Transgenic mice as a model to study the regulation of human transferrin expression in Sertoli cells

C. Lécureuil1, M.C. Saleh2, I. Fontaine1, B. Baron2, M.M. Zakin2 and F. Guillou1,3

1 UMR 6175 Institut National de la Recherche Agronomique, Centre National de Recherche Scientifique, Université de Tours, Haras Nationaux ‘Physiologie de la Reproduction et des Comportements’, 37380 Nouzilly and 2 Unité d’Expression des Gènes Eucaryotes, Institut Pasteur, Paris, France

3 To whom correspondence should be addressed. e-mail: guillou{at}tours.inra.fr


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Understanding the regulation of proteins secreted by human Sertoli cells is important for identifying the causes of infertility in men. However, experiments with Sertoli cells purified from healthy testes are difficult to perform, for obvious ethical reasons. Therefore, experiments with transgenic mouse models could provide an alternative approach to study the function and regulation of a human gene in Sertoli cells. METHODS: To validate this approach, transgenic mice were generated using phage P1 containing an 80 kbp insert encompassing the complete human transferrin (hTf) gene. The expression pattern of hTf in the mouse background was analysed by isolating Sertoli cells from transgenic mice and comparing the regulation of the human and mouse Tf genes by hormones, retinoids and a cytokine in vitro. RESULTS: The hTf gene in transgenic mice shows a tissue-specific expression pattern that mimics the pattern observed in the human. In Sertoli cell cultures, FSH, insulin, retinol or tumour necrosis factor-{alpha} (TNF-{alpha}) stimulated hTf secretion, while testosterone alone had no effect. A combination of FSH, insulin, retinol and testosterone or a combination of TNF-{alpha} and retinol stimulated hTf secretion, but no additive effect was observed. CONCLUSION: Besides their well-known advantages, transgenic mice seem to be useful models to recapitulate the normal regulation of a human gene.

Key words: PAC/regulation/Sertoli/transferrin/transgenic mice


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Sertoli cells play a key role in the process of spermatogenesis. By secreting many proteins and by forming the blood–testis barrier, they provide a unique environment within the testis, in which germinal cells divide and differentiate into spermatozoa. Understanding the regulation of Sertoli cell functions is required to elucidate the mechanisms involved in the maintenance and control of spermatogenesis. Since methods for maintaining Sertoli cells in culture have been developed, numerous results have been obtained on hormonal gene regulation in Sertoli cells, mainly in rat (Sylvester and Griswold, 1994Go). In contrast, few data have been reported using human Sertoli cells, and most of the available data have been obtained from estrogen-treated patients or pubescent men (Lipshultz et al., 1982Go; Holmes et al., 1984Go; Foucault et al., 1992Go). Actually, for ethical reasons, it is difficult to obtain healthy testes of pre-pubertal men. To our knowledge, no human Sertoli cell line is available. However, understanding the regulation of proteins secreted by human Sertoli cells is an important requirement for identifying the causes of infertility in men.

Today, with the completion of the human genome project and the facility for handling it using long genomic DNA fragments contained in yeast artificial chromosome (YAC), phage P1-derived artificial chromosome (PAC) or bacterial artificial chromosome (BAC) vectors (Burke et al., 1987Go), human gene function and regulation can be studied in transgenic mouse models. Actually, these artificial chromosomes offer the opportunity potentially to capture all regulatory elements of a gene, and some studies have reported a tissue specificity of transgene expression (Nielsen et al., 1997Go; Sinn et al., 1999Go). This technology could provide an alternative means for studying the regulation of human genes involved in Sertoli cell functions. To address this point, we have chosen the human transferrin (hTf) gene as a model. The Tf gene contains 17 exons, included in 33.5 kbp, and encodes a 79 kDa glycoprotein. In all species studied, the protein is synthesized by the liver and by oligodendrocytes in the brain, and in rodents it is additionally synthesized by epithelial cells of the choroid plexus (Tu et al., 1991Go) and of the mammary gland (Lee et al., 1987Go). In all species, Tf is also expressed by Sertoli cells of the testis (Skinner and Griswold, 1980Go). In humans, many studies show that the Tf level in the seminal plasma is correlated with sperm yield (Sueldo et al., 1984Go; Orlando et al., 1985Go). A low concentration of Tf in the seminal plasma has been correlated with severity of oligospermia. Thus, Tf in the seminal plasma appears to be an important marker of the functional integrity of Sertoli cells. However, little is known on Tf gene regulation in human Sertoli cells (Holmes et al., 1984Go; Foucault et al., 1992Go). On the other hand, regulation of Tf gene expression has been studied extensively in rat Sertoli cell cultures. The regulation of Tf secretion has been shown to vary under a complex control involving hormones (FSH, insulin and testosterone), retinoids, cytokines and the surrounding germ cells (Skinner and Griswold, 1982Go; Perez-Infante et al., 1986Go; Sigillo et al., 1999Go). The cis-activating elements involved in Tf gene expression have been identified in human and mouse 3.6 kbp promoters by transient transfection in rat primary Sertoli cell cultures (Chaudhary and Skinner, 1998Go; Guillou et al., 1991Go; Schaeffer et al., 1993Go). The major response elements involved in regulating the hTf gene are conserved in the proximal mouse transferrin (mTf) promoter. In particular, proximal region I (PRI), proximal region II (PRII), a C/EBP-binding site, distal region I (DRI) and distal region II (DRII) are conserved and functional in the mTf promoter. The transcription factors necessary for the basal activity of these promoters may be similar. The work of Chaudhary and Skinner (1998Go) suggests that regulation of Tf transcription in response to various stimuli may be more complex and may require additional sequences upstream of the 3.0 kbp 5'-regulatory region as well as a combination of different transcription factors for the regulation of mouse or human Tf promoters.

