Spermatogenesis Associated 4 Promotes Sertoli Cell Proliferation Modulated Negatively by Regulatory Factor X1

Spermatogenesis associated 4 (Spata4), a testis-specific and CpG island associated gene, is involved in regulating cell proliferation, differentiation and apoptosis. To obtain insight into the role of Spata4 in cell cycling control, we characterized the promoter region of Spata4 and investigated its transcriptional regulation mechanism. The Spata4 promoter is unidirectional transcribed and possesses multiple transcription start sites. Moreover, we present evidence that regulatory factor X1 (RFX1) could bind the typical 14-bp cis-elements of Spata4 promoter, modulate transcriptional activity and endogenous expression of Spata4, and further regulate the proliferation of Sertoli cells. Overexpression of RFX1 was shown to down-regulate both the promoter activity and mRNA expression of Spata4, whereas knockdown of RFX1 demonstrated the opposite effects. Our studies provide insight into Spata4 gene regulation and imply the potential role of RFX1 in growth of Sertoli cells. RFX1 may have negative effect on cell proliferation of Sertoli cells via modulating Spata4 expression levels by binding the conserved 14-bp cis-elements of Spata4 promoter.


Introduction
Spermatogenesis associated 4 (Spata4) was initially identified in human testes and plays an important role in cryptorchidism development [1]. All of human, mouse and rat Spata4 genes have specifically high expression in testis, and among numerous genes that expressed in testis, only minority are testis specific genes [2][3][4][5]. Therefore, it was proposed that Spata4 participates in the spermatogenesis process [1], and may function during adolescence and maintain spermatogenesis ability, as transcript expression profiles indicated [1,6]. In addition, recent studies indicate that Spata4 modulates cell growth, proliferation and differentiation in various cell types [1,7,8]. Sertoli cells produce essential factors for germ cell development and modulate male fertility [9]. And these factors are of great importance in the process of spermatogenesis [10,11]. Clearly, the number and function of Sertoli cells determine testicular size, germ cell numbers and spermatozoa output [12]. Thus, Sertoli cell growth and proliferation may contribute to maintaining the normal physiological functions of male reproductive system.
CpG islands are often linked to promoter regions of genes [13]. The accepted definition of CpG island is regions of DNA longer than 200 bp, with a guanine/cytosine content above 50% and an ratio of observed to expected CpG presence above 0.6 [14]. The human, mouse and rat Spata4 genes showed excellent similarity of the promoter regions, especially a highly conserved region around the start codon between human and mouse Spata4 genes [6], and the CpG islands of these species also have a high similarity with each other, suggesting that Spata4 genes of these species may have some common transcription factors.
The RFX (regulatory factor for X-box) family of transcription factors consists of seven members in mammals and shares a highly conserved DNA-binding domain (DBD) [15,16]. Besides the central located DBD domain, a conserved C-terminal region exists in RFX1 protein [17], mediating the dimerization and transcriptional repression of winged-helix RFX1 protein. RFX1 is involved in transcriptional regulation of a variety of viral and cellular genes, possesses both transcription activation and repression properties [18]. RFX1 was implicated in positive regulation of transcription from enhancer of hepatitis virus (HBV) [17] and MHC class II genes [19,20], and negative effect was observed in transcriptional regulation of cellular genes including c-myc [21,22], collagen a2 [23,24], proliferating cell nuclear antigen (PCNA) [25], inhibitor of DNA binding 2 (Id2) [26], and fibroblast growth factor 1 (FGF1) [27][28][29]. Nevertheless, the regulatory targets of RFX1 are still not fully understood. RFX1 is expressed in various tissues. Moreover, RFX1 is prominent both in the somatic Sertoli cells of seminiferous tubules and also detected in early haploid germ cells [30], which suggested a possible role of RFX1 in male reproductive system.
Although the function of Spata4 is partially uncovered, the transcriptional regulatory mechanism of Spata4 remains to be clarified. In this study, we presented evidence that expression of Spata4 could be modulated by RFX1, which binds to the conserved 14-bp cis-elements of Spata4 promoter region. Overexpression of RFX1 could down-regulate Spata4 expression, whereas RNAi knockdown of RFX1 demonstrated the opposite effects. Moreover, RFX1 is involved in Spata4 mediated Sertoli cell growth. Our findings provide insights into Spata4 gene regulation and suggest the possible role of the RFX1, the negative transcriptional regulator of spata4, in Sertoli cell growth.

