Gene silencing by RNAi in mouse Sertoli cells.

BACKGROUND
RNA interference (RNAi) is a valuable tool in the investigation of gene function. The purpose of this study was to examine the availability, target cell types and efficiency of RNAi in the mouse seminiferous epithelium.


METHODS
The experimental model was based on transgenic mice expressing EGFP (enhanced green fluorescent protein). RNAi was induced by in vivo transfection of plasmid vectors encoding for short hairpin RNAs (shRNAs) targeting EGFP. shRNAs were transfected in vivo by microinjection into the seminiferous tubules via the rete testis followed by square wave electroporation. As a transfection reporter, expression of red fluorescent protein (HcRed 1) was used. Cell types, the efficiency of both transfections and RNAi were all evaluated.


RESULTS
Sertoli cells were the main transfected cells. A reduction of about 40% in the level of EGFP protein was detected in cells successfully transfected both in vivo and in vitro. However, the efficiency of in vivo transfection was low.


CONCLUSION
In adult seminiferous epithelial cells, in vivo post-transcriptional gene silencing mediated by RNAi via shRNA is efficient in Sertoli cells. Similar levels of RNAi were detected both in vivo and in vitro. This also indicates that Sertoli cells have the necessary silencing machinery to repress the expression of endogenous genes via RNAi.


Background
RNA interference (RNAi) describes any process in which double stranded RNA (dsRNA) triggers post-transcriptional gene silencing. Strategies for inducing gene silencing, either for the study of gene function or in a therapeutic context, have been developed [1]. Small interference RNAs (siRNAs) and short hairpin RNAs (shRNAs) have been used in vitro and in vivo for interfering with RNA [2][3][4][5]. siRNAs are dsRNAs of 21-23 base pairs (bp) generated by chemical synthesis [6], enzymatic cleavage [7] or expression systems [8], while shRNAs are dsRNA molecules that mimic endogenous pre-micro RNAs (pre-miR-NAs). shRNAs consist of two palindromic sequences of 19-29 nucleotides (nt) with a short loop of singlestranded RNA (4-10 nt) at one end [9]. The RNAse III family of nucleases known as 'Dicer' binds and cleaves both pre-miRNAs and shRNAs into their mature 21-25 bp forms [9][10][11][12]. One strand of these miRNAs or siRNAs is incorporated into the RNA-induced silencing complex (RISC), which then either identifies, binds and cleaves the complementary messenger RNA [13,14] or induces translational repression [15]. Recent work indicates that shR-NAs are more potent inducers of RNAi than siRNAs [16].
Silencing of specific mRNAs by RNAi has been used in vivo in the eye [17,18], brain [19][20][21][22], lung [23][24][25][26], skeletal muscle [27][28][29][30], liver, kidney, spleen [31][32][33][34][35][36][37][38][39] skin [40], and pancreas [41]. In the testis, the seminiferous epithelium of adults is organized into a complex structure composed of the germ cells and Sertoli cells. Sertoli cells, a somatic cell type, extend from the basement membrane of the seminiferous tubules to reach the lumen. The architectural pattern of these cells provides a structural framework for Sertoli cell-Sertoli cell and Sertoli cell-germ cell interactions. These interactions are based on intimate contacts through different types of junctions (e.g. occluding junctions, anchoring junctions and communicating junctions), supporting a specific microenvironment required by developing germ cells [42][43][44][45]. This work reports the use of a transgenic mouse model expressing EGFP to determine which cells of the seminiferous epithelium are preferentially transfected by shRNAcoding plasmids for the induction of gene silencing and its efficiency. In vitro experiments were also performed to verify the efficiency of RNAi in Sertoli cells, the main transfected target cell seen in in vivo transfections.

