SOHLH2 is essential for synaptonemal complex formation during spermatogenesis in early postnatal mouse testes

Spermatogenesis- and oogenesis-specific helix-loop-helix transcription factor 2 (SOHLH2) is exclusively expressed in germ cells of the gonads. Previous studies show that SOHLH2 is critical for spermatogenesis in mouse. However, the regulatory mechanism of SOHLH2 during early spermatogenesis is poorly understood. In the present study, we analyzed the gene expression profile of the Sohlh2-deficient testis and examined the role of SOHLH2 during spermatogenesis. We found 513 genes increased in abundance, while 492 genes decreased in abundance in 14-day-old Sohlh2-deficient mouse testes compared to wildtype mice. Gene ontology analysis revealed that Sohlh2 disruption effects the relative abundance of various meiotic genes during early spermatogenesis, including Spo11, Dmc1, Msh4, Prdm9, Sycp1, Sycp2, Sycp3, Hormad1, and Hormad2. Western blot analysis and immunostaining showed that SYCP3, a component of synaptonemal complex, was significantly less abundant in Sohlh2-deficient spermatocytes. We observed a lack of synaptonemal complex formation during meiosis in Sohlh2-deficient spermatocytes. Furthermore, we found that SOHLH2 interacted with two E-boxes on the mouse Sycp1 promoter and Sycp1 promoter activity increased with ectopically expressed SOHLH2. Taken together, our data suggest that SOHLH2 is critical for the formation of synaptonemal complexes via its regulation of Sycp1 expression during mouse spermatogonial differentiation.

Scientific RepoRts | 6:20980 | DOI: 10.1038/srep20980 significantly decreased in the Sohlh2-deficient testes. In addition, we showed that the regulation of meiotic factors by SOHLH2 is crucial for synaptonemal complex formation during spermatogenesis.

Results
Defect of early spermatogenesis in Sohlh2-deficient testes. To investigate SOHLH2 effect on early spermatogenesis, we used 2-week-old WT and Sohlh2 KO mice. The deficiency of Sohlh2 expression in KO mice was confirmed by RT-PCR analysis (Fig. 1A). Histological analysis showed an abnormal spermatogenesis due to a reduced cell number in the seminiferous tubules of KO testes (Fig. 1B). Cell types in the seminiferous tubules were examined by immunohistochemistry using the cell type-specific markers; SOHLH1 for spermatogonia, SYCP3 for spermatocytes, and GATA4 for sertoli cells. The numbers of SOHLH1-positive cells were increased in KO testes and the numbers of GATA4-positive cells in KO testes were similar to that in WT testes (Fig. 1B,C). By contrast, SYCP3-positive spermatocytes were significantly reduced in KO testes compared with WT testes (Fig. 1B,C). These findings suggest that abnormal spermatogenesis in Sohlh2-deficient testes during early postnatal development is caused by disruption of spermatocytes but not by abnormality of spermatogonia or sertoli cells. To further investigate that the potential disruption of spermatocytes is related with cell death, we performed cell death analysis using a terminal deoxynucleotidyl transferase mediated dUTP-biotin nick end labeling (TUNEL) assay. TUNEL-positive cells were significantly increased in KO testes (Fig. 1D,E). These data suggest that differentiation of spermatogonia to early spermatocytes can be occurred in Sohlh2-knockout testes during early postnatal development. Therefore, Sohlh2-deficiency may lead to the apoptosis of early prophase I spermatocytes, resulting in abnormal spermatogenesis.
Effect of Sohlh2-deficiency on gene expression in the testes. To characterize which genes are affected by Sohlh2-deficiency, we performed microarray analysis using Sohlh2 KO and WT testes when spermatogonia entered meiosis in 2-week-old mice. We discovered at least a 2-fold increase or decrease in the expression of 1005 genes in the KO testes ( Fig. 2A). Among these genes, 513 were increased (Supplementary Table S1) and 492 were decreased in their mRNA expression (Supplementary Table S2). Based on gene ontology analysis (Fig. 2B), we found that many of these genes are involved in cell differentiation (e.g., Runx3, Kit and Piwil1), signal transduction (e.g., Dll3, Jag2, and Prok2), transport (e.g., slc9a6 and Cebpe), transcription (e.g., Tbx1, Ybx2, and Lin28b), apoptosis (e.g., Aifm3, Scp2, and Scp3), proliferation (e.g., Kit and Ppar-gamma), protein biosynthesis (e.g., Nmt2 and Paip1), homeostasis, and inflammation. We also observed that Sohlh2-deficiency caused abnormal expression of genes associated with spermatogenesis, spermiogensis, function of sperm, and morphology of seminiferous tubules and spermatids ( Fig. 2C and Supplementary Table S3). It was most likely that the defects in gene expression of meiosis factors were related to loss of germ cells in Sohlh2-deficient mice (Fig. 2D).