Results
Identification and characterization of the mouse Spata4 promoter region The mouse Spata4 gene locates on chromosome 8 B3.1 and consists of 6 exons ( Figure 1A). The mouse Spata4 promoter contains a CpG island approximately 400-bp-long ( Figure 1B) with GC % = 54% and ObsCpG/ExpCpG = 0.604, which comply with the accepted definition of CpG island [14]. To identify the promoter region of mouse Spata4, we first determined its transcription start site using the 59-rapid amplification of cDNA ends method (59-RACE). The Spata4 promoter contains dispersed transcription start sites located within a 90 bp region ( Figure 1C). Transcription of Spata4 initiates with a purine at +1 at a frequency of 98% and with a pyrimidine-purine dinucleotide at position 21, +1 at a frequency of 88%, which is consistent with previous studies that RNA polymerase II-mediated transcription initiation in mammals is preferentially initiated at a pyrimidine-purine dinucleotide at position 21, +1 [31].
Searching the mouse Spata4 promoter showed no TATA box, initiator (Inr) or other typical core promoter elements around the transcription start sites. To determine the region that is essential for Spata4 transcription, we serially truncated the 59-flanking region of the Spata4 promoter and analyzed its transcriptional activity through luciferase reporter assays. Deletion of the 59flanking region of the Spata4 promoter from 22819 to 2131 retained its transcriptional activity, while further deletion to 232 abolished its transcriptional activity. Therefore, the region between 2131 and +128 showed basal transcriptional activity ( Figure 1D). Besides, the Spata4 promoter constructs showed little reverse transcription activity, indicating that the Spata4 promoter is unidirectional transcribed ( Figure 1D).

Spata4 promoter contains a functional RFX1 binding site
Bioinformatics analysis through combination of Mulan and multi-TF scanning [32,33] showed that the Spata4 promoter contains conserved binging sites of several transcription factors, including c-jun (key component of AP-1 complex), RREB1, HSF1, SP1 and RFX1 (Figure 2A). To address which factor could regulate Spata4 promoter activity, we serially deleted their binding sites from the 21425/+128 reporter construct. Deletion of the RFX1 binding site resulted in 4.2-fold increase of luciferase activity, while deletion of SP1 binding site decreased the luciferase activity by 0.67-fold. Deletion of AP1, RREB1 and HSF1 binding sites had little effect ( Figure 2B), compared to the luciferase activity of 21425/+128 construct, respectively. These results indicate that RFX1 is a negative modulator of mouse Spata4 promoter and SP1 binding sites which locate between 2131, 232 might be essential for the basal transcriptional activity of Spata4.
Deletion of RFX1 binding site revealed markedly up-regulation of Spata4 promoter activity, indicating that RFX1 may function as a significant transcription repressor of Spata4. The RFX1 binding sites in human, rat and mouse Spata4 promoters are highly conserved, and have high similarity with the consensus and other published RFX binding sequences ( Figure 3A). Chromatin immunoprecipitation assays using two different RFX1 antibodies showed that RFX1 binds to the mouse Spata4 promoter in cultured cells. In addition, the sequences containing the GAPDH promoter are not precipitated by RFX1 antibodies ( Figure 3B). These results indicated that the RFX1 binding site in the Spata4 promoter is functional.
Then, we constructed a mutant of mouse RFX1 protein which lacks the DNA-binding domain to further confirm the functional binding of RFX1 to the Spata4 promoter. Deletion of the DNAbinding domain abolished the regulation of Spata4 promoter activity by RFX1, demonstrating that the DNA-binding domain is essential for the binding of RFX1 to the Spata4 promoter ( Figure 3C).

RFX1 regulates endogenous Spata4 expression in TM4 Sertoli cells
Spata4 is localized in the cytoplasm of TM4 cells (Figure S1 A). The mRNA level of Spata4 is dramatically increased on postnatal days 15 in mouse testes (Figure S1 B). Furthermore, the expression of Spata4 is abundant in primary Sertoli cells harvested from mouse testis (Figure S1 C). To investigate whether RFX1 regulates endogenous Spata4 expression, we transiently overexpressed different amounts of RFX1 in TM4 cells and detected its effect on the promoter activity and the mRNA levels of Spata4. Overexpression of RFX1 was confirmed by western blotting and immuno-fluorescent assay ( Figure 4A). Not only the promoter activity but also the mRNA level of Spata4 was significantly decreased by RFX1 in a dose dependent manner (Figure 4 B and C).
Next, RFX1 expression was knocked down by RNA interfere in TM4 cells ( Figure 5 A and B). The mRNA level of Spata4 was significantly up-regulated by RFX1 siRNA interference ( Figure 5C). In addition, endogenous Spata4 gene expression could be markedly increased by RFX1 knockdown in a time-dependent manner ( Figure 5 D and E).

