Identification of a novel FRMD7 splice variant and functional analysis of two FRMD7 transcripts during human NT2 cell differentiation

Cell culture and ATRA/BMP-2-induced differentiation

Human NT2 cells (Cell Culture Center of Peking Union Medical College, Peking, China) were maintained in DMEM/F12 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. Prior to the induction of differentiation, the NT2 cells were incubated in 25 cm² cell culture flasks in 4 ml of medium in a 37 °C, high humidity, 5% CO₂ incubator. The cells were trypsinized and sub-cultured at a ratio of 1:3 twice weekly. For the time course analyses, the NT2 cells were seeded on a 3.0×10⁶/10 cm² plate and treated the following day with 10 μM all-trans retinoic acid (ATRA) dissolved in DMSO [16,17] or with 50 ng/ml bone morphogenetic protein-2 (BMP-2) dissolved in normal saline [18,19]. The cells were collected at 12, 24, 36, 48, and 72 h after treatment for the early time points. For the later time points (5, 8, and 12 days), the cells were reseeded every three days in the continued presence of ATRA or BMP-2 and collected on the corresponding days.

Collection of human and mouse fetal tissue

Freshly isolated 16-week post-conception (wpc) aborted human fetal tissues were collected from the Women's Hospital School Of Medicine Zhejiang University. The experimental procedures were approved by the Human Ethical Committee of the Zhejiang University. Mouse fetal brain tissue was isolated from 18-day mouse fetuses. A scissors was used to produce 1–3 mm³ pieces of tissue, which were snap frozen in cryogenic vials in liquid nitrogen and stored at −80 °C. Total RNA from each sample was prepared for subsequent RT–PCR experiments.

RNA isolation, reverse transcription PCR, and quantitative real-time PCR analysis

Total RNA was isolated from human and mouse fetal tissues and NT2 cells using TRIZOL reagent (Invitrogen, San Diego, CA), according to the manufacturer’s instructions. Real-time quantitative PCR (RT-qPCR) assays were performed using a Light Cycler^TM real-time PCR thermocycler (Roche, Shanghai, China). Briefly, 1.5 μg of total RNA was reverse transcribed using oligo (dT) 12–18 with M-MLV Reverse Transcriptase (Promega, Madison, WI) according to the manufacturer's instructions. Relative levels of gene expression were analyzed using the SYBR^® Premix Ex Taq™ (TaKaRa, Dalian, China) using the recommended PCR conditions. The sequences of the RT-qPCR primers are shown in Table 1. The relative mRNA levels of target genes were calculated relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) using the (=2^-△△C_T) method [24].

cDNA clone and plasmid construction

Human FRMD7-FL and FRMD7-S genes were isolated from NT2 cell cDNA (synthesized from the total RNA of natural NT2 cells) by PCR amplification using the following primers: 5′-ATG CTA CAT TTA AAA GTG CAG TTT-3′ and 5′-TTA AGC TAA AAA GTA ATT ACA TGG T-3′. The mouse FRMD7-FL and FRMD7-S genes were cloned from 18-day fetal mouse brain cDNA using the following primers: 5′-GTA TGG ATC CAT GCT CCA TTT AAA AGT G-3′ and 5′-AGC TAA GAA ATA ATT GCA TGG CTT TAG C-3′. PCR was performed in 25 μl of a reaction mixture containing DNA polymerase from Thermococcus kodakaraensis (KOD) DNA polymerase buffer and 1.25 U KOD DNA polymerase (Toyobo, Osaka, Japan). Subsequent cloning of PCR products into the pEASY-Blunt vector (TransGen Biotech, Beijing, China) was performed using the pEASY-Blunt Cloning Kit for Sequencing (TransGen Biotech). Sequence analyses were performed by the Invitrogen Company (Shanghai, China). To generate eukaryotic expression vectors for hFRMD7 splice variants, cloned RT–PCR fragments in pEASY-Blunt were used as the template. The hFRMD7-FL cDNA was sub-cloned into pcDNA3.1(+) in frame with an NH₂-terminal HA tag (hFRMD7-FL-HA) and pEGFP-N1 (pEGFP-hFRMD7-FL) with the pAf/pAr and pCf/pCr primer sets, respectively (Table 1). The hFRMD7-S gene was subcloned into pcDNA3.1(+) with an NH₂-terminal Myc tag (hFRMD7-S-Myc) and plasmid pDsRed-N1 (pDsRed-hFRMD7-S) using the pBf/pBr and pDf/pDr primer sets, respectively (Table 1). NheI and XhoI were chosen as the cloning sites for all three vectors, and the sequences of all of the expression cassettes were verified.

Cell transfection and immunofluorescence microscopy

For transient transfection experiments, the NT2 cells were placed on glass coverslips or cell culture flasks until they reached approximately 60% confluency and were transfected with FuGENE HD transfection reagent (Roche) at a ratio of 1 μg of DNA to 3 μl of FuGENE HD per coverslip. The cells were processed for immunofluorescence at 48 h post-transfection. For stable overexpression experiments, G418 (1.2 μg/ ml medium; Invitrogen) antibiotic selection was performed for 4 weeks. For the immunofluorescence experiments, the induced or transfected NT2 cells were washed with phosphate-buffered saline (PBS) and fixed for 15 min at room temperature in 4% paraformaldehyde (PFA) in PBS. The coverslips were washed twice with ice-cold PBS, and the cells were permeabilized with PBS containing 0.2% Triton X-100 for 10 min at 4 °C. After washing the cells in PBS three times for 5 min (each wash), a blocking solution containing PBS and 10% normal goat serum was applied for 1 h at room temperature. Afterwards, primary antibodies (monoclonal anti-Myc or anti-HA or antibodies; Abmart Inc., Shanghai, China) diluted in a blocking solution were applied overnight at 4 °C or for 1 h at room temperature. In the double-labeling experiments, the cultures were incubated simultaneously with two primary antibodies. After three washes in PBS, the secondary antibodies (Dylight 488, Dylight 594; Jackson ImmunoResearch Laboratories, Inc. West Grove, PA) were diluted 1:250 in a blocking solution and added for 1 h at room temperature in the dark. The secondary antibody solution was decanted and washed three times with PBS for 5 min each in the dark. Finally, the cells were incubated for 5 min with 0.1 μg/ml DAPI as a nuclear counterstain. After three washes, the coverslips were mounted with a drop of Antifade Mounting Medium (Beyotime Institute of Biotechnology, Suzhou, China) and stored in the dark at −20 °C or 4 °C. The cells were viewed using a Nikon eclipse inverted fluorescence microscope equipped with a Nikon digital camera (Nikon Corporation, Tokyo, Japan). Image overlays and contrast enhancement were performed using Adobe Photoshop CS2 (Adobe Systems Inc. San Jose, CA) software.

