regulation of sertoli cell number and activity by follicle-stimulating hormone and androgen during...

Regulation of Sertoli Cell Number and Activity byFollicle-Stimulating Hormone and Androgen duringPostnatal Development in the Mouse

HEATHER JOHNSTON, PAUL J. BAKER, MARGARET ABEL, HARRY M. CHARLTON,GARY JACKSON, LYNNE FLEMING, T. RAJENDRA KUMAR, AND PETER J. O’SHAUGHNESSY

Institute of Comparative Medicine, University of Glasgow Veterinary School (H.J., P.J.B., G.J., L.F., P.J.O.), Glasgow,United Kingdom G61 1QH; Department of Human Anatomy and Genetics, University of Oxford (M.A., H.M.C.), Oxford,United Kingdom OX1 3QX; and Departments of Pathology and Molecular and Cellular Biology, Baylor College of Medicine(T.R.K.), Houston, Texas 77030

The roles of FSH and androgen in the postnatal develop-ment of Sertoli cell number and function have been investi-gated using mice that lack FSH (FSH�KO), FSH-receptors(FSHRKO), or androgen receptors (Tfm). At birth and d 5,Sertoli cell number was normal in FSHRKO and FSH�KOmice, but was significantly reduced on d 20 and in adulthood.In contrast, Sertoli cell number was reduced at birth in Tfmmice and remained significantly less than normal up to adult-hood. Sertoli cell activity was determined through measure-ment of 11 different mRNA transcript levels. From birth toadulthood, the expression of most transcripts increased, witha significant rise occurring between d 5 and 10. In animalslacking FSH stimulation, mRNA expression (measured perSertoli cell) was largely normal on d 5, but was reduced in

seven transcripts on d 20 and in five transcripts at adulthood.In Tfm mice two transcripts showed reduced expression on d5, and four were reduced on d 20, although expression in adultTfm mice did not differ from that in normal cryptorchid con-trols. The results show that 1) testosterone, but not FSH, isrequired for Sertoli cell proliferation during fetal and earlyneonatal life; 2) FSH and testosterone both regulate the latestages of Sertoli cell proliferation; 3) FSH has a general tro-phic effect on Sertoli cell activity in the pubertal and adultmouse; and 4) androgens are required for specific transcriptexpression during prepubertal development. Specific effectsof androgens were not seen in the adult, although these maybe masked by the effects of cryptorchidism. (Endocrinology145: 318–329, 2004)

THE SERTOLI cell acts as the central regulator of tes-ticular development and function. The initial genetic

events determining testis differentiation give rise to theSertoli cells (1, 2), with development of other gonadalsomatic cell types likely to depend on Sertoli cell differ-entiation. In addition, the Sertoli cells act to regulate pri-mordial germ cell proliferation and development duringfetal growth (3). Postnatally, the Sertoli cells are essentialfor development and maintenance of spermatogenesisthrough direct interactions with the developing germ cells,and in the adult animal the overall germ cell number isdependent upon Sertoli cell number (4). In addition, it islikely that adult Leydig cell activity, function, and survivalare dependent on the continued presence of Sertoli cells (5,6). It is clear, therefore, that regulation of Sertoli cell pro-liferation and activity during development and in theadult animal is crucial for normal adult fertility.

Studies over a number of years have shown that FSH canact to regulate Sertoli cell function. Initial evidence camefrom the demonstration that FSH could maintain spermat-

ogenesis in hypophysectomized rats (7), and it was subse-quently shown that FSH could stimulate aspects of Sertolicell function in culture (8, 9). More recently, the use of mi-croarrays has allowed identification of more than 300 genesthat are regulated by FSH in cultured Sertoli cells (10). De-spite these clear effects of FSH on Sertoli cell function in vitro,however, FSH �-subunit knockout (FSH�KO) mice and FSHreceptor knockout (FSHRKO) mice are fertile, albeit with areduced germ cell number and sperm quality (11–13). Anumber of studies have shown that FSH acts to regulateSertoli cell number in vivo (14–18), which would explain inpart the reduced germ cell number in the knockout mousemodels, but also raises questions about the definitive role ofFSH in regulating Sertoli cell function in vivo.

Early studies of the regulation of testicular function alsosuggested that in addition to FSH, testosterone could act tomaintain spermatogenesis in hypophysectomized rats (19).This has subsequently been confirmed by a number of stud-ies and is clearly illustrated by the failure of spermatogenesisin Tfm mutant mice, which lack functional androgen recep-tors (20), and in hypogonadal (hpg) mice, which show fullspermatogenesis after testosterone treatment (18, 21). Theexperimental chimera studies by Lyon et al. (22) and the factthat germ cells do not express the androgen receptor gene(23) provide strong evidence that androgenic control of sper-matogenesis must be mediated by Sertoli cells. Androgenshave been shown to affect Sertoli cell function in culture (24),although these effects are generally less marked than the

Abbreviations: ABP, Androgen-binding protein; AMH, anti-Mulle-rian hormone; Cys-TE, cystatin-TE; Dhh, desert hedgehog; FABP, fattyacid-binding protein; FSHR, FSH receptor; KO, knockout; PDGF-A,platelet-derived growth factor-A; tPA, tissue plasminogen activator;ZO-1, zonula occludens 1.Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving theendocrine community.

0013-7227/04/$15.00/0 Endocrinology 145(1):318–329Printed in U.S.A. Copyright © 2004 by The Endocrine Society

doi: 10.1210/en.2003-1055

318

effects of FSH, and in many studies androgens have not hadany direct effect in vitro. Thus, although a role for androgensin regulating spermatogenesis through the Sertoli cell isclear, the mechanism by which this is achieved remains to befully elucidated.

Studies into the roles of FSH and androgen in the regu-lation of Sertoli cell function have come largely through workwith isolated, cultured Sertoli cells in vitro or indirectlythrough a measurement of spermatogenesis. To examine di-rectly the role of FSH and androgen in the development,regulation, and function of Sertoli cells, we have measuredSertoli cell number and gene expression profiles in FSH�KOand FSHRKO mice, in which FSH activation of the Sertoli celldoes not occur, and in Tfm mice, which lack a functionalandrogen receptor.