RFX1 may affect proliferation of Sertoli cells through regulating Spata4 expression
Cell proliferation is closely linked to cell cycling, and cell cycle arrest can lead to suppression of proliferation rate. Exogenous expression of Spata4 in MCF7 cells significantly accelerates cell growth by traversing the S-phase and entering the G2-phase [8]. Meanwhile, stable overexpression of Spata4 could also promote Sertoli cell cycling; thereby improve Sertoli cell proliferation ( Figure S2 A). Knockdown of RFX1 increased the expression of Spata4 protein in TM4 cells and the cell growth rate ( Figure 6 A and C, Figure S2 B), while overexpression of RFX1 could decrease the cell growth rate ( Figure 6B, Figure S2 B), suggesting that RFX1 affects the proliferation of Sertoli cells in vitro and that the inhibitory effect was consistent with its regulatory effect on Spata4. The regulation of Spata4 by RFX1 and their roles in Sertoli cell proliferation are illustrated in Figure 6D. Taken together, these results indicated that endogenous expression of Spata4 is important for Sertoli cell growth.

Discussion
Spata4, conserved in male reproductive system, is important in several biological processes, including cell growth, proliferation and differentiation. However, the transcriptional regulation of Spata4 was not thoroughly investigated. Here we identified the promoter region of Spata4 and studied the mechanism of its transcriptional regulation. The mouse Spata4 promoter contains a CpG island and possesses dispersed transcription start sites, whereas no core promoter elements, such as the TATA box, Initiator (Inr) and Downstream Promoter Element (DPE) were found around the transcription start sites of Spata4.
Bioinformatics prediction revealed that the Spata4 promoter contains potential binding sites of several transcription factors, including AP1, SP1, RFX1, RREB1 and HSF1. Luciferase reporter assays showed that the region between 2131 and +128 of the Spata4 promoter accounts for its basal transcriptional activity. SP1, which binds to the consensus sequence of CCCGCCC that exists in many CpG island with high GC content, may involve in the transcription initiation of the CpG island promoters [34]. Deletion of the SP1 consensus binding sites located between 2131, 232 markedly decreased the promoter activity, suggesting that SP1 may function as an important factor in recruiting transcription apparatus to Spata4 promoter and regulating the transcription activity of Spata4.
RFX1, a dual function transcription regulator, has both transcriptional activation and repression properties depending on the promoter context [18]. Mutation of the RFX1 binding site increased the transcriptional activity of the Spata4 promoter significantly. Furthermore, overexpression of RFX1 reduced both the Spata4 promoter activity and the mRNA levels of Spata4, while knockdown of RFX1 had the opposite effects. Consequently, RFX1 was a remarkable negative regulator of Spata4 expression.
The RFX proteins bind to DNA as dimers and RFX1-4 can form either heterodimers or homodimers [17,[35][36][37], resulting in the RFX complex possesses similar specificities of binding sites. RFX1 is ubiquitously detected in various tissues with little variability [17]. RFX2 was the most prominent RFX mRNA detected in mouse testis. RFX3 and RFX4 were prevalent in testis [17], yet the RFX4 splice variants in testis are defective due to lack of the transcriptional activation region or the DNA-binding domain [35,38]. RFX1, RFX2 and RFX3, share high similarity and have identical binding site specificities and the ability of crossdimerization. The interaction among RFX1, RFX2 and RFX3 and their roles in regulation of Spata4 promoter activity are therefore worthy of further investigation. Of particular interest, it remains to be uncovered whether other testis specific co-activators are required for RFX1 or other RFX proteins.
In conclusion, we identified the transcription factor RFX1 as a novel negative regulator of Spata4 gene promoter activity, and provided insight into the regulation of Spata4 gene. Given the specificity of gene expression pattern of Spata4 in testis, identification of such a transcription factor brings us closer to the understanding of Spata4 dependent processes including spermatogenesis and cryptorchidism development, and also helps to uncover more biological functions of Spata4.