Expression of meiotic factors in Sohlh2-deficient testes.
To confirm Sohlh2 deficiency-induced changes in the relative abundance of meiotic genes, we carried out quantitative real-time reverse transcription polymerase chain reaction (RT-PCR) using total RNA from Sohlh2 KO and WT testes of 2-week-old mice. Genes specific to prophase I (i.e., leptotene, zygotene, pachytene and diplotene stages) were selected according to previous studies 7 , and their primers were designed. Of the leptotene-related factors, Spo11 expression was significantly reduced in KO testes (Fig. 3A), whereas Mei1 expression was unchanged (Fig. 3B). Of the zygotene-and pachytene-related factors, the expression of Cdk2, Sycp1, Sycp2, Sycp3, Hormad1, and Hormad2 was significantly decreased in KO testes (Fig. 3C,H). Among these genes, Sycp1, Sycp2, Sycp3, Hormad1, and Hormad2 are critical for the formation of synaptonemal complexes during meiosis. Other zygotene-and pachytene-related factors, including Psmc3ip, Msh4, Msh5, Prdm9, Syce1, Syce2, and Piwil2, were also significantly reduced in the KO testes (Fig. 3I,O). However, Rec8 expression was significantly increased (Fig. 3P) and Dmc1 expression was unchanged (Fig. 3Q) in KO testes. The pachytene-related factors Cpeb1, H2afx, and Tex11 were significantly diminished in KO testes (Fig. 3R,T), whereas Smc1b expression was unchanged (Fig. 3U). Expression of diplotene-related factors Ccna1 (Fig. 3V) and Hspa2 (Fig. 3W) was significantly decreased in KO testes, whereas Ccnb1ip1 expression was unaffected (Fig. 3X). The gene expression profile and early spermatogenic arrest in Sohlh2-deficient mice raise questions on if SOHLH2 is key regulator of meiotic gene expression, which might be crucial for successful meiotic division of germ cells during spermatogenesis.
Relationship between SOHLH2 and synaptonemal complex protein SYCP1. To examine the protein levels of major meiotic proteins including SYCP1, SYCP3, and RAD51 in the Sohlh2 KO testis, we performed western blot and immunofluorescence analyses. The expression of SYCP1, which is a transverse element in the synaptonemal complex, was significantly reduced in KO testes (Fig. 4A). Also, the expression of SYCP3, which is a lateral element in the synaptonemal complex, was reduced in KO testes (Fig. 4A). Consistent with these results, immunofluorescence staining showed a decrease in SYCP1 signal intensity in spermatocytes from KO testes compared with those from WT testes. The signal intensity of SYCP3 was not affected by Sohlh2 deficiency (Fig. 4B). Instead, the number of SYCP3-positive cells was significantly decreased, affecting SYCP3 expression level in western blot (Fig. 4A). γ H2AX is a phosphorylated form of H2AX, which is associated with meiotic DNA double-strand break. DNA double-strand break normally occurs in the nuclei of primary spermatocytes, but not in sertoli cells, spermatogonia, and spermatid. γ H2AX is known to be highly detectable in premeiotic S-phase and leptotene stage of spermatocytes 8 . Shown in Fig. 4B, there were no defects in γ H2AX staining in Sohlh2 KO spermatocytes. This suggests that the reduction of synaptonemal complex component, SYCP1, might affect male meiosis through synaptonemal complex formation.
