help button home button Endocrine Society Molecular Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2003-0364
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
18/4/979    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xia, Y.
Right arrow Articles by Schneyer, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xia, Y.
Right arrow Articles by Schneyer, A.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Compound via MeSH
*Substance via MeSH
Hazardous Substances DB
*TESTOSTERONE
Molecular Endocrinology 18 (4): 979-994
Copyright © 2004 by The Endocrine Society

Overexpression of Follistatin-Like 3 in Gonads Causes Defects in Gonadal Development and Function in Transgenic Mice

Yin Xia, Yisrael Sidis and Alan Schneyer

Reproductive Endocrine Unit, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts 02114

Address all correspondence and requests for reprints to: Alan Schneyer, Ph.D., Reproductive Endocrine Unit, BHX-5, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: Schneyer.alan{at}mgh.harvard.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Activin has numerous biological activities including regulation of follicular development, spermatogenesis, and steroidogenesis within the gonads. Activities of activin are regulated by follistatin (FST), an activin binding protein, and perhaps follistatin-like 3 (FSTL3; also known as FLRG and FSRP). FSTL3 is a recently described member of the FST family having an overall structure and activity profile similar to that of FST, including binding and neutralization of activin. FSTL3 is most highly expressed in the placenta and testis, whereas FST is highest in the ovary and kidney, suggesting that FSTL3 has biological actions that do not entirely overlap those of FST. To investigate the role of local FSTL3 as a potential regulator of activin action in gonad development and function, we examined FSTL3 expression in the mouse testis. FSTL3 protein was localized to Leydig cells, spermatagonia, and mature spermatids in normal male mice. We then created transgenic mice using a human FSTL3 cDNA driven by the mouse {alpha}-inhibin promoter. Three of five transgenic founders were fertile and were bred to establish lines. In the highest expressing line 3, transgene expression was largely restricted to gonads, with pituitary, adrenal, brain, and uterine expression being substantially lower. Gonad weights, sperm counts, and fertility were significantly reduced in transgenic males, and reduced litter size was evident in line 3 females. Within the testis, highest transgene expression was observed in Sertoli cells, and although most tubules showed evidence of normal spermatogenic development, degenerating tubules devoid of germ cells and Leydig cell hyperplasia were also evident in every line 3 animal examined. Ovaries from line 3 females contained fewer antral follicles and more apparent follicular atresia. Although circulating human FSTL3 levels were undetectable, FSH and LH levels in adult transgenic mice were not significantly different from wild-type animals. However, testosterone levels were significantly increased at d 21 and significantly reduced at d 60 compared with wild-type males. These results suggest that FSTL3 is likely to be a local regulator of activin action in gonadal development and gametogenesis and, further, that activin appears to have important actions in gonadal development and function that are critical for normal reproduction.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
NORMAL REPRODUCTION IN mammals requires the coordinated action of both endocrine hormones and locally derived growth factors. Among these gonadal factors are members of the TGF-ß superfamily including activin and inhibin, as well as modulators of their activity such as follistatin (FST) and follistatin-like 3 (FSTL3). Activins and inhibins were originally isolated from ovarian follicular fluid based on their ability to stimulate or inhibit (respectively) FSH secretion from gonadotrophs (reviewed in Ref.1). However, it is now recognized that activin has a range of biological actions principally involving autocrine/paracrine mechanisms in many adult tissues, including pituitary, gonads, bone, and muscle, as well as in tissue differentiation during embryological development (reviewed in Ref.2). Within the testis, activin can stimulate proliferation of Sertoli cells (3) and regulate testosterone production from Leydig cells (4). In the ovary, activin-A appears to be important for normal folliculogenesis because replacement of the ßA gene by the ßB gene rescued skeletal defects in ßA null mice but did not restore normal reproduction (5). Moreover, earlier in vitro studies demonstrated that activin treatment increased the number of FSH receptors and FSH action in granulosa cells (6). Taken together, these studies indicate that activin has important roles in maintaining normal gonadal function.

A number of extracellular regulators of activin action have been identified, including FST, FSTL3, and inhibin. FST is a selective binding and neutralization protein of TGF-ß superfamily members including activin (7), myostatin (8), and bone morphogenetic proteins (9), although activin and myostatin appear to be the preferred ligands based on affinity estimates (8, 10, 11). In vitro studies indicate that most, if not all, of FST’s activities are mediated via its ability to neutralize activin through formation of largely irreversible complexes (10) that prevent activin from binding its receptor (12). The physiological significance of FST is underscored by the severity of the defects in FST knockout mice in which pups die within hours of birth (13). In addition, transgenic overexpression of FST under control of the metallothionein promoter resulted in disruption of gonadal architecture as well as function, suggesting that FST can also neutralize activities of activin and/or other TGFß superfamily members in vivo (14).

Recent studies comparing FST and FSTL3 have identified a number of similarities that may be important for characterizing their respective biological actions in vivo (reviewed in Ref.2). FSTL3, originally called follistatin-related gene [FLRG (15)], and also known as follistatin-related protein [FSRP (16)], has significant structural and functional similarity to FST, including high-affinity binding and neutralization of activin (17). Moreover, for both FSTL3 and FST, the affinity for activin was more than 500-fold greater than for any other TGFß family member examined (11), suggesting that neutralization of activin is the primary activity of both FST and FSTL3 under normal physiological conditions. Incidentally, mouse and human FSTL3 are 82% identical homologous at the primary sequence level whereas the identity for FST is 98%, and both have two potential glycosylation sites.

A number of differences exist between FSTL3 and FST, which suggest that they might have differing activities in vivo. FSTL3 lacks the heparin-binding sequence found in FST, a difference that accounts for the inability of FSTL3 to bind cell-surface heparin-sulfated proteoglycans and, consequently, is a weaker antagonist of endogenous or autocrine activin in comparison with FST288 (11). Another functional difference is that FST is primarily a secreted protein whereas FSTL3 was found to be both secreted as well as localized within the nucleus of many FSTL3-producing cells (17). FSTL3 mRNA is most highly expressed in placenta and testis, whereas FST is most highly expressed in the ovary and kidney, raising the possibility that in normal animals, their nonoverlapping bioactivities may be sexually dimorphic and tissue specific (17, 18). Moreover, whereas FST and FSTL3 are differentially regulated in cultured granulosa cells (19, 20), the cellular sources of FSTL3 within the testis have not yet been investigated. Nevertheless, these critical distinctions between FST and FSTL3 suggest that, despite their similar structures and affinity for binding activin, there are important differences in their actions and regulation in vitro. It would thus seem reasonable to hypothesize that FSTL3 may have unique biological roles in vivo not shared by FST.