Materials and MethodsAnimals

The generation of both FSH�KO and FSHRKO mice has been de-scribed previously (11, 13). Both colonies were on a mixed C57BL6/129background and were maintained at the University of Oxford underUnited Kingdom Home Office regulations. As both FSHRKO andFSH�KO adult males are fertile, the colonies were maintained by breed-ing homozygous males with heterozygous females. Heterozygous maleswere used as controls for these mutants in this study. Tfm mice were bredon a C3H/HeH-101/H genetic background from stock animals obtainedoriginally from the Medical Research Council Radiobiology Unit (nowthe Medical Research Council, Mammalian Genetics Unit, Harwell, UK).The testes of normal mice undergo final descent to the scrotum on about25 d, whereas in Tfm mice the testes remain intraabdominal into adultlife. To control for the failure of testicular descent in Tfm mice, normalanimals on the same background were surgically rendered cryptorchidat 21 d and used for experiments when adults (25). Studies of Sertoli cellgene expression in normal mice at different developmental stages usedanimals on a C3H/HeH-101/H background. In all experiments the dayof birth was designated d 1, and animals were killed between d 1 andadulthood (d 90–180) as indicated in the text. The testes from eachanimal were either frozen in liquid N2 for subsequent study of specific

mRNA levels or fixed overnight in Bouin’s fluid and stored in 70%ethanol for subsequent measurement of cell number.

Stereology

For stereology, testes were fixed in Bouin’s fluid and stored in 70%ethanol. Tissue was embedded in Technovit 7100 resin, cut into sections(20 �m thickness), and stained with Harris’ hematoxylin. The total testisvolume was estimated using the Cavalieri principle (26), and the slidesused to estimate the number of cells were also used to estimate testisvolume. The optical disector technique (27) was used to count the num-ber of Sertoli cells in each testis. Sertoli cells were identified by theirdistinctive nucleus and position on the periphery of the tubule (28). Thenumerical density of Sertoli cells was estimated using an Olympus BX50microscope fitted with a motorized stage (Prior Scientific Instruments,Cambridge, UK) and Stereologer software (Systems Planning Analysis,Alexandria, VA).

Measurement of mRNA levels

For quantification of the content of specific mRNA species in testesduring development, a real-time PCR approach was used that employedthe TaqMan PCR method after RT of the isolated RNA (29). To allowspecific mRNA levels to be expressed per testis and to control for theefficiency of RNA extraction, RNA degradation, and the RT step, anexternal standard was used (30, 31). The external standard was luciferasemRNA (Promega UK, Southampton, UK), and 5 ng were added to eachtestis at the start of the RNA extraction procedure. Testis RNA wasextracted using TRIzol (Life Technologies, Paisley, UK), and residualgenomic DNA was removed by deoxyribonuclease treatment (DNA-free,Ambion, Inc., supplied by AMS Biotechnology, Abingdon, UK). TheRNA was reverse transcribed using random hexamers and Moloneymurine leukemia virus reverse transcriptase (Superscript II, Life Tech-nologies, Paisley, UK) as described previously (32, 33).

The sequences of the primers and probes used for real-time PCR wereeither previously described (34) or are shown in Table 1. The membrane-bound kit-ligand (KLm) transcript is generated by alternate splicing,which removes exon 6 of the primary KL transcript. The reverse primerused in the real-time studies spans the KLm-specific exon/exon bound-ary. The FSHR primer/probe set was designed to hybridize across exons9 and 10 of the gene and will, therefore, only detect transcripts con-taining those exons (33).

TABLE 1. Sequences of primers/probes used in this study

Gene GenBank Primer/probe Sequence

Claudin-11 NM�008770 F CGTCATGGCCACTGGTCTCTR GGCTCTACAAGCCTGCACGTAP CCCCTGGTGGACATCCTCATCCTTC

FSHR NM�013523 F GGCGGCAAACCTCTGAACTR CCAGGCTGAGTCATATCATCAATATCP CATCCAATTTGCAACAAGTCTATTTCAAGGCA

Dhh NM�007857 F CCACGTATCGGTCAAAGCTGAR ATTTCCCGGAAAGCAGCC TP CTCACTGGCGGTCCGAGCCG

KLm (kit ligand, membrane-bound) NM�013598 F GCCGGCAATGCCATGR AGGTCCCGAGAAAGGGAAAP CTGTCAATTGTAGGCCCGAGTCTTCA

AMH NM�007445 F CCACCCTGAGCATCGATCAR GTCATCCGCGTGAAACAGCP CTCCAAGCCTTTCTATTTGGCTCTGATTCC

PDGF-A M29464 F GCTCGAAGTCAGATCCACAGCR CCTCAGCCCCTACGGAGTCTP CCGGGACCTCCAGCGACTCTTG

ZO-1 NM�009386 F GGAGCTACGCTTGCCACACTR GGTCAATCAGGACAGAAACACAGTP TCCGAATTACCTTGTTGGAGCC

Luciferasea F TCGAAGTATTCCGCGTACGTGR GCCCTGGTTCCTGGAACAAP TGTTCACCTCGATATGTGCATCTGTAAAAGCA

a Derived from sequence of mRNA provided by Promega (Madison, WI). F, Forward; R, reverse; P, probe.

Johnston et al. • Development of Sertoli Cell Function Endocrinology, January 2004, 145(1):318–329 319

The real-time PCRs were carried out in a 25-�l volume using a 96-wellplate format. Components for real-time PCR were purchased from PEApplied Biosystems (Warrington, UK), apart from primers and probesthat were obtained from MWG Biotech (Milton Keynes, UK). Each PCRwell contained reaction buffer (with passive reference), 5 mm MgCl2, 200�m deoxy-NTPs, 300 nm of each primer, 200 nm probe, and 0.02 U/�lenzyme (AmpliTaq Gold, Warrington, UK). Reactions were carried out,and fluorescence was detected on a GeneAmp 5700 system (PE AppliedBiosystems). For each sample a replicate was run, omitting the RT step,and a template negative control was run for each primer/probe com-bination. The quantity of each measured cDNA was then expressedrelative to the internal standard luciferase cDNA in the same sample.This method allows direct comparison of expression levels per testisbetween different samples (30). To determine mRNA levels per Sertolicell (see Fig. 4), values were divided by the total Sertoli cell number.

Statistical analysis

Differences between groups at each age were determined by a single-factor ANOVA, followed by Fisher’s post hoc test. To examine the effectsof age on Sertoli cell number, a single-factor ANOVA was carried outwithin each group, followed by Fisher’s post hoc test. Where there was

significant heterogeneity of variance, data were log-transformed beforeanalysis.