Ethics statement
The Animal Protocol (AP) was approved by the Institutional Animal Care and Use Committee of Tsinghua University (IACUCTHU) (AP#: 11-WZ1) with animal care performed strictly according to established institutional guidelines. All surgery was performed under pentobarbital anesthesia, and all efforts were made to minimize suffering.

Plasmid constructs
The mouse genomic sequence of the 59-flanking region of the Spata4 gene including the transcription start site was obtained from GenBank (accession no. NC_000074). DNA extracted from the whole blood cells from C57/BL6 mice with TIANamp Blood DNA Kit (TIANGEN Biotech., Beijing, China) was used as the template. Different truncations of mouse Spata4 promoter were cloned into the Kpn I and Xho I restriction sites of the pGL4.17 vector (Promega, Madison, WI). The expression plasmid for wildtype RFX1 and was constructed by cloning the corresponding sequence into Kpn I and Xho I restriction sites of pCDNA3.1 myc-His A vector (Invitrogen, Carlsbad, CA). Mutation PCR reactions were performed with PrimSTAR HS polymerase (Takara Bio Inc., Shiga, Japan), 59-end of the purified PCR products was kinated and ligated using DNA Kination Kit (Takara) referring to the manufacturer's protocol. The primers used during the plasmid construction are shown in Table S1 and S2. Cell culture Mouse Sertoli cell TM4 were cultured in DMEM/F-12 (Thermo Scientific HyClone, Logan, UT), supplemented with 10% fetal bovine serum (Thermo Scientific HyClone), 100 U/ml penicillin and 100 mg/ml streptomycin, at 37uC in a humidified atmosphere containing 5% CO 2 .

Rapid amplification cDNA ends
Rapid Amplification of cDNA Ends (RACE) was performed using the SMART TM RECE cDNA Amplification kit (Clontech, Shiga, Japan) following manufacturers' protocols, to determine the transcription start sites of mouse Spata4. Ploy A+ RNA was purified from testis total RNA of 2-months-old C57 BL/6J mice by Oligotex TM mRNA Purification system (Qiagen Inc., Valencia, CA). 59-RACE-Ready cDNA was synthesized using 59-CDS (Clontech) primer and SMART II A oligo (Clontech). 59 RACE PCR was performed with 59-RACE UPM (Clontech) and Spata4 specific primer: 59-GCAG-GACAGAACGACTCAGGCGGG-39, followed by a nested PCR using 59-nested UPM (Clontech) and a nested Spata4-specific primer: 59-TCCTGCGGGAACTGCTGGTGGTGTAG-39. As a negative control, 59-RACE Ready cDNA was amplified with 59-RACE UPM only. The nested PCR products were then sub-cloned into pMD19-T vector (Takara) and a total of 60 positive clones were sequenced.

Bioinformatics analysis
Mulan (Multiple-sequence local alignment) analysis was carried out comparing the genomic sequences of about 2000 bp upstream Spata4 first exon of human, mouse and rat, combined multiTF (http://multitf.dcode.org/) scanning for conserved transcription factor binding sites (TFBS).  washed gently with phosphate-buffered saline (PBS) buffer, and cultured in fresh serum-supplemented media for an additional 18 h. Cell lysates were collected and used for measurement of the relative promoter activity of each fragment with the dual luciferase reporter assay system (Promega). pGL4.17 [luc2] vector contains no promoter region was used as a control.  III) against RFX1 were tested, and RFX1-siRNA-III were used for further RNAi knockdown experiments.

Chromatin Immunoprecipitation (ChIP) Assay
ChIP was performed using the Magnetic Chromatin Immunoprecipitation Kit (Active Motif, Carlsbad, CA) as described in the manufacturer's protocol. Briefly, TM4 cells were washed once with phosphate buffered saline and then fixed in culture medium containing 1% formaldehyde at room temperature for 10 min. Cells were then collected and lysed to release the nucleus. The nucleus was digested with the enzyme mix at 37uC for 10 min to shear the chromatin into small segments. 2 mg of RFX1 (I- 19) antibodies (Santa Cruz Biotechnology, Santa Cruz, CA), 2 mg of RFX1 (D-19) antibodies (Santa Cruz), or a negative control IgG (Abcam, Cambridge, UK) were used for immunoprecipitation of the sheared chromatin at 4uC for 4 h. The pulled-down chromatin was washed, reverse cross-linked and purified. PCR primers for mouse Spata4 promoter were forward: 59-CACAGAGCCTTCTTG-GAGG-39 and reverse: 59-GGGAACTGCTGGTGGTGTA-39. PCR primers for mouse GAPDH promoter were forward: 59-GCCCTTGAGCTAGGACTGGAT-39 and reverse: 59-CCTGG-CACTGCACAAGAAGAT-39.