To examine potential biological activities of gonadally expressed FSTL3 in vivo and to elucidate potential actions of activin through its neutralization within the gonads, we created transgenic mice overexpressing human FSTL3 under the control of a mouse {alpha}-inhibin promoter fragment that was previously shown to drive transgene expression most highly in gonadal tissues (21). Our results indicate that in wild-type males, FSTL3 is expressed in Leydig cells, spermatogonia, and mature spermatids, whereas in transgenic mice, human (h)FSTL3 was also expressed in Sertoli cells. In addition, transgenic male and female mice were subfertile, and their gonads were significantly smaller than wild-type littermates. These studies indicate that FSTL3 may have important regulatory actions in the gonads and, further, that neutralization of activin can have deleterious actions both in gonadal development and in physiological processes that are necessary for full reproductive potential in adults.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Generation of FSTL3 Transgenic Mice
To study the effect of FSTL3 overexpression within gonads, transgenic mice were created that elaborated human FSTL3 mRNA under the control of the mouse {alpha}-inhibin promoter (Fig. 1AGo). Five (two male and three female) of the 37 mice produced from microinjected eggs were transgenic based on detection of a 1.2-kb band in genomic DNA digested with EcoRV (Fig. 1Go, A and B). The two male founders (nos. 2 and 4) were completely infertile so that lines could not be established from these animals, although their DNA appeared to contain fewer copies of the transgene compared with the female founders (Fig. 1BGo). Interestingly, testis weight and histology were not different from wild-type males at 1 yr of age, leaving the cause of the infertility unexplained (data not shown). Nevertheless, the three fertile female founders were bred with wild-type male mice to establish three separate FSTL3 transgenic lines (lines 1, 3, and 5). Southern blot and PCR analysis of genomic DNA from offspring from the three founders demonstrated that all three lines transmitted the transgene. Breeding of transgenic F1 through F3 offspring with wild-type mates resulted in a transgenic-wild-type ratio of 43:57 for all three lines.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 1. Schematic Representation of hFSTL3 Transgene Construct and Southern Blot Analysis of Founder Mice

A, The transgene construct includes a 6-kb inhibin {alpha}-subunit promoter fused to a heterologous intron, hFSTL3 cDNA, and a bovine GH polyadenylation signal. The probe template used for identification of transgenic mice is a 480-bp sequence located at the 5'-end of hFSTL3 cDNA. The positions of primer 1 and primer 2 used for PCR analysis are also shown. B, DNA (10 µg) was digested with EcoRV, fractioned on 0.8% agarose gel, and blotted onto a nylon membrane. The membrane was hybridized with the 480-bp probe and exposed to x-ray film. The predicted 1.2-kb band was observed in five of 37 mice. The probe did not hybridize with genomic DNA from wild-type mice (not shown). The numbers of transgene copies of the two male mice (nos. 2 and 4) were significantly less than those of the three female mice (nos. 1, 3, and 5).

 
Transgene mRNA and Protein Expression
To examine the distribution of transgene expression, we performed quantitative PCR analysis on RNA samples extracted from various tissues of both transgenic and wild-type mice (Table 1Go). As expected from the use of human-specific PCR primers, no transgene mRNA expression was observed in tissues from wild-type mice. Transgene mRNA levels in line 3 testes were more than 10-fold higher compared with other {alpha}-inhibin-expressing tissues including the adrenal gland and pituitary. In the testis and ovary, the highest transgene mRNA levels were observed in line 3, with expression being intermediate in line 5 and undetectable in line 1. In addition, the expression was 7- to 10-fold higher in the testis than in the ovary in both lines 3 and 5. Unexpectedly, however, line 3 transgenic males expressed significant levels of FTSL3 mRNA in the heart. In line 3 females, transgene expression was similar for ovary and uterus.


View this table:
[in this window]
[in a new window]
  		Table 1. Expression of the hFSTL3 Transgene (x10-2)

 
To confirm production of transgenic hFSTL3 protein in gonads, testis or ovary extracts were used for Western blot analysis. Two protein bands at the appropriate molecular weight for FSTL3 were detected in the testis or ovary extracts from line 3 (Fig. 2AGo). Because both mouse and human FSTL3 have two potential Asn-linked glycosylation sites (22) that may account for the multiple forms, we performed deglycosylation experiments on the gonad extracts. Upon treatment with N-glycanase (PNGase F), the two bands were reduced to a single, lower molecular weight band, indicating that the two bands resulted from the same core protein with either one or two glycosylation sites occupied (Fig. 2AGo). The two hFSTL3 bands were also detected in line 5 testes, but the quantity was much less than those in line 3 testes (Fig. 2BGo), consistent with the reduced mRNA levels in this line. FSTL3 levels were similar for line 1 and wild-type testes, suggesting that only endogenous mouse FSTL3 was detectable in these samples (Fig. 2BGo). In ovarian extracts, FSTL3 protein was greatest in line 3 samples, whereas lines 1 and 5 were not different from wild type, suggesting that only line 3 females detectably overexpressed FSTL3 (Fig. 2BGo). Within line 3, transgenic FSTL3 protein expression was much higher in the testis as compared with the ovary (Fig. 2CGo), consistent with its higher mRNA levels. We also examined the transgenic expression across testis development in line 3. Levels of hFSTL3 protein increased dramatically from d 7 to d 21 and then decreased slightly from d 21 to d 60 (Fig. 2DGo). Interestingly, no hFSTL3 protein was detectable in the heart from line 3 transgenic male mice (data not shown) despite the high mRNA levels (Table 1Go), suggesting that translation of this mRNA may be restricted in the heart.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 2. Western Blot Analysis of Gonad Extracts

A, Total protein was extracted from testis and ovaries of line 3 mice at d 60 of age, and then subjected to Western blot analysis using hFSTL3 polyclonal antibody. Two bands at the predicted molecular sizes were detected both in the testis (lane 1) and in the ovary (lane 3). The two bands were shifted down to a single lower band when the extracts were pretreated with PNGase F before loading (lanes 2 and 4). B, Equivalent amounts (15 µg) of protein from the testis (T) and the ovary (O) from 60-d-old wild type (WT) and three transgenic lines (L1, L3, and L5) were loaded and subjected to Western blot analysis. The orders of lines with respect to the amount of hFSTL3 protein are approximated as follows: in the testis, line 3 > line 5 > line 1 {approx} WT; in the ovary, line 3 > line 5 {approx} line 1 {approx} WT. C, Equivalent amounts (15 µg) of protein from line 3 testis (T) and ovary (O) at d 60 of age were used for Western blot analysis. There was much more hFSTL3 protein in the testis than in the ovary. D, Western blot analysis using tissue extracts with an equivalent amount of protein (15 µg) showed the transgene expression increased dramatically from d 7 to d 21 and then declined slightly from d 21 to d 60 in line 3.

 
Cellular Localization of FSTL3 Protein
Although FSTL3 mRNA is highly expressed in testis (16), its precise cellular location has not been delineated. In wild-type adult males, endogenous mFSTL3 immunohistochemical staining was strongest in mature spermatids, but significant staining was also observed in spermatogonia at the base of most tubules, as well as in Leydig cells (Fig. 3Go, A and C). This staining was completely abolished when the antiserum was preincubated with FSTL3 protein, demonstrating the specificity of the antiserum (Fig. 3Go, B and D). Nevertheless, this staining pattern in the male suggests that FSTL3 may have important roles in regulating activin action in both germ and Leydig cells. We were unable to detect mFSTL3 in ovaries of adult wild-type female mice, probably due to its significantly lower expression in this tissue (17).



View larger version (122K):
[in this window]
[in a new window]
  		Fig. 3. Immunolocalization of FSTL3 Expression in the Testis of Wild-Type and Line 3 Transgenic Mice and Developmental Changes in the Expression Levels of Transgene in Line 3 Testis

All sections were stained with 3,3'-diaminobenzidine (brown) and counterstained with hematoxylin (blue). A and B, Testis sections from 60-d-old wild-type mice were incubated with hFSTL3 antibody preincubated with untransfected cell medium (A) or conditioned medium (B), which contains large amounts of hFSTL3 protein. C and D, Magnifications of the white rectangles in panels A and B, respectively. E and F, Testis sections from 60-d-old line 3 transgenic mice were incubated with hFSTL3 antibody preincubated with untransfected cell medium (E) or conditioned medium containing FSTL3 (F). G and H, Magnifications of the white rectangles in panels E and F, respectively. Black arrow, Sertoli cell; arrowhead, spermatogonia; open arrow, Leydig cells. The Sertoli cell staining was barely detectable at d 7 (I), but became very strong at d 21(J) before being slightly weaker at d 60 (K).