ResultsTestis morphology

On d 5, seminiferous tubules from normal animals con-tained mainly Sertoli cells and gonocytes, which were foundin the lumen of the tubule and on the basement membrane(Fig. 1A). Testis morphology in FSHRKO and Tfm mice on d5 was similar to that in controls (Fig. 1, B and C), althoughTfm testes were significantly smaller (1.97 � 0.1 vs. 1.0 � 0.17mm3; P � 0.05). By d 20, the tubules had enlarged in thenormal animal and contained mainly Sertoli cells, with sper-matogonia and spermatocytes undergoing proliferation (Fig.1D). In the FSHRKO mouse (Fig. 1E) at 20 d, overall testismorphology was very similar to normal, although testis sizewas significantly reduced (24.3 � 0.4 vs. 7.6 � 0.9 mm3; P �0.01). In 20-d-old Tfm mice, overall tubule diameter was

FIG. 1. Light micrographs showing normal, FSHRKO, and Tfm testes at d 5, d 20, and adulthood and cryptorchid testes from adult animals.The top row of light micrographs (A–C) shows morphology in d 5 animals, the second row (D–F) in d 20 animals, and the third row (G–I) inadult animals. A section of testis from a normal adult animal, surgically rendered cryptorchid before puberty, is shown in the bottom row (J).The arrow in I points to an intratubular cell mass, which was present in about 30% of tubules from adult Tfm mice. The bar shown represents50 �m in all micrographs.

320 Endocrinology, January 2004, 145(1):318–329 Johnston et al. • Development of Sertoli Cell Function

similar to normal, but with a higher variation (Fig. 1, D vs.F). Tubules in d 20 Tfm mice lacked a lumen and did not showthe full range of spermatogenic cell division seen in normalmales at this age (Fig. 1F). Instead, scattered spermatocytesundergoing division were seen within each tubule. Testes in20-d-old Tfm mice were significantly smaller than those incontrols (24.3 � 0.4 vs. 4.3 � 0.3 mm3; P � 0.01). In adultnormal and FSHRKO mice, all stages of spermatogenesiswere seen as previously described (13), although tubule di-ameter was reduced in FSHRKO animals (Fig. 1, G and H).In adult Tfm mice, tubular morphology was similar to that at20 d of age, with spermatogenesis not progressing beyondthe early spermatocyte stage (Fig. 1I). A similar disruption tospermatogenesis was seen in normal mice with experimentalcryptorchidism induced before testicular descent (Fig. 1J). Inabout one third of the tubules in adult Tfm mice, clumps ofcells with an indistinct morphology were seen in the centerof the tubule (arrowed in Fig. 1I). At all ages, the morphologyof testes from FSH�KO mice was indistinguishable from thatof FSHRKO mice (data not shown).

Sertoli cell number

In normal mice, Sertoli cell number increased 6-fold fromd 1 to d 20 and then remained static into adulthood (Fig. 2).Sertoli cell number in FSHRKO and FSH�KO animals werenormal on d 1 and 5. On d 20, Sertoli cell number was reducedin FSH�KO animals relative to that in control mice and didnot differ significantly from that in FSHRKO mice. In theadult animal, Sertoli cell numbers in both FSHRKO andFSH�KO mice were reduced to about 60% of normal. Sta-tistical analysis of the developmental changes in cell num-bers in these groups showed that cell number in FSH�KOmice reached a peak on d 5 and did not change thereafter(Fig. 2). FSHRKO mice tended to show more variation, butat no time did they differ significantly from FSH�KO mice.

Sertoli cell numbers in Tfm mice were significantly lessthan control values on d 1 and remained lower than all othergroups at each age measured (Fig. 1). Cell number peaked ond 20 in Tfm mice and then declined significantly to about 30%of control levels in adulthood (Fig. 2). There was no differ-ence in Sertoli cell number between normal control animalsand animals rendered cryptorchid before puberty (Fig. 2).

Developmental changes in Sertoli cell gene expression innormal mice

Real-time PCR was used to examine changes in Sertoli cellgene expression during postnatal development in normalmice (Fig. 3). Expression was measured relative to an externalstandard (luciferase) added to each tissue before RNA ex-traction. The results, therefore, indicate the relative geneexpression per testis. The 11 mRNA species examined in thisstudy were chosen because they have been shown to beexpressed in Sertoli cells and because, within the testis, theyshow little or no expression in any other cell type (35–45).The mRNA species studied can be broadly divided into fourfunctional groups: 1) tight junction components [zonula oc-cludens 1 (ZO-1) and claudin 11], 2) tissue remodeling factors[cystatin-TE (Cys-TE) and tissue plasminogen activator(tPA)], 3) lipid-binding proteins [epidermal fatty acid-

binding protein (FABP) and androgen-binding protein(ABP)], and 4) endocrine and paracrine factors [KLm,platelet-derived growth factor-A (PDGF-A), anti-Mullerianhormone (AMH), desert hedgehog (Dhh), and FSHR]. Of the11 transcript species measured in this study, 10 showed ageneral increase in expression from birth to adulthood, withCys-TE showing the highest overall expression at all ages.The one exception to this pattern was AMH, which showedno change in expression from birth to around d 10, beforedeclining significantly to adulthood. In the 10 transcriptsshowing an increase in expression after birth, the most con-sistent change was a significant increase in expression on d10, which was seen in all cases. After 10 d, expression pertestis either remained fairly static (ABP, Dhh, and tPA) orincreased further into adulthood (FABP, KLm, PDGF-A,claudin 11, ZO-1, and Cys-TE).

Sertoli cell gene expression in FSH�KO, FSHRKO, andTfm mice

To determine the roles of FSH and androgen in the de-velopment of Sertoli cell function, the expression of Sertolicell-specific genes was measured during postnatal develop-ment in normal, FSH�KO, FSHRKO, and Tfm mice. The

FIG. 2. Sertoli cell number in control, FSHRKO, FSH�KO, Tfm, andcryptorchid mice. Cell numbers in each null group and in appropriatecontrols were measured as described in Materials and Methods. Two-factor ANOVA showed that there was no difference between differentcontrol groups, and control values were, therefore, pooled. Resultsshow the mean � SEM for between three and 12 animals in each group.Overall, two-factor analysis of the data showed that there were sig-nificant age and animal group effects and that there was a significantinteraction between the two groups. Further single-factor ANOVAwas carried out within each age and within each animal group. Theresults of analysis within each age are shown in the figure, andcolumns within each age that do not share a common letter superscriptare significantly different. Further analysis within each group wascarried out to examine changes in cell number during development,and results are summarized in the table below. Within each group,ages that do not share a common letter superscript are significantlydifferent.

GroupAge

d 1 d 5 d 20 Adult

Control a b c c

FSHRKO a b,c c b

FSH�KO a b b b

Tfm a b c b


results were corrected for differences in Sertoli cell numberby dividing measured mRNA levels calculated per testis (asdescribed above) by the number of Sertoli cells in each group(Fig. 4).