Experimental animals
All the mice (Mus musculus) used in these experiments were bred at the Animal Care Facility of the Centro de Investigaciones Biológicas (CIB-CSIC) on a 12L:12D cycle. Male mice of the C57BL/6J wild type were used to investigate in vivo transfection efficiency. The C57BL/6 TgN(act-EGFP)OsbC14-Y01-FM131 (FM131) [49] transgenic mouse line, which constitutively expresses EGFP, was provided by RIKEN BRC (Japan). All procedures were performed according to the guidelines of the CSIC Bioethics Committee.

Plasmids
Plasmid pEGFP-N1 (Clontech, Palo Alto, CA, USA), expressing EGFP as a reporter, was used as an in vivo transfection control. Plasmid pGtoR (a kind gift of Dr. Masaru Okabe, University of Osaka, Japan) was used to induce RNAi in EGFP. pGtoR contains the RNA polimerase III promoter H1 driving the expression of an shRNA containing 21 nt sense and antisense sequences homologous to an EGFP encoding region (shRNA-EGFP), as well as the CAG promoter controlling the expression of HcRed1 protein [50].
A vector called pRed, used as a negative control, was generated by digestion of pGtoR with BamH1 and HindIII followed by religation to eliminate the H1-shRNA-EGFP cassette. Consequently, pRed only expresses the HcRed1 protein.

In vivo electroporation
Male mice of 30-45 days post-natal (dpn) were anaesthetized by an intraperitoneal injection of Rompun (Bayer, Kiel Germany)/Ketolar (Pfizer, Dublin Ireland) solution (315 μl/Kg; 84 mg/Kg respectively). After opening the abdominal cavity, the testes were exposed under a binocular microscope as previously described [47]. Approximately 20 μl of plasmid DNA in TE buffer (10 mM Tris, and 1 mM EDTA, pH adjusted to 7.5) (3 μg/μl) containing nigrosine (1 mg/ml) as a tracer was slowly microinjected into the rete testis using a 40-70 μm in diameter glass micropipette (Fig 1). Trypan blue, the standard tracer for procedures of this kind, was ruled out due to its autofluorescence. For in vivo electroporation, each testis was held between tweezer-type electrodes (model 520, 7 mm diameter, BTX, San Diego, CA) briefly soaked in PBS, and two sets of four electric pulses of square wave were applied (using an electric pulse generator ECM 830 [BTX]). Each pulse provided 50 V for 50 ms; the interval between the pulses was 950 ms [51]. The testes were then returned to the abdominal cavity and the skin stitched closed. Four days later the mice were sacrificed and the testes removed for analysis.

Cytological examination
The testes were fixed with 4% paraformaldehyde in PBS, and passed through a series of 10, 20 and 30% sucrose. The samples were then placed in Tissue-Tek OCT (Sakura Finetek, The Netherlands) and frozen on dry ice. Cryosections (10 μm thick) were processed for histological examination by fluorescent microscopy using an inverted microscope (Nikon ECLIPSE TE300) (Tokio, Japan). In each experimental condition 10 mice were examined and 50 to 100 sections of the whole testis were assessed per mouse. A histopathological evaluation of the testis sections was performed on each specimen. Cultured Sertoli cells were also analysed by fluorescence microscopy. In both cases, Hoechst 33258 was used to counterstain the cell nuclei with 5 min incubation of the dye in PBS at 15 μg/ml.