 
Immunohistochemical analysis of line 3 transgenic testes revealed that FTSL3 protein was now detected in Sertoli cells (Fig. 3Go, E and G) in addition to mature spermatids, spermatogonia, and Leydig cells that were detected in wild-type testes, suggesting the transgenic hFSTL3 protein was localized primarily to Sertoli cells. Consistent with the results from Western blot analysis (Fig. 2DGo), Sertoli cell staining was barely detectable in the line 3 testis at d 7, but increased substantially by d 21, and then became slightly weaker by d 60 (Fig. 3Go, I–K). No staining for the hFSTL3 transgene was detected in testicular tissue from line 5 males or ovarian sections from line 3 females (data not shown). This is most likely due to the reduced gene expression in these tissues (Table 1Go).

Fertility Studies
To assess fertility, transgenic mice were housed with wild-type mates (25–40 litters per line; five to eight mating pairs per line), and reproductive parameters were compared with wild-type pairs of the same strain (27 litters per five mating pairs) (Fig. 4Go). Although litter frequency was not significantly different between transgene lines and wild-type mice (data not shown), the fertility of line 3 transgenic males mated with wild-type female mice was reduced by 25% compared with lines 1 or 5, or to wild-type males (P < 0.01, Fig. 4AGo). Similarly, the litter size of line 3 females mated with wild-type males was reduced by more than 35% compared with both wild-type or transgene lines 1 and 5 (P < 0.001, Fig. 4BGo). Consistent with the reduced transgene mRNA and protein expression, litter sizes of lines 1 and 5 were not different from each other or from wild type.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 4. Litter Size (Mean ± SEM) of hFSTL3 Transgenic Mice

Individual transgenic (TG) or wild-type (WT) mice were continually housed with a wild-type mate. The mating pairs (n = 5–8) in each group were caged together until they had stopped producing pups for at least 3 months. A, Wild-type or transgenic male mice mated with wild-type female mice. B, Wild-type or transgenic female mice mated with wild-type male mice. **, P < 0.01; ***, P < 0.001.

 
For mice mated continuously from puberty, the mean date of first delivery varied from 65.75–71.25 d whereas the mean date of last delivery ranged from 167–217 d (Table 2Go), and there were no significant differences between wild-type and transgenic animals.


View this table:
[in this window]
[in a new window]
  		Table 2. Duration of Reproductive Lifespan (Days of Age)

 
Body and Gonad Weights
All transgenic mice appeared normal at birth and grew into adulthood. However, the body weights of line 3 male mice were significantly lower than wild-type mice at 7, 21, or 60 d of age, but these differences were not significant in 100-d-old mice (Fig. 5AGo). No differences in body weight were observed between wild-type, line 1, and line 5 males (Fig. 5AGo). Line 3 females were significantly lighter than wild type, line 1, or line 5 females at 7 and 21 d of age but were not different at d 60. Female body weight was not different between wild type, line 1, and line 5 (Fig. 5BGo).



View larger version (41K):
[in this window]
[in a new window]
  		Fig. 5. Body Weight (Mean ± SEM) of Male (A) and Female (B) hFSTL3 Transgenic and Wild-Type Mice

Different sets of transgenic and control wild-type (WT) mice were weighed at d 7, 21, 60, and 100 of age (n >= 5). *, P < 0.05; **, P < 0.01; #, P < 0.001.

 
The size of testes from line 3 males was dramatically decreased compared with those from wild-type males (Fig. 6AGo). In addition, line 3 testis weight was reduced by approximately 70% compared with wild-type or line 1 (P < 0.001) in immature (d 7 and d 21) and adult (d 60, 100, and 330) mice (Fig. 6BGo). Line 5 testis weight was approximately 20% less than wild-type or line 1 (P < 0.01) at d 21, 60, 100, and 330, but this difference was undetectable at d 7. Line 5 testes were larger than line 3 testes at all time points examined (P < 0.001). Thus, testis weight was inversely associated with gonadal hFSTL3 transgene expression.



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 6. Testis and Ovary Weights (Mean ± SEM) of Transgenic and Wild-Type Mice

A, Gross analysis of the testes of a 60-d-old wild-type mouse (left) and a littermate hFSTL3 transgenic mouse (right) in line 3. B, Testis weights of wild type (WT) and transgenic mice at 7, 21, 60, 100, and 330 d of age. At least seven mice were analyzed in each group. C, Sperm counts (mean ± SEM) in transgenic and wild-type mice. Sperm cells were collected from the epididymis of 100-d-old wild-type or transgenic mice. Five to six mice were used in each group. D, Ovary weights of 60-d-old transgenic and wild-type mice. Six to eight mice were used in each group. *, P < 0.05; **, P < 0.01; #, P < 0.001. #/* or #/**: #, line 3 vs. WT or line 1; * or **, line 5 vs. WT or line 1.

 
Consistent with testis weights, sperm counts in the epididymis were also inversely associated with transgene expression levels in the testis (Fig. 6CGo). Sperm counts in line 3 males were 70% (P < 0.01) lower than wild type whereas line 5 counts were reduced by 45% (P < 0.05) compared with wild-type or line 1 mice, which were not different from each other.

Ovarian weights in line 3 females were reduced by 40% in comparison with wild-type females or with line 1 or 5 females (P < 0.05), whereas lines 1 and 5 were not different from wild-type females (Fig. 6DGo), which was again consistent with the different levels of transgene expression in the ovary between lines.

Histological Analyses
To determine whether the reduced fertility and gonad size were associated with defects in gonad architecture, the histology of testes from transgenic and wild-type mice was examined. For the most part, the seminiferous tubules of lines 1, 3, and 5 (Fig. 7Go, B, C, and D, respectively) males at 60 d of age were quite similar to those of wild-type mice (Fig. 7AGo). However, a number of tubules were abnormal in line 3 testes (Fig. 7EGo), consisting largely of Sertoli cells without obvious germ cells. Moreover, these tubules stained intensely for the hFSTL3 transgene (Fig. 7FGo), suggesting that transgenic FSTL3 expression is associated with nonfunctional tubules. In addition, the interstitial cell mass in line 3 testes (Fig. 7Go, C and E) was substantially increased over wild type, most of which stained positive for side chain cleavage enzyme (P450scc; Fig. 7GGo), a characteristic of Leydig cells (23). For comparison, the typical number of P450scc-positive interstitial cells in wild-type testes is shown in Fig. 7HGo. Consistent with the intermediate transgene expression in line 5 testis, these tubules appeared largely normal, but there was some expansion of the Leydig cell mass (Fig. 7DGo). To examine whether the FSTL3 transgene had any effects on the aging in the testis, line 3 males were examined at 330 d of age. No substantial alterations in histology or weight were observed (Fig. 8BGo) that were not already evident at age d 60 (Fig. 7EGo), with the exception that more tubules exhibited what appeared to be cytoplasm nearly filling the tubule lumen but containing no apparent nucleus or spermatids (Fig. 8DGo). This type of tubule was less frequently observed in other lines and wild-type males (Fig. 8CGo).