In FSH�KO and FSHRKO mice, Sertoli cell RNA transcriptlevels were largely normal on d 5 after birth. The exceptionswere ABP, which was significantly reduced by about 50% inFSHRKO mice, and FSHR, which, as expected, was absent inFSHRKO mice. On d 20, in contrast, there was a significantreduction in ZO-1, PDGF-A, KLm, Cys-TE, FABP, and ABPexpression in both FSH�KO and FSHRKO mice. In addition,there was a reduction in claudin 11 expression, although this

was only significant in FSH�KO mice. On d 5, FSHR expressionwas absent in FSHRKO mice. In affected genes, the reductionin expression was between 50 and 80%. No change in Dhh,FSHR (in FSH�KO mice), AMH, or tPA expression was seen.In adult animals, levels of expression of ZO-1, claudin 11,PDGF-A, and Cys-TE remained significantly reduced, and tPAexpression was also reduced. In contrast to that in 20-d-oldanimals, KLm, FABP, and ABP expression was not differentfrom normal in adult FSHR-KO or FSH�-KO animals. Theexpression of Dhh, AMH, and FSHR (in FSH�KO mice) re-mained normal throughout postnatal development. Interest-ingly, FSHR expression was detectable at low levels in adult

FIG. 3. Changes in Sertoli cell mRNAlevels during development in the nor-mal mouse. RNA was extracted fromtestes at different ages, and cDNA wasprepared as described in Materials andMethods. Real-time PCR was used tomeasure cDNA levels relative to an ex-ternal standard (luciferase) added dur-ing RNA extraction. The mean � SEM ofbetween three and six animals at eachage are shown. Groups that do not sharethe same letter superscript are signifi-cantly different.


FSHRKO mice using the real-time primer/probe combinationthat spans exons 9 and 10 of the cDNA.

In Tfm mice, levels of the mRNA species measured in thisstudy were normal per Sertoli cell on d 5, with the exceptionsof tPA and FABP, which were significantly reduced. On d 20,levels of PDGF-A, Cys-TE, tPA, and FABP were reducedsignificantly in Tfm mice, whereas other mRNA species wereunaffected. The effects of the Tfm mutation on Sertoli cellfunction in the adult animal are more complex because adultTfm mice are cryptorchid. Induced cryptorchidism per se innormal animals had no effect on the expression of most of themRNA species measured with the exception of ZO-1,PDGF-A, and FABP, in which expression was significantly

reduced, and AMH, in which there was a significant increasein expression. Using the cryptorchid animal as the appro-priate control, the Tfm mutation had little effect on geneexpression in the adult mouse; the only exception was anincrease in KLm mRNA levels.

DiscussionSertoli cell number

Sertoli cells proliferate in the mouse through fetal andneonatal development to reach a maximum about 15–20 dpostpartum (28, 46). In mice lacking FSH stimulation, thenormal Sertoli cell number at birth and during the early

FIG. 3. Continued.


postnatal period indicates that FSH is not required for thedetermination of Sertoli cell number during the fetal andneonatal periods and only becomes necessary during thefinal establishment of Sertoli cell number. Other studieshave reported, in contrast, that late fetal Sertoli cell pro-liferation in the rat is regulated by FSH (16, 47). Thisdiscrepancy may be due to species differences or to dif-ferences in techniques used. Earlier studies altered Sertolicell FSH stimulation through fetal decapitation, use of

organ culture, or treatment with FSH antibodies; Sertolicell labeling, rather than cell number, was measured as anend point. It is possible that in these earlier studies tes-ticular androgen levels may have been affected by the invivo treatments, although the labeling index may not nec-essarily reflect final cell numbers. It is worth noting thatthe results of this study (see below) and others (48) haveindicated that fetal Sertoli cell function is largely inde-pendent of FSH stimulation. In the adult, the difference in

FIG. 4. Sertoli cell mRNA levels duringdevelopment in normal, FSH�KO,FSHRKO, and Tfm (AR-null) mice.RNA was extracted from testes at dif-ferent ages, and cDNA was prepared asdescribed in Materials and Methods.Real-time PCR was used to measurecDNA levels relative to an externalstandard (luciferase) added duringRNA extraction. Two-factor ANOVAwas used to determine whether therewas a significant difference in specificmRNA levels between each of the con-trol groups during development. A,Control groups were not significantlydifferent, and data were analyzed withpooled control data at each age as de-scribed in Materials and Methods.Within each age, groups not sharing thesame letter superscript were signifi-cantly different. B, There was a signif-icant difference between control groups.These control groups have been pooledfor illustrative purposes, but data wereanalyzed without pooling as describedin Materials and Methods. Groups withan asterisk were significantly differentfrom the appropriate control group atthat age. The mean � SEM of betweenthree and 23 animals at each age areshown.


Sertoli cell number between control and FSH�KO animalsis similar to that reported previously (17).

In contrast to animals lacking FSH stimulation, androgensappear to be essential for Sertoli cell proliferation during fetallife and throughout the postnatal, prepubertal period. Inter-estingly, Sertoli cell number in adult Tfm mice was signifi-cantly lower than on d 20, suggesting that androgens may berequired to maintain Sertoli cell number in the adult animal.In this study Sertoli cells are defined in part by their positionon the periphery of the seminiferous tubules. We have notedthat in sections of adult Tfm testes some tubules containedclumps of cells within the center of the tubule, similar tothose is seen in KL-deficient and claudin 11-KO mice (49, 50).These have been reported to contain Sertoli cells that have

become detached from the basement membrane (6, 50), andas these cells were not included in the Sertoli cell countsreported here, they may account for the apparent decrease incell number in the adult. If these cell clumps contain func-tional Sertoli cells, then gene expression levels, measured perSertoli cell in the adult Tfm mouse, will have been slightlyoverestimated.

The hpg mouse is an alternative and useful model for thestudy of Sertoli cell development because it lacks GnRHthrough a natural deletion leading to a loss of circulating LHand FSH (51). Comparison between hpg mice and those lack-ing FSH stimulation or androgen receptors can be complex,because adult hpg mice have severely depleted levels of bothFSH and testosterone, although neither hormone is com-

FIG. 4. Continued.


pletely absent. Sertoli cell number is normal during fetaldevelopment in hpg mice (28), which would support thehypothesis that FSH does not regulate fetal Sertoli cell num-ber, as androgen levels are normal during this stage of hpgdevelopment (52). After birth, Sertoli cell number is reducedduring the early postnatal period in the hpg mouse (28) withnumbers similar to those seen in Tfm mice in this study. Asignificant reduction in androgen levels occurs around birthin hpg mice (52), suggesting that the early postnatal reductionin Sertoli cell number is caused by androgen withdrawal.Treatment of hpg mice with FSH or androgen will increaseadult Sertoli cell number (18), which is consistent with theresults reported here.

Sertoli cell gene expression during normal development

Sertoli cell activity in this study has been assessed bydetailed measurement of a limited number of mRNA speciesthat provides a restricted picture of gene expression in theSertoli cell during development, but will indicate, neverthe-less, general trends in cell activity.

