentage of bound probe calculated. As 1 µl of labelled probe had been split between the two filters the c.p.m. of washed filter vial was doubled to give c.p.m./µl; this was again doubled to give the d.p.m./µl; this was then multiplied by 19 to calculate the total volume of probe in the tube (or d.p.m./µl). Routinely 13 µl of probe + hybridization mix were allocated for each slide; although only 10 µl were used. The number of slides per probe run was multiplied by 13 to give the total volume of probe mix needed, the volume of labelled probe was then calculated by dividing. The volume of the probe mix in d.p.m./total volume of probe mix, by the volume of d.p.m./µl of the original counts. To this was then added a 5% volume of 1 M DTT of the total

66


Figure 7. Dark-field (A, C, D, F, G, I, J, L) and bright-field (B, E, H, K) micrographs demonstrating in-situ hybridization of the developing digestive system. Bar = 200 µm. (A, B, C) Weak hybridization of the βa subunit in the cells surrounding the mouth (m) in a stage 19 embryo; (A) βa antisense hybridization, (B) bright-field, and (C) βa sense hybridization. (D, E, F) Weak hybridization of the βa subunit in the cells of the hepatic primordium (li) in a stage 19 embryo; (D) βa antisense hybridization, (E) bright-field, and (F) βa sense hybridization. (G,H,I) Weak hybridization of the βa subunit in the pseudostratified epithelial cells (ce) lining the stomach (st) in a stage 22 embryo; (G) βa antisense hybridization, (H) bright-field, and (I) βa sense hybridization. (J, K, L) Hybridization of the βa subunit in the pseudostratified epithelial cells (ce) in the lining of the gut (g) in a stage 22 embryo; (J) βa antisense hybridization, (K) bright-field, and (L) βa sense hybridization.


Figure 8. Quantification of inhibin α and βa subunit mRNA in the urogenital system.

Figure 9. Quantification of inhibin α mRNA in the primordial adrenals.

probe mix volume and the rest of the volume made up with hybridization buffer. In-situ hybridization

Serial sections (5 µm from every 50 µm) were floated and mounted onto pre-washed, pre-baked glass slides treated with a 2% solution of TESPA (Sigma) in acetone. Sections were deparaffinized twice in Histoclear (National Diagnostics) for 10 min each, rehydrated through a graded series of ethanols (100, 95, 85, 70 and 50%; 2 min each), washed with PBS and sterile saline for 5 min each before

67

being post-fixed in 4% paraformaldehyde pH 7.4 (PFA) in PBS for 20 min. The slides were washed twice in PBS (5 min each) before digestion with proteinase K (20µg/ml; Sigma) in 100 mM Tris and 10 mM EDTA buffer for 7.5 min; this facilitated the penetration of the probe into the tissue and helped to denature the DNA. Slides were washed again in PBS and refixed in 4% PFA in PBS, pH 7.4 for 5 min. Next, the slides were washed twice in 0.25% acetic anhydride (to reduce the electrostatic binding of the probe, inhibit proteinase K activity, and serve to block the binding of the probe to the coated slides) in 0.1 M triethanolamine hydrochloride (TEA; 10 min each); washed again in PBS and saline (5 min each), dehydrated through a graded series of ethanols (50, 70, 85, 95%, 1 min each, 100% twice for 5 min) and air-dried. The labelled probes were diluted to a concentration of 1 × 106 d.p.m./slide with a hybridization buffer (62.5% deionized formamide; 25% of a 50% dextran sulphate solution; 7.5% 5 M NaCl; 1% 1 M Tris, pH 8.0; 0.2% 500 mM EDTA; 2.5% 50× Denhardt’s solution; 1% 1 M DTT and 0.3% ultrapure water). heated to a temperature of 80C for 2 min and cooled on ice before application to sections at 10 µl/slide. Sections were overlaid with coverslips made from Gel Bond (FMC Bioproducts, Rockland, ME, USA) and the slides incubated overnight in a humidified chamber saturated with 5× standard sodium citrate (SSC; 1× = 0.15 M NaCl and 15 mM sodium citrate) and 50% formamide at 55C. After hybridization the slides were washed in 5× SSC with 0.01 M DTT at 55C for 30 min, then in 2× SSC with 50% deionized formamide and 0.1 M DTT for 20 min at 65C, washed three times in NTE buffer (0.5 M NaCl, 10 mM Tris, 5 mM EDTA, pH 7.5) for 10 min each at 37C; treated with ribonuclease type III A (Sigma, 40 µg/ml in NTE to reduce the background) for 30 min at the same temperature, then re-washed in NTE at 37C for 15 min. The wash in 2× SSC with 50% formamide and 0.1 M DTT (20 min, 65C) was repeated. The slides were then washed in buffers containing decreasing amount of salt solutions at room temperature (2× and 0.1× SSC, three times each, 10 min each). Slides were then dehydrated through a graded series of ethanols (50 and 85% containing 0.3 M NH4 acetate; 100% twice for 5 min each) and left to air dry. Autoradiographic dipping in liquid emulsion was performed in a dark room with a safety light Kodak NBT-2 (Rochester, NY, USA), diluted 1:1 in water, immediately before use, at 40–42°C. Slides were exposed in the dark for 4 weeks at 4°C, developed for 4 min in Kodak D-19 developer at 15°C, washed for 1 min in water and fixed for 5 min (Kodafix:water; 1:4) at room temperature, and finally washed in water. Slides were counterstained with haematoxylin (BDH, Glasgow, UK), dehydrated and