View larger version (126K):
[in this window]
[in a new window]
  		Fig. 7. Histology of Testes from Transgenic and Wild-Type Mice at 60 d of Age

A–D, Hematoxylin and eosin staining of testis sections from wild-type and line 1, line 3, and line 5 transgenic mice, respectively. Interstitial cells (arrowhead) appear to be more abundant in the testes of line 3 and line 5. E, Hematoxylin and eosin staining of a testis section from line 3. Abnormal seminiferous tubules (arrow) are detected. More interstitial cells are also shown (arrowhead). F, The abnormal tubules (arrow) show stronger hFSTL3 immunoactivity than the normal tubules. G and H, Immunostaining for P450scc (in brown color) in the testis sections from 60-d-old line 3 (G) and wild-type (H) mice. P450scc protein is detected in the normal Leydig cells in wild-type mice and also in hyperplastic Leydig cells in line 3 mice.

 


View larger version (156K):
[in this window]
[in a new window]
  		Fig. 8. Histology of Testes from Wild-Type (A and C) and Line 3 Transgenic (B and D) Mice at 330 d of Age

Black arrow, Seminiferous tubules devoid of germ cells; arrowhead, Leydig cell hyperplasia; open arrow, structures containing cytoplasm but not spermatids.

 
Compared with ovaries from wild-type adults (60 d of age) (Fig. 9Go, A and C), ovaries from line 3 females contained fewer large antral and preovulatory follicles (Fig. 9Go, B and D). In particular, preantral follicles in line 3 ovaries appeared normal, but many of the larger antral follicles showed signs of atresia (Fig. 9Go, D and F). In line 3 ovaries we occasionally observed corpus luteum-like structures containing an abnormal oocyte trapped in the center and infiltrated with red blood cells (Fig. 9BGo, inset), suggesting that some follicles were luteinized without ovulation. At 1 yr of age, the ovaries from both line 3 (Fig. 9HGo) and wild-type mice (Fig. 9GGo) contained only rare primordial/primary follicles along with a few antral follicles. In addition, line 3 ovaries appeared to have less antral follicles than wild type.



View larger version (116K):
[in this window]
[in a new window]
  		Fig. 9. Histology of Ovaries from Wild-Type Mice and Transgenic Line 3 at 60 and 360 d of Age

Ovary sections were stained with hematoxylin and eosin. A and C, Wild-type mouse ovary sections at low (A) and intermediate (C) power magnification, respectively. Black arrow, Antral follicle. B and D, A line 3 transgenic mouse ovary section at low (B) and intermediate (D) power magnification, respectively. Arrowhead, Normal preantral follicle (D). Black arrow, atretic antral follicles (D). The inset shows a corpus luteum (CL)-like structure with an oocyte trapped in the center surrounded by infiltrated blood cells (B). E, A wild-type mouse ovary section at high-power magnification. F, A line 3 transgenic mouse ovary section at high magnification showing a typical atretic-looking follicle with loosely connected cumulus cells around an oocyte with an abnormal looking zona pellucida (open arrow). A normal preantral follicle was also present (arrowhead). G and H, Ovary sections from wild-type (G) and line 3 transgenic (H) mice at the age of 360 d. *, Antral follicles.

 
We also examined the adrenal gland and pituitary from line 3 males and females, both of which appeared to be normal both in size and structure (data not shown).

Serum Hormone Measurements
To determine whether gonadal abnormalities in transgenic mice were associated with altered endocrine hormone levels, circulating gonadotropin and testosterone levels were assessed by RIA. No significant differences were observed for any transgenic line compared with wild-type mice for FSH (Fig. 10Go, A and B) or LH (Fig. 10Go, C and D). Normal gonadotropins were actually somewhat unexpected in line 3 animals because gonadal feedback could have been compromised due to the abnormalities observed in both testes and ovaries. In addition, testosterone levels were significantly elevated in line 3 males at 21 d of age (Fig. 10EGo), but this reversed to a significant decrease in testosterone levels by 60 d of age (Fig. 10FGo), suggesting that the expanded Leydig cell mass in line 3 males initially produced more testosterone but that altered Leydig cell response to LH might cause reductions in testosterone as the animals age. Circulating FSTL3 was not detectable by RIA in wild-type or line 3 transgenic animals (n =11; data not shown).



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 10. Serum Levels of FSH, LH, and Testosterone in Wild-Type and Transgenic Mice

A and B, Serum FSH levels for male (panel A, n = 8) and female (panel B, n = 5–6) mice at 60–90 d of age. C and D, Serum LH levels for male (panel C, n = 8–14) and female (panel D, n = 6–10) mice at 60–90 d of age. E and F, Serum testosterone levels for male mice at 21 d of age (panel E, n = 6–7) and at 60 d of age (panel F, n = 14–22). *, P < 0.05.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Within the testis of normal male mice, endogenous FSTL3 protein was observed in mature spermatids in the tubule lumen, spermatogonia at the tubule base, and in Leydig cells. Moreover, using the mouse {alpha}-inhibin promoter to direct transgenic expression of hFSTL3, we found that hFSTL3 was largely restricted to the gonads and was higher in males than females, confirming that this is a useful strategy for analyzing the reproductive actions of FSTL3 in vivo. In the highest expressing (line 3) males, testis weight and spermatogenic output were reduced by 70%, FSTL3 transgene was expressed primarily in Sertoli cells, and Leydig cell hyperplasia was evident. Line 3 females had reduced ovarian weight, but protein expression could not be localized. Fertility in both line 3 males and females was significantly reduced. Line 5 mice exhibited an intermediate level of transgene expression, and in many of these traits, demonstrated an intermediate phenotype. These results indicate that in vivo, FSTL3 overexpression results in altered gonadal development and function. Moreover, because FSTL3 is normally expressed in the testis, our observations suggest that endogenous FSTL3 may play a critical role in regulating activin-associated gonadal development and physiology, especially in males.

In the testes of line 3 males, transgenic FSTL3 was detected almost exclusively in Sertoli cells. Expression was maximal in immature males (d 21) but remained relatively strong even in adults. Moreover, reduced testis weight was observed as early as postnatal d 7. Because the increase in testis weight at this early developmental time point in rodents is primarily derived from Sertoli cell proliferation (24, 25), our results suggest that transgenic expression of FSTL3 in Sertoli cells suppresses Sertoli cell proliferation, leading to reduced testicular weights that persist into adulthood. Moreover, because Sertoli cells are capable of supporting a fixed number of developing germ cells (24), the number of Sertoli cells established during this proliferative phase constitutes an upper limit to spermatogenic output in adult animals. Thus, transgenic FSTL3 expression in immature males may limit Sertoli cell proliferation, ultimately leading to reduced spermatogenic output and fertility in these animals.

It has been known for some time that activin can promote Sertoli cell proliferation in culture (26). Moreover, it was recently demonstrated that in purified immature rat testicular cell cultures, activin promoted Sertoli cell proliferation only in cells taken between d 3 and d 9, and that the primary source for activin-A was peritubular myoid (PTM) cells (3). These results suggest that paracrine activin derived from PTM cells regulates Sertoli cell proliferation during a brief developmental window. It has been shown previously that FSTL3, like FST, is highly selective for activin and myostatin, because the affinity for inhibin and bone morphogenetic protein 7, the next most active TGFß superfamily members, were about 500-fold lower (8, 10, 11). Moreover, we have also demonstrated in vitro that whereas FST can antagonize both endogenous and exogenous (autocrine and paracrine/endocrine, respectively) activin, FSTL3 is a much weaker antagonist of endogenous activin, a difference probably accounted for by its inability to bind to the cell surface (11). Taken together with the observation that myostatin has not been localized to the testis, these results suggest that in vivo, intragonadal FSTL3 is likely to be a regulator of paracrine and/or endocrine activin as long as its concentration is sufficiently limited to prevent neutralization of related TGFß superfamily members. We therefore propose that the transgenic FSTL3 expressed in Sertoli cells inhibited PTM-derived activin in immature line 3 males, resulting in substantial suppression of Sertoli cell proliferation compared with wild-type animals. If this model is ultimately proven correct, our results would then support the concept that activin, as regulated by FSTL3, is critical for normal testicular development and function. Moreover, these results indicate that activin, or a related TGFß family member, is essential for attaining normal Sertoli cell number and that this activity can be inhibited in vivo by FSTL3.