Isolation of primary Sertoli cells and their culture
Sertoli cells were isolated from FM131 mice as previously described [52] with minor modifications. As mature Sertoli cells can not be efficiently cultured, testes from 17 days post-natal (dpn) males were decapsulated in PBS, cut into small fragments and digested in DMEM:Ham's F12 medium (1:1, Gibco BRL, Eggenstein, Germany) containing 2% foetal bovine serum (FBS) (ICN Biomedical, Costa Mesa, CA, USA), 0.2 mg/ml collagenase-dispase (Roche, Mannheim, Germany) and 0.1 mg/ml DNAse I (Roche) for 30 min at 32°C. The resultant seminiferous tubule fragments were washed with DMEM:F12 followed by two additional digestions under the same conditions, and then washed again with DMEM:F12. This material was repeatedly passed through an 18 1/2 G needle and the disaggregated cells were collected by filtration through a 70 μm Cell Strainer (BD Falcon, Lexington, TN, USA). The cells were incubated with continuous shaking in DMEM:F12 containing 2% FBS, 0.4 mg/ml hyaluronidase I-S (Sigma St Louis, MO, USA) and 0.1 mg/ml DNAse I for 30 min at 32°C. The sample was then centrifuged at 200 g for 10 min. The Sertoli cells obtained were resuspended in DMEM:F12 with 10% FBS and allowed to settle (20 min at 32°C). The settled cells were cultured at 32°C in a 5% CO2 atmosphere for three days in DMEM:F12 supplemented with 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin and 1× insulin-transferrin-sodium selenite media supplement (ITSS) (Sigma). The germ cells that had residually attached to the Sertoli cells were removed by hypotonic treatment with 20 mM Tris-HCl, pH 7.4 at 20°C for 3 min, and were cultured in supplemented DMEM:F12 medium.
To discern the presence of potential contaminant cells, analysis of transferrin (Trf) expression, as a Sertoli cell marker, was carried out by RT-PCR. As we previously reported to detect other potential contaminant cells RT-PCR analysis was performed for the expression of Hsd17 (17beta-hydroxysteroid dehydrogenase) as a Leydig cell marker and S16 (ribosomal protein) were also assessed as negative and positive controls respectively [53].

In vitro transfection
Sertoli cells growing in vitro in wells were transfected with the different plasmids using 0.4 μg of plasmid per well (1.9 cm 2 ) by FUGENE™6 reagent (Roche) according to the manufacturer's instructions. The cells were harvested three, five and seven days post-transfection. Each experiment was performed three times.

Flow cytometry analysis
Monocellular suspensions of the testis cells were obtained from in vivo electroporated testes and controls by digestion of the tubules following a procedure similar to that performed for the isolation of the Sertoli cells. After hyaluronidase digestion, the monodispersed cells from the seminiferous epithelium were resuspended in PBS. Sertoli cells from in vitro cultures and monocellular suspensions of seminiferous tubule cells were then analysed Different phases of in vivo microinjection of vectors into tes-tis tubules through the rete testis

Results and Discussion
In vivo transfection To evaluate the efficiency of in vivo gene silencing in mouse testis, after completing the first wave of spermatogenesis, we first characterized the efficiency to deliver plasmid DNA into the cells of the seminiferous epithelium. To determine which cell types were preferentially transfected either pEGFP-N1 or pGtoR was used. The cytological detection of green or red fluorescent proteins indicated that Sertoli cells (Fig. 2) were the cell type most commonly transfected (less than 1% of germ cells were also transfected). However, the efficiency of transfection of Sertoli cells was always less than 10% although no differences were found between the plasmids used. As previously described, altering the experimental conditions, i.e., increasing voltage and/or the number of electrical pulses, was found to damage the seminiferous epithelium as assessed by histopathological analysis (data not shown) [51].

In vivo gene silencing of seminiferous epithelial cells
Since shRNA molecules can induce potent gene silencing [9,16,[54][55][56], vectors expressing shRNA were used in the present work to confirm the efficiency of silencing of a specific gene both in vivo in seminiferous tubule cells and in in vitro cultures of Sertoli cells.
The vector pHcRed1-shRNA-EGFP (pGtoR) [50] allows detection of transfected cells expressing shRNA-EGFP, based on the co-expression of the red fluorescence protein HcRed1. Nevertheless, in tissue sections, an accurate measure of the fading of green fluorescence at the cellular level is difficult to detect due to the frequent superposition of adjacent cells and to the variability of EGFP expression between different cell types of the seminiferous epithelium [57]. To quantify the level of post-transcriptional silencing of EGFP in transfected cells, the reduction in green fluorescence from monocellular dispersions of seminiferous tubule cells from in vivo transfected testis was measured by fluorescence activated cell sorting (FACS). Red fluorescent cells were selected four days after transfection, and a reduction of 41.94% of green fluorescence was detected in the cells transfected with pGtoR compared to those transfected with pRed (used as a control). A significant difference in green fluorescence (t test; p = 0.047) was observed between red fluorescent cells depending on the vector used (pGtoR or pRed) (Fig. 3). This difference can only be interpreted as a specific silencing of EGFP mediated by shRNA-EGFP.