Previous studies indicate that Sycp1 and Sycp3 deficiency affects chromosome synapsis during meiosis I 9,10 . Therefore, we examined whether SOHLH2 that was critical for Sycp1 expression affects formation of synaptonemal complexes in Sohlh2-deficient testes by using SYCP3 staining to examine chromosome structure. In the testes of 2-week-old WT mice, we found normal progression of spermatocytes across leptotene, zygotene, pachytene, and diplotene stages (Fig. 5A), whereas KO spermatocytes were found only in the leptotene stage (Fig. 5A,B). To verify the defect of Sohlh2-deficient spermatocytes in synaptonemal complex formation, we investigated the structure of synaptonemal complexes using electron microscopy. Tripartite synaptonemal complex structures  were well visualized during meiosis I in WT testes, but not in KO testes (Fig. 5C). These findings indicate that SOHLH2 controls synaptonemal complex formation via regulating Sycp1 expression.
Next, we examined the localization of HORMAD1, which localizes to unsynapsed chromosome axes during prophase I 11,12 . HORMAD1 was gradually detected in unsynapsed chromosome during prophase I in WT spermatocytes (Fig. 6A). HORMAD1 was detected on the sex body complex in pachytene (Fig. 6A), consistent with previous reports 11,12 . HORMAD1 signal was detected as dispersed spots in KO spermatocytes (Fig. 6B). Next, we investigated the localization of SYCP1, which is indicative of synapsed chromosomes. In WT testes, SYCP1 localized to synapsed chromosomes only after zygotene stage, at which the synaptonemal complex begin to form (Fig. 7A). At the pachytene stage, SYCP1 signal was detected on synapsed autosomal chromosomes except for the sex chromosome, which showed partial detection due to unsynapsed region on the chromosome (Fig. 7A). However, KO spermatocytes showed dispersed spots of SYCP1 signal (Fig. 7B), similar to the HORMAD1 signal (Fig. 6B). These results suggest that SOHLH2 is very important for the formation of synaptonemal complexes through either direct or indirect regulation of meiotic factors including SYCP1.

Regulation of Sycp1 expression by SOHLH2.
To investigate whether Sycp1 expression is directly regulated by SOHLH2, we examined Sycp1 promoter (up to ~572 kb) using a computational program (TFSEARCH). We found three putative bHLH domain-binding elements, E-boxe elements (CANNTG), at − 258, − 94 and − 45 upstream from start codon on the mouse Sycp1 promoter (Fig. 8A). To examine the effect of SOHLH2 on transactivation of the Sycp1 promoter, HEK293T cells were transiently transfected with the firefly luciferase reporter vector fused with its promoter containing the three putative E-boxes. Compared with a mock vector, expression of FLAG-tagged SOHLH2 vector resulted in increasing amounts of Sycp1 promoter activity (Fig. 8B,C). Electrophoretic mobility shift assay (EMSA) showed that SOHLH2 directly bound to E-box sequences containing CACGTG or CAGCTG (Fig. 8D,F). These results suggest that Sycp1 expression might be directly regulated via two of three putative E-boxes under SOHLH2 regulation during spermatogenesis in mouse.

Discussion
The process of spermatogenesis in the seminiferous tubules is a complex biological event. Spermatogonial stem cells perpetually proliferate and differentiate into sperm cells through meiosis, and this process is highly regulated by germ-cell-specific factors and hormones 13 . In a previous study, we discovered that SOHLH2 is a germ cell transcription factor expressed in the gonad 1 . SOHLH2 plays a role in the differentiation of mature spermatogonia during spermatogenesis 3,4 as well as in oogenesis 2 . Sohlh2 deficiency disrupts normal spermatogenesis by altering the expression of many genes in the testis. However, the mechanism by which SOHLH2 regulates gene expression was unclear. Recently, two research groups showed that SOHLH2 directly regulates several gene targets including Kit, Gfra1, and Sox3 5,6 . In particular, SOHLH2 binds independently to the E-box on the Kit promoter and stimulates Kit expression with SOHLH1, another germ cell-specific transcriptional factor 14 . Both SOHLH2 and SOHLH1 play crucial roles in the differentiation of spermatogonial stem cells 3,4,14 . However, these findings are not sufficient to fully explain molecular events mediated by SOHLH2 that are essential for spermatocytes to develop normally from differentiating spermatogonia.