In adult animals, activin may also be important for maintenance of tubule structure, because treatment of mixed testicular cultures with activin results in formation of tubule-like structures in vitro (26). In testes from 60-d-old line 3 males we observed a number of apparently degenerating tubules with disorganized Sertoli cells and no germ cells. These tubule structures were highly immunoreactive with the FSTL3 antibody, suggesting that FSTL3 expression in adult tubules is associated with disruption of normal testicular architecture. Thus, the disorganized tubules suggest that suppression of activin action by FSTL3, even in the adult, can have deleterious effects on spermatogenesis independent of its actions on regulating Sertoli cell number. However, it should be noted that in older line 3 males, there was no further acceleration in the degeneration process so that the number of defective tubules at 330 d was not different from that at 60 d. It is possible that the decrease in FSTL3 transgene expression observed between d 21 and 60 may limit further tubule degeneration in older animals.

Our results with FSTL3 agree with experiments in which FST was overexpressed in testes, which also resulted in smaller testes, arrest of spermatogenesis, tubule degeneration, and Leydig cell hyperplasia (14). Although it has been known that activin subunit mRNA and protein, as well as its receptors, are expressed in testicular cells including germ cells (reviewed in Ref.27), and activin promotes proliferation of spermatogonia in vitro (26), the precise role of activin in adult spermatogenesis in vivo remains to be elucidated. Disruption of the activin ßA gene resulted in death within 24 h of birth, thereby preventing analysis of the role of activin-A in adult reproductive processes, whereas elimination of activin-B did not deleteriously impact gonadal function. Interestingly, overexpression of activin has been reported to have deleterious effects on seminiferous tubule organization and spermatogenesis, including tubule degeneration by 10 wk of age, absence of sperm in the epidydimis, and smaller testes (28). However, because transgene expression was not observed until d 21 in these animals, the effect of activin on Sertoli cell proliferation that occurs closer to birth could not be ascertained. Interestingly, disruption of the activin receptor ActRII caused delayed fertility and smaller testes as early as d 21, although spermatogenesis in the remaining tubules appeared normal (29). This could indicate that activin-A, acting through ActRII, is critical for developmental regulation of Sertoli cell proliferation (i.e. the active activin receptor on Sertoli cells) whereas the alternate activin receptor, ActRIIB, the expression of which was not altered in the ActRII knockouts, is critical for maintenance of spermatogenesis or at least tubule structure and spermatogonial proliferation in adults. Taken together, these studies indicate that activin action may be critical for normal testicular development and, further, that alteration in the concentration of activin or its regulatory proteins FSTL3 and FST can lead to defective testicular function.

The Leydig cell hyperplasia observed in line 3 males may represent an actual increase in Leydig cell number, or maintenance of a normal complement of Leydig cells in the context of reduced Sertoli cell number and tubule content. Interestingly, in line 3 males, testosterone was significantly increased in immature animals (d 21) but decreased in adults relative to wild-type males, suggesting that either the total number of cells is altered after puberty or their function is suppressed. Because LH levels were not different between line 3 and wild-type males, any alteration in testosterone output in response to LH must be due to local effects. Previous studies have demonstrated that activin can suppress LH-stimulated testosterone production from cultured immature Leydig cells (4). Thus, neutralization of activin by transgenic FSTL3 could explain the higher testosterone we observed in immature line 3 males. In adults, testosterone could be suppressed by other local factors not inhibited by FSTL3, or their collective response to LH could be reduced. In contrast, Leydig cell hyperplasia in FST transgenic mice was not associated with alteration in testosterone levels (14), suggesting that for FST, as well as FSTL3, the collective response of the larger number of Leydig cells was somehow suppressed. Nevertheless, the increase in apparent Leydig cell number is consistent with the phenotype observed when the MIS gene, another TGFß superfamily ligand, was disrupted (30). It is presently unclear whether MIS and activin interact in the same pathway or separately contribute to regulation of Leydig cell proliferation, but their elimination or suppression clearly leads to increased Leydig cell number. It is also possible that this altered Leydig cell number contributes to the observed tubular degeneration because the larger Leydig cell mass was often observed closely associated with degenerating tubules.

It is not clear why the transgene expression pattern observed in this study is somewhat different from that reported previously using the same {alpha}-inhibin promoter. For example, we expected to see higher expression in females relative to males, and a stronger female phenotype because {alpha}-inhibin is expressed more highly in females relative to males (31) whereas endogenous FSTL3 in the ovary is lower than testis (17). Moreover, {alpha}-inhibin promoter-directed transgene expression was observed in previous studies in both theca and granulosa cells of secondary and antral follicles of all stages (21, 32). The primary differences between the studies are the strain of mouse and our insertion of a heterologous intron sequence between the promoter and cDNA that may have altered activity of enhancers, thereby modifying expression. This may also explain our greater tissue-specific expression because we observe extremely low levels of the transgene in {alpha}-inhibin-expressing tissues such as adrenal and pituitary whereas other studies observed transgene expression in those tissues. Of course, the insertion position of the transgene in each experiment was different and could also explain differences in observed expression.

Based on quantitative PCR analysis, FSTL3 transgene expression in the ovary was approximately 10-fold lower than in the testis. This reduced expression probably accounts for our inability to detect transgenic FSTL3 protein production by immunohistochemistry. Interestingly, line 3 females were also subfertile, suggesting that even at the relatively low expression levels observed, transgenic FSTL3 still disrupted normal follicular processes dependent on activin. Histological evaluation of ovaries indicated that fewer developing and antral follicles were present in line 3 females and more of the antral follicles appeared atretic, which may account for the reduced fertility in these animals. This reduced fecundity occurred in the context of a tendency toward elevated FSH levels, suggesting that inadequate gonadotropic support was not responsible for the decreased follicle selection or increased atresia, both gonadotropin dependent processes. Interestingly, the reproductive interval between first and last litter was not different between transgenic and wild-type females, suggesting that either fewer follicles ovulate each cycle or that fewer gametes are capable of developing to term in line 3 females, but their total follicle pool might not be substantially altered. Taken together with results in males, it seems likely that activin, the primary ligand identified for both FST and FSTL3, is important for complete fertility in females. This is supported by the disruption of follicular maturation observed in mice overexpressing FST under the metallothionein promoter (14) and the infertility observed in ActRII receptor knockout mice in which higher rates of follicle atresia were observed (29). Moreover, replacement of the activin ßA subunit gene with the mature ßB gene was able to rescue skeletal defects in activin-A knockout mice but not the reproductive defects (5), providing further support for the concept that activin-A is critical for normal female fertility. However, because not all TGFß superfamily ligands have been evaluated for affinity of binding to FST or FSTL3, it also remains possible that a related factor other than activin A or B that also uses the ActRII signaling pathway might be inhibited in these transgenic models.

Line 3 males and females are smaller than wild-type animals at birth, but this difference decreases after puberty. Because this decreased body weight was evident in line 3 pups from wild-type females (and transgenic males), it is unlikely to reflect a defect in nursing. It is possible that lower body weights in younger animals were due to deficient suckling ability in line 3 pups because the body weight differences decreased with age.