On this basis, we have generated transgenic mice using a phage P1 containing 80 kbp of human DNA sequences encompassing the complete hTf gene. First, the expression pattern of the hTf gene was assayed. Furthermore, regulation of the human and mouse Tf genes by hormones, retinoids and cytokines in Sertoli cells purified from transgenic mice was studied.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Materials
Dibutyryl-cAMP (db-cAMP), forskolin, insulin, retinol and tumour necrosis factor-{alpha} (TNF-{alpha}) were obtained from Sigma-Aldrich Chimie S.AR.L. (St Quentin Fallavier, France). Porcine FSH was provided by J.L.Closset from the Laboratory of Endocrinology and Biochemistry (Liège, Belgium).

Animals
Transgenic mice for the hTf gene were generated in the C57/Bl6-SJL and C57/Bl6-DBA2 genetic backgrounds. A PAC containing the hTf gene used as a transgene was obtained from Genome Systems. The transgenic animals were screened for the presence of hTf in the blood. All transgenic mice used in this study were hemizygous for the hTf gene. Animals were fed with a standard laboratory diet and tap water ad libitum, and maintained on a half light:half dark photoperiod in a temperature-controlled room at 21–23°C. All animal studies were conducted in agreement with the Guides for Care and Use of Laboratory Animals (NIH Guide).

In situ hybridization
Mice were perfused intracardially with 4% paraformaldehyde. The livers, testes and brains were dissected in cold phosphate-buffered saline (PBS) (137 mmol/l NaCl, 2.7 mmol/l KCl, 1.4 mmol/l KH2PO4, 4.3 mmol/l Na2HPO4), post-fixed in 4% paraformaldehyde overnight at 4°C and embedded in paraffin. Slices were hybridized to an antisense riboprobe specific for the hTf gene that does not cross-react with endogenous mTf, or to a hTf sense riboprobe. The probes were labelled with [{alpha}-35S]UTP using SP6 and T7 polymerase or with digoxigenin-11-UTP (Roche Diagnostics, Meyla, France). The hybridization procedure using radioactive riboprobes has been described previously in detail (Saleh et al., 2003Go). The hybridization using digoxigenin riboprobes was carried out according to the manufacturer’s instructions.

Immunohistochemistry
Mice were perfused intracardially with 4% paraformaldehyde, then the brains were post-fixed for 1 h. The brains were included in tissue-tek (Sakura, The Netherlands) and immediately frozen. Cryostat sections (8 µm) were air-dried, then incubated in PBS containing 4% bovine serum albumin (BSA) for 1 h. Anti-mTf or anti-hTf antibodies (1:10 000) were incubated at 4°C overnight with 1 µg of human or mouse Tf, respectively. After saturation of unspecific sites by BSA, the sections were incubated with Tf antibodies overnight in a humid chamber at 4°C. After washes with PBS, and incubation with fluorescein isothiocyanate (FITC)-conjugated rabbit IgG (Jackson ImmunoResearch, Pennsylvania, PA) for 1 h at room temperature, the sections were counterstained with Evans blue and examined with a Leica microscope.