Stem and early progenitor spermatogonia give rise to differentiating spermatogonia that complete meiosis to produce haploid spermatogenic cells termed spermatids. SOHLH1 and SOHLH2 are co-expressed in undifferentiated and differentiating spermatogonia, but not in primitive undifferentiated spermatogonia expressing glial cell-derived neurotrophic factor receptor alpha I 6 . In previous study, accumulation of STRA8-positive differentiated spermatogonia was previously shown in Sohlh2-deficient testes 5,6 , suggesting that differentiation of spermatogonia does not occur properly. SOHLH2 binds directly to chromatin upstream of genes that are essential for spermatogonial stem cell maintenance and spermatogonial differentiation, such as Gfra1, Sox3, Sohlh1, Sohlh2, and Kit 6 . In our immunohistochemistry, we also found that SOHLH1-positive spermatogonia were more in 2-week-old Sohlh2 KO testes than in WT testes, suggesting that Sohlh2-deficiency caused an impairment in spermatogonial differentiation and resulted in accumulation of SOHLH1-positive spermatogonia. Although Sohlh2-deficiency compromised the differentiation of spermatogonia, SYCP3-positive spermatocytes were detectable in KO testes. We propose that early postnatally, Sohlh2-deficient spermatogonia are entering meiosis to form abnormally developed leptotene spermatocyte-like germ cells. These data are similar to the previous data showing Sohlh2 KO testes contain a portion of cells with thread-like chromosome condensation which is a characteristic of leptotene spermatocytes 6 , although premature HORMAD1-positive meiotic cells abnormally appears in 1-week-old Sohlh2 KO testes. Therefore, we focused on the time when meiosis begins, which is around postnatal day 10 to 14 in mice, to determine whether SOHLH2 regulates the meiotic process. When we examined the gene profiles of testes from 2-week-old male Sohlh2 KO mice, we found that numerous genes were regulated at this stage, including 1005 genes with at least 2-fold changes in expression relative to WT mice. Many of these genes were involved in meiosis, including Spo11, Sycp1, Sypc3, and Hormad1. This altered expression of meiotic genes can be derived from the direct effect of Sohlh2-deficiency on transcriptional rate of the genes. In addition, a loss of meiotic cells in Sohlh2 KO testes can result in the reduced mRNA levels due to the vast majority of leptotene to diplotene spermatocyte populations not being present in Sohlh2-deficient mice. According to our RT-PCR and immunohistochemistry analysis, it is likely that the decreased level of Sycp3mRNA originated from spermatocytes absent after leptonema, rather than Sohlh2 deficiency-induced down-regulation of gene transcription. In 1-week-old testes, Spo11 expression was not affected by Sohlh2-deficiency 6 , but its expression was drastically decreased in 2-week-old KO testes. Spo11 is normally expressed at very low level in 1-week-old testes and increased from postnatal day 12. Spo11 mRNA is detected at the highest level in pachytene spermatocytes 15 . Therefore, the decreased Spo11 expression might also be a consequence of SOHLH2-induced loss of pachytene cells in KO testes, although more extensive analysis is needed.
During meiosis, crossover between homologous chromosomes occurs during prophase I, which consists of substages: leptotene, zygotene, pachytene, and diplotene. Chromosomes are marked for double-strand break during leptotene stage, and they begin to synapse at the zygotene stage. The synapse is completed at the pachytene stage, and the synapsed chromosomes separate with one or two synapsed regions at the diplotene stage. The successful synapse of homologous chromosomes depends on the formation of the synaptonemal complex, for which SYCP1, SYCP2, SYPC3, and HORMAD1 are critical factors. SYCP1 acts as a transverse element in the synaptonemal complex 9 that bridges between homologous chromosomes, and SYCP2 and SYCP3 are lateral elements of the complex that are localized on the sister chromatids (Fig. 9). HORMAD1 is involved in the double-strand break and synaptonemal complex formation during meiosis and it localizes to unsynapsed chromosome axes during prophase I 11,12,16 . We found that SYCP1 expression was significantly decreased in Sohlh2 KO testes. The expression level of SYCP3 in KO spermatocytes was similar to that in WT cells, although the number of SYCP3-positive cells were reduced in KO testes. A chromosome spreading assay showed that Sohlh2 KO spermatocytes were arrested in leptotene stages, and electron microscopy showed that synaptonemal complexes failed to form in KO spermatocytes. In the study of Sycp1 function, spermatogenic differentiation is interrupted predominantly at pachytene stage of spermatocytes in Sycp1 KO testes, suggesting Sycp1-deficient spermatocyte can undergo meiosis from leptotene to pachytene 9 . However, Sohlh2 deficiency is more severe defect in meiosis progression than Sycp1 deficiency. These results indicate that SOHLH2 plays a role in regulation of SYCP1 expression and the lack of SYCP1 with more additional factors in Sohlh2 KO testes was involved in misregulated spermatogenesis.