With the exception of AMH, the clearest developmentalchange in normal Sertoli cell gene expression (per testis)occurred between d 5 and 10. Sertoli cell number increasesduring this period, but in most cases the increase in cellnumber would not account for the increase in gene expres-sion, suggesting an overall increase in Sertoli cell activity.This is also the period during which changes in the expres-sion of at least some other Sertoli cell genes is first seen (53),suggesting that there is an overall change in Sertoli cell ac-tivity at this time. The interval between postnatal d 5 and 10is crucial in mouse testis development. During this periodspermatagonial proliferation and meiosis begin (54), adultLeydig cells first start to differentiate (55, 56), and the blood-testis barrier begins to form (57). The Sertoli cell is known toplay a central role in regulating testis function (6), and theseearly postnatal developments in the testis may all be relatedto changes in Sertoli cell function. Between d 5 and 10 thereis little change in circulating levels of FSH (58, 59) (alsoconfirmed in our own studies; data not shown) and a dropin intratesticular androgen (60). This suggests that Sertoli cellsensitivity to hormone stimulation increases or another, asyet unknown, factor is involved in Sertoli cell regulation atthis time.

Sertoli cell gene expression in FSHRKO, FSH�KO, andTfm mice

Tight junction components. Tight junctions between Sertolicells are the main functional component of the blood-testisbarrier. Both ZO1 and claudin-11 are integral to the structureof these tight junctions (61), and claudin-11-KO mice lacktight junctions between Sertoli cells, leading to infertility (50).The significant reduction in levels of both ZO1 and clau-din-11 transcripts in mice lacking FSH stimulation on d 20indicates that FSH is involved in the regulation of thesegenes. Sertoli cell tight junctions may, therefore, be compro-mised in these animals, and this may provide a partial ex-planation of the reported reduction in sperm quality in thesemutants. In the Tfm mutants, there was little difference inclaudin 11 expression at any age during postnatal develop-

ment. Similarly, levels of ZO-1 were normal up to 20 d,although there was a dramatic decline in expression in bothadult normal cryptorchid mice and adult Tfm mice. As adultTfm mice are cryptorchid, it appears that androgen resistancedoes not affect ZO-1 expression beyond the changes inducedby cryptorchidism itself. Other studies have shown that tightjunctions are abnormal in the adult Tfm mouse (62), andcryptorchidism has been reported to be associated with de-fective development of the blood-testis barrier (63).

Tissue remodeling factors. Cys-TE is a member of the cystatinsuperfamily of cysteine proteinases, and it may have a num-ber of functions in the testis, including germ cell-Sertoli cellinteractions (43). Overall testicular expression of Cys-TE wasmore than 10 times greater than that of any other transcriptmeasured in this study, suggesting an important role inSertoli cell function. Reduced levels of Cys-TE in d 20 andadult mice lacking FSH stimulation indicate a role for FSHin regulating this gene. In contrast, the other tissue remod-eling factor, tPA, was not significantly affected by lack of FSHstimulation at any age. In Tfm mice, levels of both Cys-TE andtPA were significantly reduced on d 20, but were normal inthe adult animal. This suggests that there is a transient periodaround puberty during which these transcripts require an-drogen stimulation, but that they become refractory to theeffects of androgens in the adult animal.

Lipid-binding proteins. Both FABP and ABP were reduced ond 20 in FSH�KO and FSHRKO mice, but were normal inadult animals, showing that FSH is required during the pu-bertal period for normal expression, but that expression be-comes independent in the adult animal. FABP is involved inthe uptake and transport of fatty acids essential for the nour-ishment of the surrounding cell types (42) and may play arole in germ cell maturation. It is possible, therefore, that inthe adult animal control of FABP expression becomes de-pendent primarily on Sertoli cell-germ cell interaction asspermatogenesis becomes established. This would be con-sistent with continued reduced levels of expression of FABPin normal cryptorchid and Tfm adult mice. The role of ABPin the testis is uncertain, but it may be involved in regulatinggerm cell apoptosis during the prepubertal period (64). Re-duced levels of ABP during this period may, therefore, alsocontribute to reduced germ cell number in the adult FSHRKOand FSH�KO mice.

Endocrine and paracrine factors. The growth factors Dhh andPDGF-A are required during testicular development for nor-mal Leydig cell differentiation (41, 65). The expression ofDhh was unaffected by the mutations used in this study,whereas, in contrast, PDGF-A transcript levels were reducedin animals lacking FSH stimulation. This contrasts with ear-lier studies in the rat that suggested that PDGF-A expressionis inhibited by FSH (66). It is likely that this apparent dis-crepancy is due to species differences in the expression andcontrol of PDGF-A (67). Interestingly, cryptorchidism (andthe Tfm mutation) caused a marked reduction in PDGF-Aexpression, suggesting that adult expression is also depen-dent on germ cell maturation. This would be consistent withdemonstrated stage-dependent expression in the mouse tes-tis (41).


The membrane-bound form of Kit ligand has been shownto be necessary for the maintenance of differentiated germcells and for their entry into and/or completion of meiosis(68). The significantly lower levels of KLm transcripts on d20, when meiosis is beginning, in mice lacking FSH stimu-lation indicates that FSH regulates the expression of this genein prepubertal mice. Normal expression of KLm in the adultanimal shows, however, that this regulation is transient andis not required once spermatogenesis is established.

The FSHR is known to be down-regulated in the testis byexposure to high circulating levels of FSH (69, 70). In theresults reported here, however, the lack of circulating FSHdid not affect the expression of FSHR, suggesting that withinthe normal physiological range, FSHR expression is not reg-ulated by FSH. In adult mice lacking androgen receptors,there was a reduction in FSHR expression, indicating thatandrogens can play a role in regulating FSHR in the adultanimal. It is possible, therefore, that one of the trophic effectsof androgen on the Sertoli cell is to increase sensitivityto FSH.

AMH is produced by Sertoli cells from early fetal life topuberty (71). At puberty, synthesis is down-regulated, pos-sibly through the synergistic action of intratesticular testos-terone and meiotic germ cells (71). In adult mice, there wasa small increase in AMH mRNA levels in cryptorchid andTfm mice compared with normal controls, which would beconsistent with this hypothesis, although it is clear that theoverall drop in AMH expression and AMH serum levels (72)is not prevented by androgen insensitivity. Normal levels ofAMH expression in mice lacking FSH indicate that FSH is notrequired for AMH expression.