Cell preparation and culture
Tubular cells were isolated from 12- to 14-day-old mice. All steps of cell preparation were performed in L15 medium (Gibco-INVITROGEN, Cergy-Pontoise, France) with 100 µg/ml streptomycin, 100 U/ml penicillin G (Sarbach, Suresnes, France). Testes were decapsulated and digested by trypsin (2.5 mg/ml) and DNase (10 µg/ml) (Sigma-Aldrich Chimie S.AR.L.) for 25 min at 34°C under continuous shaking. Trypsin action was stopped by the addition of 2.5 mg/ml soybean trypsin inhibitor (Sigma-Aldrich Chimie S.AR.L.). The tubular fragments were allowed to sediment and the supernatant containing the interstitial cells was eliminated. The sedimented fragments were washed once before being disrupted by a glass potter. Finally, the cell clusters were washed and filtered through sterile surgical gauze. Sertoli cell preparations were plated in Petri dishes at 0.5 x 106 cells per 2 cm2 in Dulbecco’s modified Eagle’s medium (DMEM; Gibco-INVITROGEN) supplemented with 100 U/ml penicillin G, 2.5 µg/ml amphotericin B (fungizone), 100 µg/ml streptomycin, 2 mmol/l glutamine and 200 ng/ml vitamin E. Cells were cultured at 34°C in a humidified atmosphere of 5% CO2. The medium was replaced after 24 h to remove unattached cells, and plated tubular cells were treated with the different products as described in the Results. After 48 h of stimulation, culture media were collected, and mouse and human Tfs were measured.

The Sertoli cell preparation contains 3% peritubular cells and 20% germ cells. The peritubular cells were identified by alkaline phosphatase reaction, as previously described (Chapin et al., 1987Go). The germ cells were identified by classical histological methods.

Preparation of cellular extracts
One unit of organ weight was disrupted in 2 volumes of lysis buffer [15 mmol/l Tris pH 7.9, 60 mmol/l KCl, 15 mmol/l NaCl, 2 mmol/l EDTA, 0.4 mmol/l phenylmethylsulphonyl fluoride (PMSF)]. After four freeze/thaw cycles, the lysates were centrifuged at 6000 g for 10 min at 4°C and the supernatants were stored at –20°C.

Transferrin quantification
HTf was quantified by an antibody-sandwich enzyme-linked immunosorbent assay (ELISA), using two polyclonal antibodies directed against hTf, as previously described (de Arriba Zerpa et al., 2000Go). The detection limit was 0.1 ng/ml. HTf antibodies cross-reacted at <0.05% with mTf.

MTf was measured by radioimmunoassay. The method was performed as described previously (Cassia et al., 1997Go). The detection limit was 1 ng/ml. All standards and samples were assayed in triplicate. The specificity of the antibody was tested. MTf antibodies cross-reacted at <0.0001% with hTf.

Contamination of the brain with serum Tf was estimated by measuring the IgG level in the brain and the blood as described (Saleh et al., 2003Go).

Isolation of RNA and northern blot
Total RNAs were extracted from organs using acid guanidinium thiocyanate–phenol–chloroform extraction as previously reported (Chomczynski and Sacchi, 1987Go) and quantified by absorbance at 260 nm.

A 10 µg aliquot of total RNA was size-fractionated on 1% (w/v) agarose gels containing 1.1 mol/l formaldehyde, 20 mmol/l MOPS, 5 mmol/l sodium acetate and 1 mmol/l EDTA. After electrophoresis, the RNA was transferred overnight by capillary action to a nylon membrane (Nytran Super Charge, Schleicher and Schuell, Dassel, Germany) and covalently bound to the membrane by baking at 80°C for 2 h. Pre-hybridization of the membranes was performed in rotating bottles for 2 h at 42°C, in NorthernMax Prehybridization/Hybridization Buffer (Ambion, Inc., Austin, TX). The northern blots were hybridized with the following probes: (i) a 1.5 kbp DNA fragment of the mTf cDNA obtained after EcoRI hydrolysis of the clone named pMTF-5 (Chen and Bissell, 1987Go); (ii) a 494 bp 3' sequence of the human Tf gene; and (iii) a region of mouse cDNA 18S from Ambion, Inc. These probes were radiolabelled previously by a random-primed labelling reaction (Amersham, Buckinghamshire, UK) and [{alpha}-32P]dATP. Unincorporated nucleotides were removed by gel filtration chromatography on a Sepharose G50 column (Amersham Biosciences-Pharmacia, Saclay, France) and the probes were heat denatured. Hybridization was performed by adding 106 d.p.m./ml of the respective probes to the pre-hybridization mixture and by incubating overnight at 42°C. Membranes were washed twice for 5 min at room temperature in 2x SSC (0.3 mol/l NaCl, 30 mmol/l tri-sodium citrate pH 7.0), and twice for 15 min in NorthernMax High Stringency Wash buffer (Ambion, Inc.) at 42°C. Membranes were exposed to a PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA) at room temperature for 24 h before quantification using the ImageQuant software (Molecular Dynamics).