SYCP1 is a major component of transverse elements in the synaptonemal complex. Spermatocytes in Sycp1 KO testes are arrested during early meiosis 9 , similar to what is observed in Sohlh2 KO mice. In spite of the importance of SYCP1 in the formation of the synaptonemal complex, little is known about how SYCP1 is regulated during the meiotic process. A previous study investigating the regulation of the Sycp1 promoter region (up to 2 kb) in gonad showed that a short fragment encompassing the transcription site (− 54 to + 102) was critical for temporal and spatial regulation during meiosis in male mice 17 . As the region contains an E-box (CAGCTG) at position − 45 relative to the transcription start site 17 , this suggests that Sycp1 expression is regulated by a transcription factor containing a bHLH by way of an E-box on its promoter. When the activity of longer promoter fragment (− 260 to + 102) was compared with the activity of the short fragment (− 54 to + 102), expression from the longer promoter fragment was much higher in testes 17 , suggesting the possibility that other promoter regions of Sycp1 were involved in regulation of its promoter activity. In the present study, we found two additional E-boxes on the Sycp1 promoter region (− 258 and − 94) (Fig. 8). SOHLH2 directly bound to two E-boxes at positions − 94 and − 45 on Sycp1 promoter and induced Sycp1 promoter activity. Although a mutation study for SOHLH2-induced Sycp1 promoter activity remains, we propose that SOHLH2 might regulate the expression of Sycp1 via binding to two of E-boxes (CACGTG and CAGCTG).
In conclusion, our gene expression profiling identified putative regulatory targets of SOHLH2 as a master transcription factor during spermatogenesis. Furthermore, we revealed that SOHLH2 is associated with progression of meiosis by regulating Sycp1 expression during spermatogenesis.

Animals.
All mouse experiments were carried out in a 129S7/SvEvBrd × C57BL/6 mixed background. Sohlh2 heterozygous mice were bred to generate Sohlh2 KO mice as previously described 2 . The care of mice and experimental procedures complied with the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Agricultural Animal Care and Use Committee of CHA University.
Histology and immunostaining. Testes were fixed in 4% paraformaldehyde (PFA) and embedded in paraffin wax for histological analysis. After embedding in paraffin, sections (5-μ m thick) were cut and mounted on Superfrost Plus slides (Fisher Scientific, USA). For histology, sections were stained with Harris hematoxylin (Sigma-Aldrich, USA) for 5 minutes (min) and washed in running tap water for 5 min.
For immunofluorescence assay, deparaffinized slides were placed in a container with antigen retrieval buffer (10mM sodium citrate, 0.05% Tween 20, pH6.0). The containers were heated in a microwave at full power for 5 min. After boiling, the slides were reheated at 50% power for 5 min. The slides were then cooled to room temperature for 15 min and washed three times for 5 min in distilled water. After permeabilization in a buffer solution (0.2% Triton X− 100 in phosphate-buffered saline (PBS)) at room temperature for 45 min, the slides were washed with PBS and incubated with blocking buffer (PBS with 10% bovine serum albumin (BSA)) in a humidified chamber for 1 hour at room temperature. The blocked slides were incubated with primary antiboides at 4 °C for overnight. The primary antibodies were anti-GATA4 (1:500; Abcam), anti-HORMAD1 (1:500) 17   SOHLH2 acts as a master regulator of target genes including Kit during spermatogonial differentiation. In the testes, SOHLH2 is hypothesized to regulate gene expressions critical for meiosis during spermatogenesis. SOHLH2 deficiency leads to spermatogenic arrest early during meiotic prophase I due to defects in components of the synaptonemal complex such as SYCP1, leading to the disruption of differentiation and the apoptosis of spermatocytes.