Control of Sertoli cell activity

Overall, little difference in Sertoli cell activity was seen ond 5 between animals lacking FSH stimulation and normalmice of the same age. Receptors for FSH are expressed earlyin fetal testis development (73), but our results indicate thatfetal and early neonatal Sertoli cell development is largelyindependent of FSH control. This supports earlier studiesusing cultured rat Sertoli cells (48) and is consistent withnormal Sertoli cell numbers in animals lacking FSH stimu-lation. By d 20, the expression of seven genes was reducedin FSHRKO and FSH�KO animals. This suggests that the risein Sertoli cell activity seen around d 10 is at least partly drivenby FSH. In the adult animal dependence on FSH was largelyretained, with five of the 11 genes investigated being signif-icantly reduced compared with normal mice. Despite theseclear changes in gene expression in mice lacking FSH stim-ulation, it should be noted that the spermatogenic process isessentially normal in these animals, although there is anoverall reduction in spermatid number per Sertoli cell (17).Thus, although FSH clearly regulates the expression of anumber of genes in the Sertoli cell, the reduction in expres-sion in the absence of FSH stimulation is not sufficient tocompromise, fatally, the spermatogenic process. The re-duced spermatid number in these animals may, however, bea reflection of reduced Sertoli cell function.

In contrast to the relatively normal spermatogenesis inmice lacking FSH stimulation, mice lacking functional an-

drogen receptors show severe disruption to spermatogenesisboth before puberty and in the adult animal. The failure ofnormal germ cell proliferation and differentiation in prepu-bertal, 20-d-old Tfm mice shows that androgens are essentialfor this process and, as androgen receptors are present inSertoli cells but not germ cells (22, 23), this indicates thatSertoli cell function is disrupted during the prepubertal pe-riod. Gene expression levels per Sertoli cell on d 5 in Tfm micewere similar to those in normal mice with the exceptions ofFABP and tPA. This would indicate that, unlike cell prolif-eration, Sertoli cell activity is largely independent of andro-gen action up to d 5. By d 20, the expression levels of fourmRNA transcripts were reduced in the Tfm testis, showingthat androgens as well as FSH have specific trophic effects onthe Sertoli cells at this age. Although both hormones can thusalter Sertoli cell activity around puberty, the considerablygreater level of spermatogenic disruption in the androgen-insensitive mice suggests that androgens must regulate anessential subset of Sertoli cell genes critical for the spermat-ogenic process.

Analysis of Sertoli cell function in adult Tfm mice is com-plicated by the failure of normal testicular descent in theseanimals. In normal animals rendered surgically cryptorchidat puberty, gene expression was altered in four of the 11transcripts measured in this study. The mechanism involvedin this effect is likely to be linked to increased temperaturein the cryptorchid testis or, possibly, to germ cell loss (74). Inadult Tfm mice, only one transcript was significantly differ-ent from the cryptorchid controls. It is possible that some ofthose genes that are sensitive to cryptorchidism are alsodependent on androgen for full expression, but our studiessuggest, nevertheless, that androgens have only a limitedeffect on Sertoli cell gene expression in the adult animal.

Overall, our results using FSH�KO, FSHRKO, and Tfmmice show that both FSH and androgen are required fordevelopment of the full complement of Sertoli cells in theadult male, although only androgens are crucial during fetaland neonatal development. Sertoli cell activity increasesaround d 10 in the mouse, possibly reflecting an increase insensitivity to FSH at this time, although some genes appearto function independently of FSH action through develop-ment. Androgens also appear to be important in maintainingSertoli cell function in the pubertal period, although require-ments for androgen are less clear in the adult, and the effectsof androgen may be mediated through a limited set of genesnot measured here.

Acknowledgments

Received August 14, 2003. Accepted September 29, 2003.Address all correspondence and requests for reprints to: Prof. P. J.

O’Shaughnessy, Institute of Comparative Medicine, University of Glas-gow Veterinary School, Bearsden Road, Glasgow, United Kingdom G611QH. E-mail: p.j.o’[email protected].

This work was supported by funding from the Wellcome Trust, theBiotechnology and Biological Sciences Research Council, and The MoranFoundation (Department of Pathology, Baylor College of Medicine).

References

1. Palmer SJ, Burgoyne PS 1991 In situ analysis of fetal, prepuberal and adultXX/Xy chimeric mouse testes: Sertoli cells are predominantly, but not exclu-sively, XY. Development 112:265–268


2. Albrecht KH, Eicher EM 2001 Evidence that Sry is expressed in pre-Sertoli cellsand Sertoli and granulosa cells have a common precursor. Dev Biol 240:92–107

3. McLaren A 2000 Germ and somatic cell lineages in the developing gonad. MolCell Endocrinol 163:3–9

4. Orth JM, Gunsalus GL, Lamperti AA 1988 Evidence from Sertoli cell-depletedrats indicates that spermatid number in adults depends on numbers of Sertolicells produced during perinatal development. Endocrinology 122:787–794

5. Baker PJ, Pakarinen P, Huhtaniemi IT, Abel MH, Charlton HM, Kumar TR,O’Shaughnessy PJ 2003 Failure of normal leydig cell development in follicle-stimulating hormone (FSH) receptor-deficient mice, but not FSH�-deficientmice: role for constitutive FSH receptor activity. Endocrinology 144:138–145

6. Russell LD, Warren J, Debeljuk L, Richardson LL, Mahar PL, Waymire KG,Amy SP, Ross AJ, MacGregor GR 2001 Spermatogenesis in Bclw-deficientmice. Biol Reprod 65:318–332

7. Greep RO, Fevold H 1937 The spermatogenic and secretory function of thegonads of hypophysectomised adult rats treated with pituitary FSH and LH.Endocrinology 21:611–619

8. Dorrington JH, Roller NF, Fritz IB 1975 Effects of follicle-stimulating hormoneon cultures of Sertoli cell preparations. Mol Cell Endocrinol 3:57–70

9. Fritz IB, Rommerts FG, Louis BG, Dorrington JH 1976 Regulation by FSH anddibutyryl cyclic AMP of the formation of androgen-binding protein in Sertolicell-enriched cultures. J Reprod Fertil 46:17–24

10. McLean DJ, Friel PJ, Pouchnik D, Griswold MD 2002 Oligonucleotide mi-croarray analysis of gene expression in follicle-stimulating hormone-treatedrat Sertoli cells. Mol Endocrinol 16:2780–2792

11. Kumar TR, Wang Y, Lu N, Matzuk MM 1997 Follicle stimulating hormone isrequired for ovarian follicle maturation but not male fertility. Nat Genet15:201–204

12. Dierich A, Sairam MR, Monaco L, Fimia GM, Gansmuller A, LeMeur M,Sassone-Corsi P 1998 Impairing follicle-stimulating hormone (FSH) signalingin vivo: targeted disruption of the FSH receptor leads to aberrant gametogenesisand hormonal imbalance. Proc Natl Acad Sci USA 95:13612–13617

13. Abel MH, Wootton AN, Wilkins V, Huhtaniemi I, Knight PG, Charlton HM2000 The effect of a null mutation in the follicle-stimulating hormone receptorgene on mouse reproduction. Endocrinology 141:1795–1803