Statistics
Results are presented as means ± SD. Data analyses were performed using Mann–Whitney U-test for evaluation of differences between treated and control cells and between treated mTf/hTf ratios and control mTf/hTf ratios. The statistical computer program StatXact6 was used. All tests were performed two-sided and P < 0.05 was considered as statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Generation of Tf transgenic mice and characterization of Tf expression
The human genomic fragment was tested by FIGE (field inversion gel electrophoresis), and the results revealed the presence of an 80 kbp insert. The hTf gene encompasses 33.5 kbp of genomic sequences (Schaeffer et al., 1987Go) and is flanked by 15 kbp of DNA sequences in the 5' and by 31 kbp in the 3' direction (Figure 1A). This 80 kbp human fragment was compared with data generated by the Human Genome Resource (NCBI). Only the hTf gene was present in the fragment, but a putative gene named PRO2086 was also identified in the fourteenth intron of the human gene (Figure 1A). For comparison, in the mouse gene, an expressed sequence tag (EST) named EST BE132832 was also located in the fourteenth intron. Transgenic lines resulting from the pronuclear injection of the 80 kbp transgene were established as two lines, namely lines 801 and 803, established in the C57BL6/DBA2 and C57BL6/SJLJ genetic backgrounds, respectively.



View larger version (45K):
[in this window]
[in a new window]
 
Figure 1. Characterization of the transgene. (A) Structure of the transgene. The vector sequences are in hashed black lines, the Tf introns are in grey and the exons are in black. A horizontal black box indicates the position of the putative PRO2086 gene. The size of the transferrin gene and of the 5'- and 3'-flanking sequences is indicated. (B) Representative northern blot with mRNA extracted from transgenic line 803 (+) or wild-type (–) mice hybridized with hTf, mTf and 18S cDNA probes.

 
Tissue-specific expression of the transgene in the liver, the brain and the testis was demonstrated by detection of hTf mRNA by northern blot. A representative northern blot is shown in Figure 1B. In the testis, the hTf mRNA was detected by in situ hybridization inside the seminiferous tubules (Figure 2C). In the liver, the hTf gene is expressed in hepatocytes (Figure 2A). In the brain, the hTf mRNA was detected only in oligodendrocytes, and not in the choroid plexus cells (Figure 2E). By immunofluorescence, we confirmed that hTf was only expressed by oligodendrocytes (Figure 3A and C). In contrast, mTf was detected in both oligodendrocytes and choroid plexus (Figure 3B and D).



View larger version (170K):
[in this window]
[in a new window]
 
Figure 2. Tissue specificity of human transferrin. By in situ hybridization, the hTf mRNA was detected in the liver (A and B), testis (C and D) and brain (E and F) from sections of transgenic mice (line 803). Slides were hybridized with the antisense hTf probe (A, C and E) or the sense hTf probe (B, D and F). Probes were labelled with digoxygenin (A and B) or [32P]dCTP (C, D, E and F). Liver section (x100). Testis sections (x100) are shown on a black background and brain sections (x25) are shown on a white background. In the photograph of the brain sections, the choroid plexus (*) and oligodendrocytes (arrows) are shown.

 


View larger version (62K):
[in this window]
[in a new window]
 
Figure 3. Expression of human and mouse transferrin in the brain of hTf transgenic mouse line 803. Immunostaining with hTf antibody (A and C) or mTf antibody (B and D). The choroid plexus (*) is shown in (A) and (B) (x 100) and oligodendrocytes of the corpus callosum in (C) and (D) (x250).

 
The secretion level of hTf is weaker than that of mTf
To study the level of expression of Tf genes in transgenic mice, we compared the hTf and mTf yield in different organs and biological fluids. In both lines, the level of hTf expression was weaker than that of mTf (Table I). Nevertheless, in the brain, hTf was relatively more abundant than in other organs tested (50–100% compared with mTf). In liver, blood and the seminiferous tubules, hTf levels only reached 10–20% of mTf levels. Sertoli cells were isolated from seminiferous tubules and hTf levels were found to be 15–40% those of mTf. In milk, very little hTf was detected when compared with mTf (between 1 and 2%). This observation is in agreement with the fact that there is little Tf in human milk, in contrast to murine milk (Masson and Heremans, 1971Go).