Further analysis of these mice may reveal possible applications to human fertility. For example, locally administered activin, in combination with systemic FSH, may provide a unique treatment for infertile human males who have not yet undergone puberty, such as men with idiopathic hypogonadotrophic hypogonadism (33, 34). In those idiopathic hypogonadotrophic hypogonadism patients with complete absence of GnRH secretion, administration of pulsatile GnRH often fails to completely normalize testicular volume and sperm counts despite normalization of FSH levels (35), suggesting that it may be beneficial to first stimulate proliferation of Sertoli cells with both FSH and activin for restoring optimal Sertoli cell number and maximal fertility before administration of human chorionic gonadotropin or GnRH to mature the Sertoli cells and initiate spermatogenesis.

The reproductive defects observed in mice overexpressing FSTL3 in their gonads demonstrate that signaling by activin (or a related ligand) is critical for normal reproductive development and function. These results also demonstrate that FSTL3 can function in vivo as an activin antagonist in a manner similar to our observations in vitro, suggesting an important role for FSTL3 in reproduction. Ultimately, tissue- and age-specific expression or disruption of FSTL3 will be critical to elucidate the precise role(s) of FSTL3 in the gonads.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Generation of Transgenic Mice
A 6-kb fragment of the mouse {alpha}-inhibin promoter was originally cloned in pBluescript-KSII by Hsu et al. (32) and was generously provided by Drs. Ilpo Huhtaniemi and Nafus Rahman (University of Turku, Turku, Finland). The transgenic construct was generated by inserting sequentially a heterologous intron [intron 1 from the human ß globin gene (358–489; gi:30349216)], hFSTL3 cDNA (0.94 kb), and bovine GH polyadenylation fragment (0.25 kb) downstream of the {alpha}-inhibin promoter. The plasmid was digested with XbaI and the transgene was purified. Microinjection was carried out at the Transgene Core Facility of Massachusetts General Hospital. Thirty-seven mice were obtained from the microinjected eggs generated from B6C3F1 (C57BI/6 x C3H). For screening, genomic DNA was extracted from tail biopsies, and transgenic founders were identified by both Southern blot and PCR analyses of genomic DNA. DNA (10 µg) was digested with EcoRV, electrophoresed on 0.8% agarose gels, and subjected to Southern blotting using a sequence of 480 bp from the 5'-end of the FSTL3 cDNA as a probe (Fig. 1AGo). The forward primer used for PCR corresponded to inhibin {alpha}-subunit promoter region (primer 1: 5'-GTC CTA GAC AGA AAG GGC ACA GGG-3') and the reverse primer to FSTL3 cDNA region (primer 2: 5'-GGT GAC ATC AGT CTG GAG CAC CAG G-3'). To monitor the quality of DNA, a reverse primer (primer 3) was designed to correspond to exon 1 of inhibin {alpha}-subunit gene (5'-CTT CCT CCT CTG GTT CAG AGG TCC-3'). The amplification of the transgene and endogenous inhibin {alpha}-subunit gene was performed in the same tube for 35 cycles using an annealing temperature of 60 C. All transgenic founder mice and offspring were mated with wild-type littermates beginning at 6 wk of age. The breeding pairs were maintained until they did not produce pups for at least 3 months to test the fertility of these mice. Transgenic and wild-type mice were weighed at d 7, 21, 60, and 100 and killed, and testes from all three transgenic lines of mice and wild-type mice were weighed. Ovaries were removed at random times during the estrous cycle at 60 d of age and weighed. All animal studies were conducted in accordance with an animal use protocol approved by the institutional animal use committee in accordance with United States Department of Agriculture guidelines.

Real-Time RT-PCR
Total RNA was extracted from tissues stored in RNAlater (Ambion, Austin, TX) using Trizol (Life Technologies, Inc., Gaithersburg, MD) according to manufacturer’s protocol. Total RNA (0.5–1.0 µg) was reverse transcribed as previously described (36). Quantitation of human FSTL3 mRNA was performed on a Stratagene Mx-4000 (Stratagene, La Jolla, CA) using primers as follows: forward, 5'-TGG TGC TCC AGA CTG ATG TCA; reverse, 5'-CAG TGG ACA AGG CCC AAG AA; and a probe, 5' 6-FAM d(AAC ATT GAC ACC GCC TGG TCC AAC C) bhq-1 3' (Biosearch Technologies, Inc, Novato, CA) all within exons 1–2 of hFSTL3. To normalize for RNA quality and efficiency of reverse transcription, mouse Rpl19 mRNA was also quantitated using a forward primer, 5'-CCT GAA GGT CAA AGG GAA TGT GTT; a reverse primer, GCT TTC GTG CTT CCT TGG TCT TA; and a probe, 5'-/5HEX/TGC GAG CCT CAG CCT GGT CAG CC/3BHQ_2/-3' (Integrated DNA Technologies, Inc., Coralville, IA). Results are expressed as a ratio of FSTL3/Rpl19 as determined from standard curves run in each assay made from PCR products cloned into PCRII (Invitrogen, Carlsbad CA) and quantitated by spectrophotometer.

Western Blot Analysis
Testes from mice at d 7, 21, and 60 of age and ovaries from mice at d 60 of age were collected and immediately frozen on dry ice, and stored at -80 C. The individual frozen gonads from each group were homogenized in 800 µl CelLytic-MT mammalian tissue lysis/extraction reagent (Sigma Chemical Co., St. Louis, MO) supplemented with protease inhibitor cocktail tablets (Roche Diagnostics GmbH, Mannheim, Germany). The extracted tissue was centrifuged, and the supernatant was measured for protein concentration by protein assay (Bio-Rad Laboratories, Inc., Hercules, CA) before being aliquoted. An equal amount (15 µg) of protein from each sample was subjected to SDS-PAGE under reducing conditions using Bio-Rad precast gels (12% Tris-HCl). After electrophoresis, the separated proteins were transferred to a 0.45-µm polyvinylidene difluoride membrane (Millipore Corp., Bedford, MA). At the conclusion of the transfer, the membranes were blocked for 1 h in blocking solution (10% skim milk powder, 0.2% Tween 20 in Tris-buffered saline), before incubation overnight at 4 C with purified polyclonal antibody raised in rabbits against human FSTL3 (provided by Millennium Pharmaceuticals, Cambridge, MA) at a final concentration of 1 µg/ml in 4% milk-Tris-buffered saline-Tween solution. The membranes were washed three times before being incubated for 2 h at room temperature with second antibody (horseradish peroxidase-conjugated donkey antirabbit, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted at 1:15,000. Bound antibody was detected with ECL Reagent Plus (PerkinElmer Life Sciences, Boston, MA).

Deglycosylation was performed with line 3 gonad extracts using PNGase F according to the manufacturer’s instructions (New England Biolabs Inc., Beverly, MA) before the samples were subjected to SDS-PAGE and Western blotting.

Histology
Testes from 7-, 21-, 60-, and 330-d-old mice were fixed in Bouin’s solution overnight, and ovaries from 60- and 360-d-old mice were fixed in 4% paraformaldehyde overnight. The tissues were then transferred to 70% ethanol and rinsed daily for 4 d. The tissues were dehydrated in a series of ascending concentrations of ethanol and embedded in paraffin. Sections (6 µm in thickness) were stained with hematoxylin and eosin.