68


Figure 10. Dark-field (A, C, D, F, G, I, J, L) and bright-field (B, E, H, K) micrographs demonstrating in-situ hybridization of the inhibin βa and α subunits in the developing urogenital system. Bar = 200 µm. (A, B, C, D, E, F) Hybridization of the βa subunit in the mesonephros (mne) but not in the cells surrounding the mesonephric ducts (md) of a stage 21 (A, B, C) and a stage 22 (D, E, F) embryo; (A, D) βa antisense hybridization, (B, E) bright-field, and (C, D) βa sense hybridization. G,H,I) Intense hybridization of the βa subunit in the chromaffin cells of the adrenal primordial medulla (ad) in a stage 22 embryo; (G) βa antisense hybridization, (H) bright-field, and (I) βa sense hybridization. (J, K, L) Intense hybridization of the βa subunit in the testis cords (tc) of a stage 22 gonad (go); (J) βa antisense hybridization, (K) bright-field, and (L) βa sense hybridization.


69

Figure 11. Quantification of inhibin α and βa subunit mRNA in the yolk sac, amnion, skin and mesenchyme.

mounted in Eukitt (O.Kindler GmbH and Co, Freiburg, Germany), before examination under bright and dark-field optics using a Olympus microscope. Quantification of the probe

Slides were examined under dark field using an Olympus microscope attached to a video camera and computer. The concentration of grains × 106/cm2 were measured using an image analysis program (Olympus PC-Task Manager Version 1.20A, Cue 2 Image Analysis Morphometry) and the background for each slide subtracted from those positive tissues counted. Each hybridization run was quantified and then all of the runs for both probes were normalized against each other using trophoblast tissue included, in each run, as a control to calculate the differences. The normalized results were meaned and the SEM calculated, before the results were plotted. Student’s t-test was performed on the normalized values between the different stages for each of the systems, and the t and P values noted.

Results Each of the individual quantification results were plotted according to the developmental stage and to the system to which they belonged. The controls used were from 9 week trophoblast tissue (Figure 1) which was known to express both the α (Figure 2A–D) and βa (Figure 2E–H) subunits in the syncytiotrophoblast and cytotrophoblast layers (Tovanabutra, 1991). Each of the six chosen embryos were serial sectioned and every 10th 5 µm section was probed for either the α or βa subunits. Therefore all of the developing

systems within each embryo were probed for inhibin– activin. Cardiovascular system

Expression of the βa subunit was found at stage 13 in the endocardial tube, dorsal aortas (Figure 4A–C) and midline dorsal aortas (Figure 3a, although only weak positive expression was found. At stage 16, a highly significant increase (t = 6.6962, P = 0.0000) was noted in the amount of βa subunit expression in the dorsal aorta; in the pericardial tube no difference was noted between the different stages; and expression was also noted in the umbilical artery (Figure 4D–F). Stage 21 brought another, but non-significant, increase in the amount of βa subunit expression in the thoracic aorta (Figure 4 G–I) and in the pericardium (Figure 4J–L), and α subunit expression was noted in the pericardium. At stage 22, there was a highly significant decrease (t = 3.5285, P = 0.0020) in the amount of βa expression in the thoracic aorta (Figure 5J–L), although some of this expression could be due to a higher background than had previously been found. The βa expression for the rest of the cardiovascular system was subdivided at stage 22 into the parietal pericardium (Figure 5G–I), ventricle, endocardial cushion (Figure 5D–F) and atrio-ventricular canal (Figure 3b and Figure 5A–C). No significant difference was found between any of the different cardiovascular parts at this stage. No expression of either the α or the βa subunit was found prior to stage 13. Digestive system

Expression of the βa subunit of inhibin was detected in the cells of the hepatic primordium of the liver at stage 19; in the