View this table:
[in this window]
[in a new window]
 
Table I. The expression level of human transferrin is weaker than that of mouse transferrin in transgenic mice
 
Regulation of transferrin secretion in Sertoli cells
In the testis, FSH is an important hormonal regulator of Tf expression. Sertoli cells from wild-type or transgenic mice were cultured, and the effects of FSH on Tf secretion were estimated. In lines 801 and 803, a dose response of FSH stimulation was observed (Figure 4). In both lines, 10–500 ng/ml FSH increased the amount of hTf secreted up to 1.5-fold. A similar range of hormone concentration stimulated mTf by 2-fold. Of note, whereas the same basal levels of mTf were secreted in both cell lines, the quantity of basal hTf secreted differed (6 ng/ml in line 801 versus 18 ng/ml in line 803). For line 801, no difference in sensitivity was observed between hTf and mTf stimulation by FSH as levels increased in the same fashion for all concentrations tested. On the other hand, whereas hTf secretion in line 803 was already stimulated at 10 ng/ml of FSH, mTf secretion was increased only beyond this concentration (Figure 4).



View larger version (14K):
[in this window]
[in a new window]
 
Figure 4. FSH stimulates mouse and human transferrin secretion. The Sertoli cells isolated from the testes of mouse line 801 (A) and 803 (B) were incubated with FSH at different concentrations for 48 h. hTf and mTf were assayed in the medium by specific hTf ELISA and mTf radioimmunoassay. Note that the scales are different for mTf versus hTf.

 
A difference in maximum FSH sensitivity between both lines also exists. In line 801, FSH stimulated hTf secretion at concentrations of up to 5000 ng/ml. In line 803, the maximal effect was observed at 100 ng/ml FSH.

FSH is known to mediate its effects by increasing the production of cAMP in Sertoli cells. Hence, we have tested whether forskolin and db-cAMP stimulate FSH-induced hTf expression. In both lines, the stimulatory effects of FSH on hTf secretion can be mimicked by forskolin (1 µmol/l) or by db-cAMP (10–4 mol/l) (Table II). Moreover, FSH, db-cAMP or forskolin also stimulated significantly mTf secretion in wild-type C57/Bl6-DBA2 and C57/Bl6-SJL or in transgenic mice. In line 803, db-cAMP stimulated hTf secretion more efficiently than mTf.


View this table:
[in this window]
[in a new window]
 
Table II. FSH, forskolin and db-cAMP stimulate secretion of mouse and human transferrin
 
Next, we have studied various factors (insulin, testosterone, retinol and TNF-{alpha}) at concentrations known to stimulate Tf secretion. These factors have the same effects on mTf derived from wild-type or transgenic mice (Table III). Testosterone (1 µmol/l) and retinol at low concentration (0.35 µmol/l) had no effect on hTf or mTf. In contrast, 5 µg/ml insulin, 3.3 µmol/l retinol, 2.5 ng/ml TNF-{alpha} and FIRT mix composed of FSH (100 ng/ml), insulin (5 µg/ml), retinol (0.35 µM) and testosterone (1 µmol/l) or a combination of 2.5 ng/ml TNF-{alpha} and 3.3 µmol/l retinol significantly stimulated hTf secretion as well as that of mTf. A combination of FIRT and TNF-{alpha} + retinol had no synergistic effect. Whereas the factors used have the same qualitative effect (stimulation or no effect) on mTf or hTf, hTf expression exhibits a tendency to be stimulated more potently than mTf, which is statistically significant with retinol 3.3 µmol/l.


View this table:
[in this window]
[in a new window]
 
Table III. Effect of factors relevant to the Sertolian function on the secretion of mouse and human transferrin from transgenic or wild-type mice
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
We have set up a model which emphasizes the usefulness of transgenic mice to study the mechanisms which regulate the expression of a human gene involved in Sertoli cell functions.

This gene is the 80 kbp long Tf gene comprising its long 5'- and 3'-regulatory sequences. We obtained two lines of transgenic mice that both exhibited the expected tissue-specific pattern of hTf expression in the liver and in the testis. In the brain, the human gene was expressed solely in oligodendrocytes, while the expression of the mTf gene was detected in the oligodendrocytes and also in the choroid plexus. These observations are in agreement with a previous report indicating that the Tf mRNA from mouse, but not from human, is weakly expressed in the choroid plexus (Tu et al., 1991Go). In addition, we have found a high concentration of mTf and a weak concentration of hTf in milk. High quantities of Tf in the milk of rodents in contrast to human were reported previously (Masson and Heremans, 1971Go). More recently, it has been shown that Tf was expressed by glial Müller cells of human and macaque retinas. In transgenic mice, hTf was also detected in glial Müller cells (Y.Courtois and J.C.Jeanny, personal communication). Thus, and probably due to its size, the transgene seems to have maintained most characteristics of human tissue specificity despite being embedded in a murine genome. This was not observed in previous experiments using 3.9 kbp of Tf 5'-regulatory sequences upstream of the LacZ gene (Cassia et al., 1997Go), although the proximal regulatory region of the transferrin gene is highly conserved in human and mouse. Thus, our present results strongly suggest that the species specificity of Tf expression is likely to be born by cis-elements included in the 80 kbp fragment, confirming that BACs, PACs or YACs are useful tools in transgenesis to obtain tissue specificity of a gene, because they permit the manipulation of long DNA fragments.