Immunohistochemistry
Sections were deparaffinized with Safeclear II (Fisher Scientific Co. L.L.C., Middletown, VA) and rehydrated in a descending ethanol series. Antigen retrieval was performed in 0.01 M citrate buffer (pH 6.0) using a microwave oven. Sections were heated for 10 min at full power, maintained for 15 min at 30% power, and allowed to cool for 20 min. Sections were then treated with 3% H2O2 for 30 min and were blocked for 1 h using 1.5% normal goat serum. Tissue sections were incubated overnight at 4 C with 5 µg/ml rabbit polyclonal antibody against hFSTL3 or 1:200 rabbit anti-rat P450 side chain cleavage (P450scc) polyclonal antibody (catalog no. AB1244, Chemicon Intl., Temecula, CA). The following day, sections were incubated for 1 h with biotinylated goat antirabbit IgG, and then for 30 min with Vectastain Elite ABC kit (Vector Laboratories, Inc., Burlingame, CA). Color product was developed by 3,3'-diaminobenzidine (ICN Biomedicaia, Inc., Aurora, OH). The reactions were stopped in water, and sections were counterstained with Harris’ hematoxylin, dehydrated in ethanol, cleared in xylene, and mounted with Cytoseal (Richard-Allan Scientific, Kalamazoo, MI).

The hFSTL3 staining specificity was assessed by replacement of primary antibody with normal rabbit IgG, as well as by preincubating the hFSTL3 antibody for 2 h with conditioned medium from transfected HEK 293 cells secreting large amounts of hFSTL3 before applying to tissue sections.

Sperm Production
Epididymides were collected from 100-d-old male mice into 2 ml of M2 medium (Sigma). The sperm were squeezed from the epididymis into the medium using watchmaker’s forceps and then incubated at 37 C for 15–30 min. A 1:10 or 1:20 dilution was used for sperm counting on a hemocytometer as described previously (37).

Hormone Assays
The hFSTL3 RIA was performed as previously described (17) except that purified 3xFlag-hFSTL3 was used for radioiodination and standard. All the samples were measured in a single assay. The detection limit of the assay was 4 ng/ml, and the intraassay coefficient of variance was 6.50%. Serum levels of FSH and LH were analyzed by RIA using reagents from the NHPP, National Institute of Diabetes and Digestive and Kidney Diseases and purchased from or provided by Dr. A. Parlow. The FSH assay used rFSH-I-9 (AFP-128288) as tracer, mouse FSH (AFP-3080) as standard, and guinea pig antimouse FSH (AFP-1760191). Dilutions of normal mouse serum were parallel to the FSH standard curve, and the detection limit of the FSH assay was 2.78 ng/ml. The intra- and interassay coefficients of variance were 2.45% and 13.79% respectively. The LH assay used rLH-I-10 (AFP-11536B) as tracer, mouse LH (AFP-6306A) as standard, and rabbit antirat LH-S-11(AFPC697071P). Dilutions of mouse serum were parallel to the LH standard curve, and the detection limit of the LH assay was 0.04 ng/ml. The intra- and interassay coefficients of variance were 3.60% and 6.25%, respectively. Serum testosterone levels were measured by Coat-A-Count RIA kit (Diagnostic Products Corp., Los Angeles, CA) according to the manufacturer’s protocol. Dilutions of male mouse serum pool gave a dose-response curve parallel to the testosterone standard curve. The detection limit of the assay was 0.025 ng/ml, and the intra- and interassay coefficients of variance were 4.25 and 12.54%, respectively. Castrated mouse serum contained 0.08 ng/ml, indicating that nontesticular testosterone was less than 2–5% of total testosterone using this assay.

Statistics
Statistical analysis was performed with Student’s t test to evaluate the differences between transgenic and wild-type mice. All values were expressed as the mean ± SEM. Differences were considered statistically significant at P < 0.05.


   ACKNOWLEDGMENTS
 
The {alpha}-inhibin promoter construct was generously provided by Drs. Ilpo Huhtaniemi and Nafus Rahman along with critical advice on transgene construction. The FSTL3 cDNA and additional advice on transgene construction were kindly provided by Dr. William Holmes and Millennium Pharmaceuticals. We are also indebted to Dr. Patrick Sluss who helped with testosterone assay design and validation. Dr. Martin Dym provided helpful suggestions for evaluation of testicular phenotypes. We also thank Drs. Abir Mukherjee, Nelly Pitteloud, and William F. Crowley, Jr., for helpful discussion. Dr. Ernestine Schipocni and Janet Saxton provided expertise and access to critical facilities for the histological and immunocytochemical analyses.


   FOOTNOTES
 
This work was supported by National Institutes of Health Grant R01HD39777.

Results from this work were presented in part at the 85th Annual Meeting of The Endocrine Society, Philadelphia, Pennsylvania, 2003, and the International Symposium on Inhibins, Activins and Follistatins, Siena, Italy, 2003.

Abbreviations: FST, Follistatin; FSTL3, FST-like 3; PNGase F, N-glycanase; PTM, peritubular myoid.