Quantitative real-time polymerase chain reaction (RT-qPCR)
Total RNA was extracted from TM4 cells using Trizol reagent (Invitrogen) according to the manufacturer's protocol. cDNA was transcribed with RevaTra Ace reverse transcriptase (Toyobo, Osaka, Japan) using oligo (dT) 20 as primers. Each cDNA transcribed from 0.8 mg RNA was subsequently used for RT-qPCR analysis with SYBR Green Real-time PCR Mater Mix (Toyobo) in CFX96 Real-Time PCR Detection Systems (Bio-Rad, Hercules, CA). PCR primers for mouse Spata4 were forward: 59-CAGATACAAGTCAA-GAGGTTC-39 and reverse: 59-GTGTTCTCACAAGGATTTC-CAC -39. PCR primers for mouse RFX1 were forward: 59-TGCTGCCCCTGCACCCTCACA-39 and reverse: 59-GGTG-GGAACACCGGTCTGGCTT-39. PCR primers for mouse GAPDH were forward: 59-CATGGCCTTCCGTGTTCCTA-39 and reverse: 59-CCTGCTTCACCACCTTCTTGAT-39. All the primers are intron-spanning. The specificity of primers was confirmed by the melting curves which performed on the amplified products. The thermal cycling conditions were as follows: initial denaturation at 95uC for 10 min, followed by 40 cycles of denaturing at 95uC for 15 s, annealing at 60uC for 30 s and extension at 72uC for 30 s.

Western blotting
Cells were collected and lysed with the cell lysis buffer (Beyotime, Shanghai, China). Alternatively, nuclear protein was isolated with the Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime). Protein concentrations were determined using the BCA Protein Assay Kit (Thermo Scientific). Protein samples were separated by electrophoresis through SDS-PAGE gels and transferred to polyvinylidene difluoride membranes (Millipore, Schwalbach, Germany). The membranes were blocked at room temperature for 1.5 h in Tris-buffered saline containing 0.05% Tween 20 with 5% non-fat dry milk. The membranes were then incubated with primary antibodies specific for RFX1 (1:500; Santa Cruz), Spata4 (1:1000; prepared by our laboratory), GAPDH (1:1000; Cwbiotech, Beijing, China) or b-tublin (1:1000; Cwbiotech) at 4uC overnight, followed by an incubation with horseradish peroxidase-conjugated anti-goat IgG (1:5000; Santa Cruz) or antirabbit IgG (1:10000) at room temperature for 1 h. The chemiluminescence reaction was performed using ECL reagent (Thermo Scientific).

Immunofluorescence staining
TM4 cells were seeded on a 6-well dish with a density of 1610 5 cells/ml and fixed with methylate-acetone (1:1) for 30 min. After blocking with 10% bovine serum albumin (BSA) for 1 hour, a primary antibody against RFX1 (1:100; Santa Cruz) was added for 2 h, followed by an anti-goat IgG conjugated with FITC (1:100; Santa Cruz) for 1 h. DAPI (0.1 mg/ml) was added to label the nucleus for 10 min. A negative control without the primary antibody was used to monitor the staining background. Images were obtained through Leica DX100 (Leica Microsystems Wetzlar GmbH, Germany).

CCK-8 assay
TM4 cells were suspended at a final concentration of 2610 4 cells/ml and seeded in 96-well plates, then transiently transfected with pcDNA3.1-vector, pcDNA3.1-RFX1, non-specific siRNA, siRNA-RFX1, respectively. After incubation for the indicated time as 0, 24 and 48 h, cell growth was determined by the CCK-8 assay (Beyotime). In brief, 10 ml CCK-8 solutions were added to each well and cultures were incubated at 37uC for 2 h. The absorbance at 450 nm was measured with a Benchmark microplate reader (Bio-Rad).

Statistical analysis
The values reported in graphs are the means 6 S.E.M. from three independent experiments. One-way ANOVA followed by Bonferroni multiple comparison was performed to assess the differences. A value of P,0.05 was considered statistically significant.