Although tissue specificity of the hTf gene expression in both lines of transgenic mice was conserved by using 80 kbp of human DNA, we observed that in both lines and in all organs or fluids except the brain, the hTf gene was expressed at lower levels than the mTf gene (only 10–40% of mTf levels) and was also reduced when compared with normal levels in human plasma (2.5 mg/ml) (Bowman et al., 1988Go). Several hypotheses can be proposed to explain this observation: first, the neighbouring genes in which the transgene was inserted could reduce the expression of hTf. This, however, is unlikely since the level of expression was reproducibly weaker in both transgenic lines which were generated independently, with probably different insertion sites. Secondly, the DNA fragment inserted may not contain all regulatory elements controlling the level of expression, such as long distance enhancers. Thirdly, some species-specific regulatory elements, present in the 80 kbp fragment and controlling the level of expression, may not be recognized as efficiently by mouse factors. Fourthly, a maximal level of Tf could be reached and mTf could regulate the hTf level. We could address this question by generating mice with a knocked-out mTf gene and expressing the hTf gene.

It has been reported that Sertoli cells from estrogen-treated patients or young men (16–47 years) produce Tf (Holmes et al., 1984Go; Foucault et al., 1992Go). We show that primary Sertoli cell cultures from mice transgenic for the hTf gene could express and secrete hTf constitutively. The regulation of hTf and mTf expression by different factors known to control Sertoli cell functions was also analysed in our model. A stimulatory effect of FSH on Tf expression has already been reported in rat (Skinner and Griswold, 1982Go), human (Holmes et al., 1984Go) and murine (Huleihel and Lulenfeld, 2002Go) Sertoli cells. Here we show that FSH as well as db-cAMP or forskolin are able to stimulate hTf and mTf expression in Sertoli cell cultures from both lines of transgenic mice. However, the magnitude of Tf stimulation by FSH we observed is difficult to compare with other data obtained in humans (Holmes et al., 1984Go; Foucault et al., 1992Go) because of differences in culture conditions (e.g. stimulation time, days in culture).

Regulation of the mTf gene in mouse Sertoli cells has been studied previously only in response to FSH or interleukins (Huleihel and Lunenfeld, 2002Go). We show here that insulin and retinol can also stimulate mTf while testosterone was ineffective. These results were similar to those obtained in rat (Skinner and Griswold, 1982Go; Hoeben et al., 1996Go; Sigillo et al., 1999Go). Interestingly, some differences between mouse and rat have been highlighted in our study. In mouse, a low concentration of retinol (0.35 µmol/l) had no effect on mTf secretion, 2.5 ng/ml TNF-{alpha} stimulated mTf, and a FIRT mix or a combination of TNF-{alpha} and 3.3 µmol/l retinol did not have additive effects. Thus, regulation of Tf seems to be different between rat and mouse. Of note, regulation of mTf secretion was not altered by the presence of the hTf gene in transgenic animals. Whatever the factors used, the effects were identical in wild-type or transgenic mice.

Up to now, two studies have analysed the regulation of hTf by insulin, retinol and testosterone in cultured human Sertoli cells (Holmes et al., 1984Go; Foucault et al., 1992Go). Both have shown that insulin and retinol stimulate the secretion of hTf, although the retinol data are not entirely conclusive. In transgenic mice, we confirmed that insulin and retinol stimulate hTf. In contrast, the effects of testosterone are a matter of debate: some authors (Holmes et al., 1984Go) did not observe stimulation of hTf, whereas others (Foucault et al., 1992Go) did. In our model, no testosterone effect was observed. Thus, we conclude that expression of the hTf gene in mouse Sertoli cells seems to be regulated as in human Sertoli cells. This observation prompted us to analyse the effects of other factors known to regulate Tf secretion in rat Sertoli cells and not yet studied in human Sertoli cells. Our study shows that the hTf gene could be regulated by the same factors as the mTf gene, but we cannot exclude that the effects observed are due to the murine context. It would be interesting to verify these effects on human Sertoli cells in culture, but such a study is difficult for ethical reasons. Another way to evaluate the influence of the murine context in regulation of gene expression could be to study mice transgenic for the rat Tf gene. Nevertheless, we can observe that the hTf gene is regulated in transgenic mice by the same factors as the mTf gene, but a difference in the magnitude of the response exists as the hTf gene is stimulated more potently than the mTf gene by 3.3 µmol/l retinol.