Received for publication September 17, 2003. Accepted for publication January 15, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Vale W, Rivier C, Hsueh A, Campen C, Meunier H, Bicsak T, Vaughan J, Corrigan A, Bardin W, Sawchenko P, Spiess J, Rivier J 1988 Chemical and biological characterization of the inhibin family of protein hormones. Recent Prog Horm Res 44:1–34
  2. Welt C, Sidis Y, Keutmann H, Schneyer A 2002 Activins, inhibins, and follistatins: from endocrinology to signaling. A paradigm for the new millennium. Exp Biol Med (Maywood) 227:724–752[Abstract/Free Full Text]
  3. Buzzard JJ, Farnworth PG, de Kretser DM, O’Connor AE, Wreford NG, Morrison JR 2003 Proliferative phase Sertoli cells display a developmentally regulated response to activin in vitro. Endocrinology 144:474–483[Abstract/Free Full Text]
  4. Hsueh AJW, Dahl KD, Vaughan J, Tucker E, Rivier J, Bardin CW, Vale W 1987 Heterodimers and homodimers of inhibin subunits have different paracrine action in the modulation of luteinizing hormonestimulated androgen biosynthesis. Proc Natl Acad Sci USA 84:5082–5086[Abstract/Free Full Text]
  5. Brown CW, Houston-Hawkins DE, Woodruff TK, Matzuk MM 2000 Insertion of Inhbb into the Inhba locus rescues the Inhba-null phenotype and reveals new activin functions. Nat Genet 25:453–457[CrossRef][Medline]
  6. Findlay JK 1993 An update on the roles of inhibin, activin, and follistatin as local regulators of folliculogenesis. Biol Reprod 48:15–23[Abstract]
  7. Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H 1990 Activin-binding protein from rat ovary is follistatin. Science 247:836–838[Abstract/Free Full Text]
  8. Lee SJ, McPherron AC 2001 Regulation of myostatin activity and muscle growth. Proc Natl Acad Sci USA 98:9306–9311[Abstract/Free Full Text]
  9. Iemura S, Yamamoto TS, Takagi C, Uchiyama H, Natsume T, Shimasaki S, Sugino H, Ueno N 1998 Direct binding of follistatin to a complex of bone-morphogenetic protein and its receptor inhibits ventral and epidermal cell fates in early Xenopus embryo. Proc Natl Acad Sci USA 95:9337–9342[Abstract/Free Full Text]
  10. Schneyer AL, Rzucidlo DA, Sluss PM, Crowley Jr WF 1994 Characterization of unique binding kinetics of follistatin and activin or inhibin in serum. Endocrinology 135:667–674[Abstract]
  11. Sidis Y, Tortoriello DV, Holmes WE, Pan Y, Keutmann HT, Schneyer AL 2002 Follistatin-related protein and follistatin differentially neutralize endogenous vs. exogenous activin. Endocrinology 143:1613–1624[Abstract/Free Full Text]
  12. de Winter J, Ten Dijke P, de Vries CJM, van Achterberg TAE, Sugino H, De Waele P, Huylebroeck D, Verschueren K, Eijnden-van Raaij AJM 1996 Follistatins neutralize activin bioactivity by inhibition of activin binding to its type II receptors. Mol Cell Endocrinol 116:105–114[CrossRef][Medline]
  13. Matzuk MM, Lu N, Vogel HJ, Sellheyer K, Roop DR, Bradley A 1995 Multiple defects and perinatal death in mice deficient in follistatin. Nature 374:360–363[CrossRef][Medline]
  14. Guo Q, Kumar TR, Woodruff T, Hadsell LA, DeMayo FJ, Matzuk MM 1998 Overexpression of mouse follistatin causes reproductive defects in transgenic mice. Mol Endocrinol 12:96–106[Abstract/Free Full Text]
  15. Hayette S, Gadoux M, Martel S, Bertrand S, Tigaud I, Magaud JP, Rimokh R 1998 FLRG (follistatin-related gene), a new target of chromosomal rearrangement in malignant blood disorders. Oncogene 16:2949–2954[CrossRef][Medline]
  16. Schneyer A, Tortoriello D, Sidis Y, Keutmann H, Matsuzaki T, Holmes W 2001 Follistatin-related protein (FSRP): a new member of the follistatin gene family. Mol Cell Endocrinol 180:33–38[CrossRef][Medline]
  17. Tortoriello DV, Sidis Y, Holtzman DA, Holmes WE, Schneyer AL 2001 Human follistatin-related protein: a structural homologue of follistatin with nuclear localization. Endocrinology 142:3426–3434[Abstract/Free Full Text]
  18. Arai KY, Tsuchida K, Uehara K, Taya K, Sugino H 2003 Characterization of rat follistatin-related gene: effects of estrous cycle stage and pregnancy on its messenger RNA expression in rat reproductive tissues. Biol Reprod 68:199–206[Abstract/Free Full Text]
  19. Tuuri T, Ritvos O 1995 Regulation of the activin-binding protein follistatin in cultured human luteinizing granulosa cells: characterization of the effects of follicle-stimulating hormone, prostaglandin E2, and different growth factors. Biol Reprod 56:1508–1516
  20. Liu J, Vanttinen T, Hyden-Granskog C, Voutilainen R 2002 Regulation of follistatin-related gene (FLRG) expression by protein kinase C and prostaglandin E(2) in cultured granulosa-luteal cells. Mol Hum Reprod 8:992–997[Abstract/Free Full Text]
  21. Kananen K, Markkula M, El Hefnawy T, Zhang FP, Paukku T, Su JG, Hsueh AJ, Huhtaniemi I 1996 The mouse inhibin {alpha}-subunit promoter directs SV40 T-antigen to Leydig cells in transgenic mice. Mol Cell Endocrinol 119:135–146[CrossRef][Medline]
  22. Tsuchida K, Arai KY, Kuramoto Y, Yamakawa N, Hasegawa Y, Sugino H 2000 Identification and characterization of a novel follistatin-like protein as a binding protein for the TGF-ß family. J Biol Chem 275:40788–40796[Abstract/Free Full Text]
  23. Yu RN, Ito M, Saunders TL, Camper SA, Jameson JL 1998 Role of Ahch in gonadal development and gametogenesis. Nat Genet 20:353–357[CrossRef][Medline]
  24. Sharpe RM, McKinnell C, Kivlin C, Fisher JS 2003 Proliferation and functional maturation of Sertoli cells, and their relevance to disorders of testis function in adulthood. Reproduction 125:769–784[Abstract]
  25. Walker WH 2003 Molecular mechanisms controlling Sertoli cell proliferation and differentiation. Endocrinology 144:3719–3721[Free Full Text]
  26. Mather JP, Attie KM, Woodruff TK, Rice GC, Phillips DM 1990 Activin stimulates spermatogonial proliferation in germ-Sertoli cell cocultures from immature rat testis. Endocrinology 127:3206–3214[Abstract]
  27. Mather JP, Moore A, Li RH 1997 Activins, inhibins, and follistatins: further thoughts on a growing family of regulators. Proc Soc Exp Biol Med 215:209–222[Abstract]
  28. Tanimoto Y, Tanimoto K, Sugiyama F, Horiguchi H, Murakami K, Yagami K, Fukamizu A 1999 Male sterility in transgenic mice expressing activin ßA subunit gene in testis. Biochem Biophys Res Commun 259:699–705[CrossRef][Medline]
  29. Matzuk MM, Kumar TR, Bradley A 1995 Different phenotypes for mice deficient in either activins or activin receptor type II. Nature 374:356–360[CrossRef][Medline]
  30. Mishina Y, Rey R, Finegold MJ, Matzuk MM, Josso N, Cate RL, Behringer RR 1996 Genetic analysis of the Mullerian-inhibiting substance signal transduction pathway in mammalian sexual differentiation. Genes Dev 10:2577–2587[Abstract/Free Full Text]
  31. Meunier H, Rivier C, Evans R, Vale W 1988 Gonadal and extragonadal expression of inhibin {alpha}, ß-A, and ß-B subunits in various tissues predicts diverse functions. Proc Natl Acad Sci USA 85:247–251[Abstract/Free Full Text]
  32. Hsu SY, Lai RJ, Nanuel D, Hsueh AJ 1995 Different 5'-flanking regions of the inhibin-{alpha} gene target transgenes to the gonad and adrenal in an age-dependent manner in transgenic mice. Endocrinology 136:5577–5586[Abstract]
  33. Seminara SB, Crowley Jr WF 2002 Genetic approaches to unraveling reproductive disorders: examples of bedside to bench research in the genomic era. Endocr Rev 23:382–392[Abstract/Free Full Text]
  34. Pitteloud N, Hayes FJ, Boepple PA, DeCruz S, Seminara SB, MacLaughlin DT, Crowley Jr WF 2002 The role of prior pubertal development, biochemical markers of testicular maturation, and genetics in elucidating the phenotypic heterogeneity of idiopathic hypogonadotropic hypogonadism. J Clin Endocrinol Metab 87:152–160[Abstract/Free Full Text]
  35. Pitteloud N, Hayes FJ, Dwyer A, Boepple PA, Lee H, Crowley Jr WF 2002 Predictors of outcome of long-term GnRH therapy in men with idiopathic hypogonadotropic hypogonadism. J Clin Endocrinol Metab 87:4128–4136[Abstract/Free Full Text]
  36. Sidis Y, Fujiwara T, Leykin L, Isaacson K, Toth T, Schneyer AL 1998 Characterization of inhibin/activin subunit, activin receptor, and follistatin messenger ribonucleic acid in human and mouse oocytes: evidence for activin’s paracrine signaling from granulosa cells to oocytes. Biol Reprod 59:807–812[Abstract/Free Full Text]
  37. World Health Organization 1992 Laboratory manual for the examination of human semen and semen-cervical mucus interaction. Cambridge, UK: Cambridge University Press



This article has been cited by other articles:


Home page
 EndocrinologyHome page
A. L. Schneyer, Y. Sidis, A. Gulati, J. L. Sun, H. Keutmann, and P. A. Krasney
Differential Antagonism of Activin, Myostatin and Growth and Differentiation Factor 11 by Wild-Type and Mutant Follistatin
Endocrinology, September 1, 2008; 149(9): 4589 - 4595.
[Abstract] [Full Text] [PDF]