14. Atanassova N, McKinnell C, Walker M, Turner KJ, Fisher JS, Morley M,Millar MR, Groome NP, Sharpe RM 1999 Permanent effects of neonatalestrogen exposure in rats on reproductive hormone levels, Sertoli cell number,and the efficiency of spermatogenesis in adulthood. Endocrinology 140:5364–5373

15. Sharpe RM, Walker M, Millar MR, Atanassova N, Morris K, McKinnell C,Saunders PT, Fraser HM 2000 Effect of neonatal gonadotropin-releasing hor-mone antagonist administration on Sertoli cell number and testicular devel-opment in the marmoset: comparison with the rat. Biol Reprod 62:1685–1693

16. Orth JM 1984 The role of follicle-stimulating hormone in controlling Sertoli cellproliferation in testes of fetal rats. Endocinology 115:1248–1255

17. Wreford NG, Rajendra Kumar T, Matzuk MM, de Kretser DM 2001 Analysisof the testicular phenotype of the follicle-stimulating hormone �-subunitknockout and the activin type II receptor knockout mice by stereologicalanalysis. Endocrinology 142:2916–2920

18. Haywood M, Spaliviero J, Jimemez M, King NJ, Handelsman DJ, Allan CM2003 Sertoli and germ cell development in hypogonadal (hpg) mice expressingtransgenic follicle-stimulating hormone alone or in combination with testos-terone. Endocrinology 144:509–517

19. Walsh EL, Cuyler WK, McCullagh DR 1934 The physiologic maintenance ofmale sex glands: the effect of androtin on hypophysectomized rats. Am JPhysiol 107:508–518

20. Lyon MF, Hawkes SG 1970 X-linked gene for testicular feminization in themouse. Nature 227:1217–1219

21. Singh J, Oneill C, Handelsman DJ 1995 Induction of spermatogenesis by an-drogens in gonadotropin-deficient (Hpg) mice. Endocrinology 136:5311–5321

22. Lyon MF, Glenister PH, Lamoreux ML 1975 Normal spermatozoa from an-drogen-resistant germ cells of chimaeric mice and the role of androgen inspermatogenesis. Nature 258:620–622

23. Zhou Q, Nie R, Prins GS, Saunders PT, Katzenellenbogen BS, Hess RA 2002Localization of androgen and estrogen receptors in adult male mouse repro-ductive tract. J Androl 23:870–881

24. Sar M, Hall SH, Wilson EM, French FS 1993 Androgen regulation of Sertolicells. In: Russell LD, Griswold MD, eds. The Sertoli cell. Clearwater, FL: CacheRiver Press; 509–516

25. Murphy L, O’Shaughnessy PJ 1991 Testicular steroidogenesis in the testicularfeminized (Tfm) mouse: loss of 17�-hydroxylase activity. J Endocrinol 131:443–449

26. Mayhew TM 1992 A review of recent advances in stereology for quantifyingneural structure. J Neurocytol 21:313–328

27. Wreford NG 1995 Theory and practice of stereological techniques applied tothe estimation of cell number and nuclear volume of the testis. Microsc ResTechnol 32:423–436

28. Baker PJ, O’Shaughnessy PJ 2001 Role of gonadotrophins in regulating num-bers of Leydig and Sertoli cells during fetal and postnatal development in mice.Reproduction 122:227–234

29. Bustin SA 2000 Absolute quantification of mRNA using real-time reversetranscription polymerase chain reaction assays. J Mol Endocrinol 25:169–193

30. Baker PJ, O’Shaughnessy PJ 2001 Expression of prostaglandin D synthetaseduring development in the mouse testis. Reproduction 122:553–559

31. O’Shaughnessy PJ, Johnston H, Willerton L, Baker PJ 2002 Failure of normaladult Leydig cell development in androgen-receptor-deficient mice. J Cell Sci115:3491–3496

32. O’Shaughnessy PJ, Murphy L 1993 Cytochrome P-450 17�-hydroxylase pro-tein and mRNA in the testis of the testicular feminized (Tfm) mouse. J MolEndocrinol 11:77–82

33. O’Shaughnessy PJ, Marsh P, Dudley K 1994 Follicle-stimulating hormonereceptor mRNA in the mouse ovary during post-natal development in thenormal mouse and in the adult hypogonadal (hpg) mouse: structure of alternatetranscripts. Mol Cell Endocrinol 101:197–201

34. O’Shaughnessy PJ, Fleming L, Baker PJ, Jackson G, Johnston H 2003 Iden-tification of developmentally-regulated genes in the somatic cells of the mousetestis using serial analysis of gene expression. Biol Reprod 69:747–808

35. Byers S, Graham R, Dai HN, Hoxter B 1991 Development of Sertoli celljunctional specializations and the distribution of the tight-junction-associatedprotein ZO-1 in the mouse testis. Am J Anat 191:35–47

36. Wang. Y-M., Sullivan PM, Petrusz P, Yarbrough W, Joseph DR 1989 Theandrogen-binding protein gene is expressed in CD1 mouse testis. Mol CellEndocrinol 63:85–92

37. Morita K, Sasaki H, Fujimoto K, Furuse M, Tsukita S 1999 Claudin-11/OSP-based tight junctions of myelin sheaths in brain and Sertoli cells in testis. J CellBiol 145:579–588

38. Kliesch S, Penttila TL, Gromoll J, Saunders PTK, Nieschlag E, Parvinen M1992 FSH receptor messenger-RNA is expressed stage-dependently during ratspermatogenesis. Mol Cell Endocrinol 84:R45–R49

39. Munsterberg A, Lovell-Badge R 1991 Expression of the mouse anti-Mullerianhormone gene suggests a role in both male and female sexual differentiation.Development 113:613–624

40. Bitgood MJ, Shen L, McMahon AP 1996 Sertoli cell signaling by deserthedgehog regulates the male germline. Curr Biol 6:298–304

41. Gnessi L, Basciani S, Mariani S, Arizzi M, Spera G, Wang C, Bondjers C,Karlsson L, Betsholtz C 2000 Leydig cell loss and spermatogenic arrest in platelet-derived growth factor (PDGF)-A-deficient mice. J Cell Biol 149:1019–1026

42. Kingma PB, Bok D, Ong DE 1998 Bovine epidermal fatty acid-binding protein:determination of ligand specificity and cellular localization in retina and testis.Biochemistry 37:3250–3257

43. Li Y, Friel PJ, Robinson MO, McLean DJ, Griswold MD 2002 Identificationand characterization of testis- and epididymis-specific genes: cystatin SC andcystatin TE-1. Biol Reprod 67:1872–1880

44. Zhang T, Zhou HM, Liu YX 1997 Expression of plasminogen activator andinhibitor, urokinase receptor and inhibin subunits in rhesus monkey testes.Mol Hum Reprod 3:223–231