70


Figure 12. Dark-field and light-field micrographs demonstrating in-situ hybridization of the α and βa subunits of inhibin in the body cavities, skin and mesenchyme of developing embryos. Bar = 200 µm. (A, B, C, D, E, F) Hybridization of the α (A, B, C) and βa (D, E, F) subunits in the walls of the secondary yolk sac (ys) of a stage 12 embryo; (A, D) antisense subunit, (B, E) light-field, and (C, F) sense subunit. (G, H, I) Hybridization of the βa subunit in the walls of the amnion (a) in a stage 16 embryo; (G) antisense subunit, (H) light-field, and (I) sense subunit. (J, K, L) Intense hybridization of the βa subunit in the surface ectoderm of the skin (sk); (J) antisense subunit, (K) light-field, and (L) sense subunit. (M, N, O) Weak hybridization in the mesenchymal cells (mes) surrounding the sclerotome near the somites; (M) antisense subunit, (N) light-field, and (O) sense subunit. NB: both F and O micrographs show false positives where blood cells are detailed in white. This is not a positive silver grain count but a specific effect of hybridization on blood cells.


cells surrounding the mouth (Figure 7A–C) and the hepatic primordium of the liver of stage 21, no significant difference was found between the two stages for expression in the liver. The βa subunit was also noted in the pseudostratified epithelium of the stomach (Figure 7G–I) and the gut (Figure 7J–L) at stage 22. In small amounts, the α subunit expression was detected in the liver of both stage 21 and 22 (Figure 6 and Figure 7D–F), with a significant rise between the two stages (t = 2.5053; P = 0.0366). At no stage prior to stage 19 was either the α or βa subunit mRNA expression detected. Urogenital system

Expression was only detectable for the βa subunit after stage

71

21 and for the α subunit after stage 22. Expression of the βa subunit was detected in the tissue surrounding the mesonephric ducts at both stage 21 (Figure 10A–C) and 22 (Figure 10D–F). No significant difference was seen between the two stages (t = 2.3300; P = 0.0802). Specific expression of the α subunit was detected in the testis cords at stage 22 (Figure 8 and Figure 10J–L). The adrenal glands showed expression of the α subunit at stages 19, 21 and 22 (Figure 10G–I) in the chromaffin cells of the adrenal primordial medulla; prior to this no expression could be detected (Figure 9). No significant difference was noted between stages 19 and 21 (t = 1.3829; P = 0.3008), and a small difference was noted between stages 21 and 22 (t = 3.0768; P = 0.0179).

Table I. Summary showing the localization of the inhibin α and βa mRNA subunits of the human embryo during embryogenesis. All developmental systems for each of the six different embryos were probed for both of the subunit mRNAs Stage

12

13

16 α

19 βa

α

21 βa

α

22

α

βa

α

βa

βa

α

βa

Cardiovascular system Aorta

–

–

–

+

–

++

–

–

–

+++

–

++

Umbilical artery

–

–

–

–

–

+

–

–

–

–

–

–

Pericardium

–

–

–

–

–

++

–

–

++

++

–

++

Endocardial tube

–

–

–

+

–

–

–

–

–

–

–

–

Endocardial cushion

–

–

–

–

–

–

–

–

–

–

–

+

A–V canal

–

–

–

–

–

–

–

–

–

–

–

++

Ventricle

–

–

–

–

–

–

–

–

–

–

–

++ –

Digestive system Liver

–

–

–

–

–

–

–

++

+

++

+

Mouth

–

–

–

–

–

–

–

–

–

++++

–

–

Gut

–

–

–

–

–

–

–

–

–

–

–

+

Stomach

–

–

–

–

–

–

–

–

–

–

–

+

Urogenital system Mesonephros

–

–

–

–

–

–

–

–

–

++

–

+

Gonad

–

–

–

–

–

–

–

–

–

–

++

+

Adrenal

–

–

–

–

–

–

++

–

+++

–

++

–

Extraembryonic membrane, skin and mesenchyme Yolk sac

+

+

–

–

–

–

–

–

–

–

–

–

Amnion

–

–

–

++

–

+

–

–

–

–

–

–

Skin

–

–

–

–

–

+

–

++

–

+++

–

+

Mesenchyme

–

–

–

–

–

+

–

+

–

–

–

+

Neural tube (antisense)

+

+

–

–

–

–

–

–

–

–

–

–

Neural tube (sense)

++

++++

–

–

–

–

–

–

–

–

–

–

Cartilage

–

–

–

–

–

–

–

+

–

+++

–

++

Skeletal/Neural

Respiratory system Nasal

–

–

–

–

–

–

–

–

–

++

–

+

Vomeronasal

–

–

–

–

–

–

–

–

–

–

–

+

–

–

–

–

–

–

–

–

–

–

–

+

Miscellaneous Rudimentary thymus

A–V canal = atrio–ventricular canal. – represent no hybridization of the subunit occurs at this stage; + represent intensity of silver grains with increasing symbols indicating increasing colour.