For immunohistochemical staining, deparaffinized slides were incubated in 0.3% H 2 O 2 in PBS and 100% methanol for 15 min to quench endogenous peroxidase activity after antigen retrieval. The slides were incubated with a blocking buffer (10% normal serum with 10% BSA) for 2 hours at room temperature. After tapping off the blocking agent, biotin/avidin blocking solutions (Vector Labs, USA) were added to block endogenous biotin and avidin for 15 min. Primary antibodies were applied as described above. On the next day, the sections were further processed with an avidin-biotin-horseradish peroxidase complex (Vector Labs, USA). The peroxidase signals were developed with a 0.01% 3,3′ -diaminobenzidine solution (Vector Labs) according to the manufacturer's protocol and counterstained with hematoxylin. The slides were rehydrated with ethanol and xylene, and mounted with Permount mounting medium (Fisher Scientific, USA). TUNEL assay. Analysis of TUNEL staining was performed using a TUNEL assay kit (Roche, USA) according to the manufacturer's instructions. TUNEL-positive cells in testis were counted under microscopy.
Analysis of gene expression profile. The testes were separated from 2-week-old WT and KO mice (n = 4 per group), and total RNA was extracted using an RNeasy mini kit (Qiagen, Germany). A 12-bay MAUI hybridization system was used for the Agilent mouse whole genome 8 × 60 K array platform (Agilent, USA). Image and signal data extraction was performed using Agilent Scanner, Feature Extract v10.7.3.1. Because four separate experiments were performed on WT and KO RNA, signal intensities for particular genes were averaged across experiments, and the ratio of WT-to-KO signal was calculated. Data were normalized using the robust multi-array average method 18 . Differentially expressed gene selection, clustering, and functional analysis were performed by GenoCheck (Korea).

Real-time RT-PCR.
Total RNAs were converted to cDNA using iScript cDNA Synthesis kit (Bio-Rad Laboratories, USA). Quantitative real-time PCR was performed using QuantiTect SYBR Green PCR reagents (Qiagen, Germany) and iCycler (Bio-Rad, USA). The primers used for the PCR were listed in supplementary  Table S4. Results were evaluated with iQ5 TM Optical system software. RNA expression level was normalized to Gapdh mRNA level and calculated using the relative quantification approach based on the Δ Δ Ct method . The relative amount of cDNA was determined as 2 −ΔΔCt .
Chromosome spreading assay. Testes were removed from 2-week-old mice and incubated in trypsin-EDTA solution with DNas1 at 37 °C for 10 min and washed in PBS. Trypsinized testes were pipetted repeatedly and filtered by mesh (70 μ m) to remove debris. After centrifugation, cells were resuspended in 1 ml PBS. Suspended cells were placed on poly-L-lysine-coated slides containing sucrose solution (40 mg/ml) and treated with permeabilization solution containing 0.005% Triton X-100. The slides were fixed in 2% PFA and 0.02% SDS for 1 hour at room temperature. The incubated slides were washed six times in distilled water, dried briefly, and stored at − 80 °C until use.

Western blot analysis.
Protein samples were loaded on SDS-PAGE (4-20% gradient gel; Invitrogen), transferred to nitrocellulose membranes and blocked with 5% non-fat milk in tris-buffered saline (TBS) containing 0.1% Tween 20 (TBS-T) for 1 hour at room temperature. The membrane was incubated with specific primary antibody diluted in TBS-T/5% milk at 4 °C overnight followed by appropriate HRP-conjugated secondary antibody (Invitrogen) diluted at 1:1000 in TBS-T/5% milk for 1 hour at room temperature. The membrane was developed using the ECL Western Blotting substrate kit (Gendepot, USA). The relative expression of protein was analyzed using the ChemiDoc XRS system (Bio-Rad, USA). Electron microscopy. Testes were fixed with 2% formaldehyde and 3% glutaraldehyde, for 2 hour at room temperature. Samples were treated with 0.5% uranyl acetate and osmium tetraoxide, dehydrated with ethanol, and embedded in LX− 112 medium. The tissue was polymerized in a 70 °C oven for 2 days and then cut into ultrathin sections (70-100 nm) with an ultracut microtome. The sections were stained in 1% aqueous uranyl acetate and placed for 2 min in 1% aqueous lead citrate at room temperature in an M Stainer, The sections were then examined using a transmission electron microscope.
Electrophoretic mobility shift assay. For in vitro translation of SOHLH2, DNA encoding FLAG-tagged SOHLH2 was subcloned into pcDNA3.1/Hygro vector (Invitrogen). FLAG-tagged SOHLH2 was in vitro translated using TNT ® quick coupled transcription/translation system (Promega) according to manufacturer's instructions. Successful translation of SOHLH2 was confirmed by western blot analysis. One microliter of protein products was electrophoresed through 10% SDS-polyacrylamide gel and transferred into nitrocellulose