The model of hTf transgenic mice can bring new enlightenment on the regulation of this gene in Sertoli cells. Moreover, transgenic mice offer the opportunity to study the impact of factors during different events of testis development, for example the fetal period or before and after the blood–testis barrier has formed.

In summary, we demonstrate here that the use of hTf transgenic mice constructed with a large fragment from the human genome allows study of the spatial specificity of transgene expression and its regulation. Secondly, since hTf gene expression seems to be regulated differently from that of the rat Tf gene, it is possible to envisage using these ‘humanized’ mice to study the impact of pharmacological agents on this expression. In Sertoli cells, expression of a human transgene in a murine context may mimic the human physiology more closely than the rat model.


    Acknowledgements
 
The authors thank P.Crépieux for critical reading of the manuscript, M.Vignuzzi for reviewing the English of the manuscript, A.Aubert for his assistance with the statistical analysis, and J.C.Lenoir for his help in designing the figures. We also thank Atelier de transgenèse (Institut Pasteur Paris), E.Jean-Pierre for his help in mice manipulation, J.Dupont for her expert tutelage in mammary gland dissection, C.Cahier and J.C.Braguer for animal husbandry. C.Lécureuil was supported by a fellowship from the INRA and the Région Centre.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Bowman BH, Yang FM and Adrian GS (1988) Transferrin: evolution and genetic regulation of expression. Adv Genet 25,1–38.[ISI][Medline]

Burke DT, Carle GF and Olson MV (1987) Cloning of large segments of exogenous DNA into yeast by means of artificial chromosome vectors. Science 236,806–812.[ISI][Medline]

Cassia R, Besnard L, Fiette L, Espinosa de los Monteros A, Ave P, Py MC, Huerre M, de Vellis J, Zakin MM and Guillou F (1997) Transferrin is an early marker of hepatic differentiation and its expression correlates with the postnatal development of oligodendrocytes in mice. J Neurosci Res 50,421–432.[CrossRef][Medline]

Chapin RE, Phelps JL, Miller BE and Gray TJ (1987) Alkaline phosphatase histochemistry discriminates peritubular cells in primary rat testicular cell culture. J Androl 8,155–161.[Abstract/Free Full Text]

Chaudhary J and Skinner MK (1998) Comparative sequence analysis of the mouse and human transferrin promoters: hormonal regulation of the transferrin promoter in Sertoli cells. Mol Reprod Dev 50,273–283.[CrossRef][Medline]

Chen LH and Bissell MJ (1987) Transferrin mRNA level in the mouse mammary gland is regulated by pregnancy and extracellular matrix. J Biol Chem 262,17247–17250.[Abstract/Free Full Text]

Chomczynski P and Sacchi N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal Biochem 162,156–159.[CrossRef][ISI][Medline]

deArribaZerpa GA, Saleh MC, Fernandez PM, Guillou F, Espinosa de los Monteros A, de Vellis J, Zakin MM and Baron B (2000) Alternative splicing prevents transferrin secretion during differentiation of a human oligodendrocyte cell line. J Neurosci Res 61, 388–395.[CrossRef][ISI][Medline]

Foucault P, Carreau S, Kuczynski W, Guillaumin JM, Bardos P and Drosdowsky MA (1992) Human Sertoli cells in vitro. Lactate, estradiol-17 beta and transferrin production. J Androl 13,361–367.[Abstract]

Guillou F, Zakin MM, Part D, Boissier F and Schaeffer E (1991) Sertoli cell-specific expression of the human transferrin gene. Comparison with the liver-specific expression. J Biol Chem 266,9876–9884.[Abstract/Free Full Text]

Hoeben E, Van Damme J, Put W, Swinnen JV and Verhoeven G (1996) Cytokines derived from activated human mononuclear cells markedly stimulate transferrin secretion by cultured Sertoli cells. Endocrinology 137,514–521.[Abstract]

Holmes SD, Lipshultz LI and Smith RG (1984) Regulation of transferrin secretion by human Sertoli cells cultured in the presence or absence of human peritubular cells. J Clin Endocrinol Metab 59,1058–1062.[Abstract]

Huleihel M and Lunenfeld E (2002) Involvement of intratesticular IL-1 system in the regulation of Sertoli cell functions. Mol Cell Endocrinol 187,125–132.[CrossRef][Medline]