45. Manova K, Huang EJ, Angeles M, De L, V, Sanchez S, Pronovost SM, BesmerP, Bachvarova RF 1993 The expression pattern of the c-kit ligand in gonads ofmice supports a role for the c-kit receptor in oocyte growth and in proliferationof spermatogonia. Dev Biol 157:85–99

46. Vergouwen RPFA, Huiskamp R, Bas RJ, Roepers-Gajadien HL, Davids JAG,de Rooij DG 1993 Postnatal development of testicular populations in mice. JReprod Fertil 99:479–485

47. Sasaki M, Yamamoto M, Arishima K, Eguchi Y 2000 Effect of follicle-stim-ulating hormone on Sertoli cell division in cultures of fetal rat testes. BiolNeonate 78:48–52

48. Ueda H, Ziomek C, Hall PF 1988 Isolation and characterization of rat fetalSertoli cells. Endocrinology 123:1014–1022

49. Parreira GG, Ogawa T, Avarbock MR, Franca LR, Brinster RL, Russell LD1998 Development of germ cell transplants in mice. Biol Reprod 59:1360–1370

50. Gow A, Southwood CM, Li JS, Pariali M, Riordan GP, Brodie SE, Danias J,Bronstein JM, Kachar B, Lazzarini RA 1999 CNS myelin and Sertoli cell tightjunction strands are absent in Osp/claudin-11 null mice. Cell 99:649–659

51. Cattanach BM, Iddon CA, Charlton HM, Chiappa SA, Fink G 1977 Go-nadtrophin releasing hormone deficiency in a mutant mouse with hypogo-nadism. Nature 269:338–340

52. O’Shaughnessy PJ, Baker P, Sohnius U, Haavisto A-M, Charlton HM,Huhtaniemi I 1998 Fetal development of Leydig cell activity in the mouse isindependent of pituitary gonadotroph function. Endocrinology 139:1141–1146

53. Lindsey JS, Wilkinson MF 1996 Pem: a testosterone-regulated and LH-regulated homeobox gene expressed in mouse sertoli cells and epididymis.Dev Biol 179:471–484

54. McCarrey JR 1993 Development in the germ cell. In: Desjardins C, Ewing LL,eds. Cell and molecular biology of the testis. Oxford: Oxford University Press;58–89

55. Nef S, Shipman T, Parada LF 2000 A molecular basis for estrogen-inducedcryptorchidism. Dev Biol 224:354–361

56. Baker PJ, Sha JA, McBride MW, Peng L, Payne AH, O’Shaughnessy PJ 1999Expression of 3�-hydroxysteriod dehydrogenase type I and VI isoforms in themouse testis during development. Eur J Biochem 260:911–916

57. Cyr DG, Hermo L, Egenberger N, Mertineit C, Trasler JM, Laird DW 1999


Cellular immunolocalization of occludin during embryonic and postnatal de-velopment of the mouse testis and epididymis. Endocrinology 140:3815–3825

58. Michael SD, Kaplan SB, Macmillan BT 1980 Peripheral plasma concentra-tions of LH, FSH, prolactin and GH from birth to puberty in male and femalemice. J Reprod Fertil 59:217–222

59. Jean-Faucher C, el Watik N, Berger M, De Turckheim M, Veyssiere G, JeanC 1983 Ontogeny of the secretory pattern of LH and FSH in male mice duringsexual maturation. Int J Androl 6:575–584

60. O’Shaughnessy PJ, Willerton L, Baker PJ 2002 Changes in Leydig cell geneexpression during development in the mouse. Biol Reprod 66:966–975

61. Cheng CY, Mruk DD 2002 Cell junction dynamics in the testis: Sertoli-germcell interactions and male contraceptive development. Physiol Rev 82:825–874

62. Lyon MF, Cattanach BM, Charlton HM 1981 In: Austin CR, Edwards RG, eds.Mechanisms of sex differentiation in animals and man. London: AcademicPress; 197–238

63. Pinart E, Sancho S, Briz MD, Bonet S, Garcia N, Badia E 2000 Ultrastructuralstudy of the boar seminiferous epithelium: changes in cryptorchidism. J Mor-phol 244:190–202

64. Jeyaraj DA, Grossman G, Petrusz P 2003 Dynamics of testicular germ cellapoptosis in normal mice and transgenic mice overexpressing rat androgen-binding protein. Reprod Biol Endocrinol 1:48

65. Yao HH, Whoriskey W, Capel B 2002 Desert hedgehog/Patched 1 signalingspecifies fetal Leydig cell fate in testis organogenesis. Genes Dev 16:1433–1440

66. Gnessi L, Emidi A, Jannini EA, Carosa E, Maroder M, Arizzi M, Ulisse S, Spera

G 1995 Testicular development involves the spatiotemporal control of PDGFs andPDGF receptors gene expression and action. J Cell Biol 131:1105–1121

67. Mariani S, Basciani S, Arizzi M, Spera G, Gnessi L 2002 PDGF and the testis.Trends Endocrinol Metab 13:11–17

68. Ohta H, Yomogida K, Dohmae K, Nishimune Y 2000 Regulation of prolif-eration and differentiation in spermatogonial stem cells: the role of c-kit andits ligand SCF. Development 127:2125–2131

69. O’Shaughnessy PJ 1980 FSH receptor autoregulation and cyclic AMP pro-duction in the immature rat testis. Biol Reprod 23:810–814

70. Griswold MD, Kim JS, Tribley WA 2001 Mechanisms involved in the ho-mologous down-regulation of transcription of the follicle-stimulating hor-mone receptor gene in Sertoli cells. Mol Cell Endocrinol 173:95–107

71. Rey R 1998 Endocrine, paracrine and cellular regulation of postnatal anti-Mullerian hormone secretion by Sertoli cells. Trends Endocrinol Metab 9:271–276

72. AlAttar L, Noel K, Dutertre M, Belville C, Forest MG, Burgoyne PS, JossoN, Rey R 1997 Hormonal and cellular regulation of Sertoli cell anti-Mullerianhormone production in the postnatal mouse. J Clin Invest 100:1335–1343

73. Rannikki AS, Zhang FP, Huhtaniemi IT 1995 Ontogeny of follicle-stimulat-ing-hormone receptor gene: expression in the rat testis and ovary. Mol CellEndocrinol 107:199–208

74. Yomogida K, Ohtani H, Harigae H, Ito E, Nishimune Y, Engel JD, YamamotoM 1994 Developmental stage- and spermatogenic cycle-specific expression oftranscription factor GATA-1 in mouse Sertoli cells. Development 120:1759–1766

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving theendocrine community.


regulation of sertoli cell number and activity by follicle-stimulating hormone and androgen during...

Documents