72


Miscellaneous tissue

Expression of the βa subunit was detected in the area surrounding the pseudostratified epithelium of the nasal passages at stage 21 (Figure 16A–C) and 22 with no significant difference noted between the two stages (t = 0.1563; P = 0.0565); and in the vomeronasal organ (Figure 15 and Figure 16D–F) and the rudimentary thymus at stage 22 (Figure 16G–I). Discussion

Figure 13. Quantification of inhibin α and βa mRNA subunit in the skeletal and muscular systems.

Extraembryonic membranes, skin and mesenchyme

Expression of the βa and α subunits was detected in this group of tissues, at stage 12 in the secondary yolk sac (Figure 12A–C α subunit, D–F βa subunit); βa expression was detected in the amnion at stages 13 and 16 (Figure 12G–I) with a significant difference between the two stages (t = 139.4387; P = 0.0046). At stage 16 expression was visible in the skin (Figure 12J–L) and mesenchyme (Figure 12M–O) surrounding the sclerotome near the somites. Stages 19, 21 and 22 all showed βa subunit expression in the skin and mesenchyme (Figure 11), with a significant rise in the expression in the skin between stages 19 and 21 (t = 2.0660; P = 0.0434) and a highly significant difference between stages 21 and 22 (t = 5.9152; P = 0.0000). No significant difference was seen in the mesenchyme between any of the different stages. Skeletal and neural systems

Expression of both the α and βa sense and antisense subunits was noted at stage 12 in the neural tube; the expression was tissue specific and a higher degree of expression was detected in the sense, with the antisense only just detectable (Figure 14A–C for α subunit; D–F for βa subunit). At stages 19, 21 and 22 expression of the βa subunit of inhibin was only found in proliferating chondroblasts (Figure 13 and Figure 14G–I). No significant difference was found between any of the different stages; stages 19–21, t = 0.8714; P = 0.4006; stages 21–22, t = 0.8412; P = 0.4154.

The data show that both the inhibin α and βa subunits are expressed in a variety of tissues throughout 3–10 weeks gestation in the human. It is known that the subunits of inhibin have close structural homology to the transforming growth factor (TGF)-β family. The subunits have been found to belong to a group of ancestral genes which have been highly conserved during evolution and it is clear that inhibin and activin share many of the growth regulatory properties involved in embryogenesis and cell differentiation. Reports on growth factors in embryogenesis in different species have been previously published (Hill et al., 1987; Gatherer et al., 1990), establishing the link with growth factors and developmental changes occurring during embryogenesis. The results from this chapter are summarized in Table I. Due to the number of slides involved in the study, it was not possible to include all of them in the same run and this meant that the quantification of the mRNA expression had to be validated; this was done in two different ways. Firstly, measurements of sense probes at all stages studied and in all developmental systems were made and these were then subtracted from the same tissue which was probed with antisense. Secondly, 7 week post-ovulatory trophoblast tissue was used in every run as a quality control. The concentration of grains/cm2 was normalized for each of the runs from the trophoblast tissue and the density of silver grains in the embryonic tissue was then normalized for each run, thus eliminating problems due to different probe labelling and different runs. All slides were treated with the same rigorous methods, which also allowed the quantification of the density of grains/cm2 to be analysed. No expression of inhibin or its subunits was found prior to stage 13 in the cardiovascular system. Both the α and βa mRNA subunit expression was found in the pericardium with βa expression, from stage 13, increasing in small but significant amounts until stage 21. After stage 21 differentiation had progressed to enough of an extent to separate the differentiating tissues from the general term of pericardium into ventricle, endocardial cushion, atrio-ventricular canal and parietal pericardium. The overall expression of the βa subunit appeared constant at both stages 21 and 22;


73

Figure 14. Dark-field and light-field micrographs of in-situ hybridization of the inhibin α and βa subunits in the developing skeletal and neural systems. Bar = 200 µm. (A, B, C, D, E, F) Hybridization of the α (A, B, C) and βa (D, E, F) subunits in the neural tube (nt) of a stage 12 embryo; less hybridization occurred in the antisense micrographs (A, D), than in the sense micrographs (C, F). (G,H,I) Hybridization of the βa subunit in the proliferating chondroblasts (ch) of one of the developing digits in a stage 20 embryo; (G) βa antisense hybridization, (H) bright-field, and (I) βa sense hybridization.

74


Figure 15. Quantification of inhibin α and βa subunit mRNA in the digestive and respiratory systems.