Lee EY, Barcellos-Hoff MH, Chen LH, Parry G and Bissell MJ (1987) Transferrin is a major mouse milk protein and is synthesized by mammary epithelial cells. In vitro Cell Dev Biol 23,221–226.[ISI][Medline]

Lipshultz LI, Murthy L and Tindall DJ (1982) Characterization of human Sertoli cells in vitro. J Clin Endocrinol Metab 55,228–237.[Abstract]

Masson PL and Heremans JF (1971) Lactoferrin in milk from different species. Comp Biochem Physiol 39,119–129.[CrossRef][ISI]

Nielsen LB, McCormick SP, Pierotti V, Tam C, Gunn MD, Shizuya H and Young SG (1997) Human apolipoprotein B transgenic mice generated with 207- and 145-kilobase pair bacterial artificial chromosomes. Evidence that a distant 5'-element confers appropriate transgene expression in the intestine. J Biol Chem 272,29752–29758.[Abstract/Free Full Text]

Orlando C, Caldini AL, Barni T, Wood WG, Strasburger CJ, Natali A, Maver A, Forti G and Serio M (1985) Ceruloplasmin and transferrin in human seminal plasma: are they an index of seminiferous tubular function? Fertil Steril 43, 290–294.[ISI][Medline]

Perez-Infante V, Bardin CW, Gunsalus GL, Musto NA, Rich KA and Mather JP (1986) Differential regulation of testicular transferrin and androgen-binding protein secretion in primary cultures of rat Sertoli cells. Endocrinology 118,383–392.[Abstract]

Saleh MC, Espinosa de los Monteros A, de Arriba Zerpa GA, Fontaine I, Piaud O, Djordjijevic D, Baroukh N, Garcia Otin AL, Ortiz E, Lewis S et al. (2003) Myelination and motor coordination are increased in transferrin transgenic mice. J Neurosci Res 72,587–594.[CrossRef][Medline]

Schaeffer E, Lucero MA, Jeltsch JM, Py MC, Levin MJ, Chambon P, Cohen GN and Zakin MM (1987) Complete structure of the human transferrin gene. Comparison with analogous chicken gene and human pseudogene. Gene 56,109–116.[CrossRef][ISI][Medline]

Schaeffer E, Guillou F, Part D, Zakin MM, Boissier F, Py MC and Cohen GN (1993) A different combination of transcription factors modulates the expression of the human transferrin promoter in liver and Sertoli cells. Cell type-specific expression of the human transferrin gene. Role of promoter, negative and enhancer elements. J Biol Chem 268,23399–23408.[Abstract/Free Full Text]

Sigillo F, Guillou F, Fontaine I, Benahmed M and Le Magueresse-Battistoni B (1999) In vitro regulation of rat Sertoli cell transferrin expression by tumor necrosis factor alpha and retinoic acid. Mol Cell Endocrinol 148,163–170.[CrossRef][Medline]

Sinn PL, Davis DR and Sigmund CD (1999) Highly regulated cell type-restricted expression of human renin in mice containing 140- or 160-kilobase pair P1 phage artificial chromosome transgenes. J Biol Chem 274,35785–35793.[Abstract/Free Full Text]

Skinner MK and Griswold MD (1980) Sertoli cells synthesize and secrete transferrin-like protein. J Biol Chem 255,9523–9525.[Abstract/Free Full Text]

Skinner MK and Griswold MD (1982) Secretion of testicular transferrin by cultured Sertoli cells is regulated by hormones and retinoids. Biol Reprod 27,211–221.[Medline]

Sueldo C, Marrs RP, Berger T, Kletzky OA and O’Brien TJ (1984) Correlation of semen transferrin concentration and sperm fertilizing capacity. Am J Obstet Gynecol 150, 528–531.[Medline]

Sylvester SR and Griswold MD (1994) The testicular iron shuttle: a ‘nurse’ function of the Sertoli cells. J Androl 15,381–385.[Abstract/Free Full Text]

Tu GF, Achen MG, Aldred AR, Southwell BR and Schreiber G (1991) The distribution of cerebral expression of the transferrin gene is species specific. J Biol Chem 266,6201–6208.[Abstract/Free Full Text]

Submitted on February 9, 2004; accepted on April 8, 2004.





This Article
Abstract
Full Text (PDF )
All Versions of this Article:
19/6/1300    most recent
deh297v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (1)
Request Permissions
Google Scholar
Articles by Lécureuil, C.
Articles by Guillou, F.
PubMed
PubMed Citation
Articles by Lécureuil, C.
Articles by Guillou, F.