Gene silencing in primary cultures of Sertoli cells
Since Sertoli cells were the main cell type transfected in vivo, transfection and gene silencing were assessed in cultured primary Sertoli cells. Sertoli cells from C57B/6J mice were independently transfected with pEGFP-N1, pGtoR or pRed vectors. Comparative analysis of the transfection efficiency showed that 38% of the cells had been transfected with pEGFP-N1 and 25% with pRed or pGtoR.
In order to compare gene silencing by RNAi in cultured primary Sertoli cells and in in vivo transfected cells, Sertoli cells isolated from the EGFP transgenic mouse line FM131 were cultured. A reduction in green fluorescence due to EGFP protein was observed in the red fluorescent cells transfected with pGtoR compared to those that were not transfected (Fig. 4).
To indirectly quantify the silencing of EGFP, EGFP fluorescence intensity in transfected Sertoli cells was determined by using flow cytometry. Compared to the cells transfected with pRed, a significant reduction in EGFP fluorescence intensity was seen in the cells transfected with pGtoR at three (p = 0.0165), five (p = 0.0199) and seven days (p = 0.0171) post-transfection. The reduction of EGFP fluorescence was more significant at seven days (41.77%) and five days (36.55%) than at three days posttransfection (28.63%) (Fig. 5).
The efficiency of vector transfection in vivo after electroporation, the method employed here, was relatively low with all the vectors used. This might be attributed to an intrinsic characteristic of these cells or to the nature of the constructs used in these experiments. However, a preference for Sertoli cells and a similar transfection rate were observed with both the pEGFP-N1 and pGtoR vectors.
In the mature mouse, Sertoli cells occupy approximately 15-20% of the volume of the seminiferous epithelium and a large proportion of the Sertoli cell surface is in contact with elongated spermatids and the tubular lumen Hasuwa et al.
[50] developed a transgenic approach to deliver EGFP-targeted shRNAs into mice ubiquitously expressing EGFP. In this way, they studied the effectiveness of transgene-mediated gene silencing in different cells and tissues, however, no analysis in the testis was performed. We used the same vector and, as expected, the Sertoli cells were the main target cell type for transfection and EGFP silencing.
As the low efficiency of transfection of nonviral vectors is a technical limitation in the use of this approach to silence genes in seminiferous epithelium, alternative methodologies are also being explored.
Comparative results from flow cytometry analysis of EGFP-expression in seminiferous epithelium cells transfected in vivo with pRed or pGtoR

Conclusion
In conclusion, gene silencing by RNAi via shRNA, was demonstrated both in vivo and in primary culture of Sertoli cells. In Sertoli cells from the mouse model used, the reduction of 40% in the amount of target (EGFP) was significant. This also indicates that Sertoli cells have the necessary silencing machinery to repress the expression of endogenous genes via RNAi.

Competing interests
The authors declare that they have no competing interests.

Authors' contributions
EGG participated in designing the study and performed the experimental transfection of shRNAs. He also participated in the analysis and discussion of the results and drafted the manuscript. PPLC participated in the experimental design of the study, supervised the analysis, discussion of the results and critical revision of the manuscript. JdM was responsible for the conception, design, funding and supervision of this work. He also participated in the analysis and discussion of the results, drafting and critical revision of the manuscript. All authors read and approved the final manuscript.
Comparative results from flow cytometry analysis of EGFP-expressing cultured Sertoli cells in vitro transfected with the pGtoR or pRed