Co-immunoprecipitation and immunoblot analyses

For co-immunoprecipitation (co-IP) experiments, NT2 cells were transiently co-transfected with 2 μg of each of the hFRMD7-FL-HA and hFRMD7-S-Myc vectors. At 48 h post-transfection, the total protein been extracted from the cells by using M-PER^® Mammalian Protein Extraction Reagent (Thermo Fisher Scientific Inc., Rockford, IL) following the manufacturer's instructions. Then the ProFound™ Co-Immunoprecipitation Kit (Thermo Fisher Scientific Inc.) was used for co-IP assays according to the manufacturer's instructions. Control immunoprecipitations were performed in the presence of non-specific immunoglobulins.

Statistical analyses

All experiments were done in triplicate, and the data are represented as mean±SEM. Statistical significance between groups was determined using the Kolmogorov–Smirnov test and Student’s t-test, as implemented by SPSS13.0 (SPSS Inc., Chicago, IL). Results were considered to be statistically significant when p<0.05.

Identification of a novel FRMD7 splice variant and tissue distribution of two FRMD7 transcripts

In RT–PCR experiments to clone the hFRMD7-FL gene from NT2 cells, two PCR products were produced. Cloning and sequence analysis of the PCR products indicated that they corresponded to two distinct FRMD7 mRNAs: the expected original hFRMD7-FL mRNA and another unexpected mRNA not described previously, which is referred to herein as hFRMD7-S. This novel transcript, which lacks a 45 bp segment in the 5′ region of exon 4, results in the fusion of the 3′ end of exon 3 to the 5′ end of truncated exon 4; therefore, hFRMD7-S encodes a putative protein of 699 amino acids (Figure 1A,B). We then performed RT–PCR to confirm the presence of mFRMD7-S in 18-day-old fetal mice (Figure 1C). The expression of hFRMD7-FL and hFRMD7-S transcripts was examined in tissues from 16-wpc human fetal tissue using sequence-specific primers (Figure 1D). We detected high expression of hFRMD7-FL in the cerebellum and moderate expression in the cerebral cortex, optic nerve, kidney, testis, and gastric wall. We detected high expression of hFRMD7-S in the cerebellum and moderate expression in the kidney, testis, and gastric wall. However, hFRMD7-S was not expressed in the optic nerve or cerebral cortex. Neither hFRMD7-FL nor hFRMD7-S were expressed in the medulla oblongata, pons, or liver. Taken together, these results revealed that hFRMD7-S exhibited a similar but limited tissue distribution compared to hFRMD7-FL (Figure 1E).

Levels of both hFRMD7 transcripts significantly increased in ATRA-induced differentiating NT2 cells

As shown in Figure 2A,B, the levels of hFRMD7-FL and hFRMD7-S mRNA expression in NT2 cells started to increase at 12 h after ATRA treatment and reached a maximal increase at 8 days (81 fold for hFRMD7-FL and 96 fold for hFRMD7-S relative to the 0 h time point; Figure 2A,B). The control group did not exhibit any change relative to the 0 h time point. Compared to ATRA, BMP-2 had a repressive effect on hFRMD7-FL and hFRMD7-S expression. After 24 h of BMP-2 treatment, the levels of hFRMD7-FL and hFRMD7-S transcripts started to decline relative to the 0 h time point, and continued to decrease after 36 h; after 8 days, the expression levels of hFRMD7-FL and hFRMD7-S transcripts began to increase (Figure 2C,D).

hFRMD7-FL and hFRMD7-S co-localized with class III β-tubulin and exhibited differential effects on neurite development

Stable NT2 cell lines overexpressing hFRMD7-FL-HA or hFRMD7-S-Myc were established for use in experiments to study their role during differentiation. We examined the co-localization of class III β-tubulin, which is not expressed at detectable levels in neural progenitor cells, with the two FRMD7 isoforms by indirect immunofluorescence. Both isoforms exhibited strong co-localization with neuronal class III β-tubulin in differentiating NT2 cells induced with ATRA for 5 days (Figure 3A,B). Having established that the hFRMD7-FL or hFRMD7-S overexpressed cell lines, we next sought to find how this change would affect neurite development in differentiated NT2 cells. Notably, following ATRA treatment for 5 days, a significant increase in the length and number of neurites was observed in NT2 cells stably overexpressing hFRMD7-FL-HA NT2 (Figure 3A,C and Table 2; p<0.05). However, the degree of neurite branching did not differ between the control and hFRMD7-S overexpressing NT2 cells (Figure 3B,C and Table 2; p>0.05).

hFRMD7-FL co-localizes and interacts with hFRMD7-S in NT2 cells

To investigate the subcellular co-localization of FRMD7 isoforms and to differentiate between the localization of endogenous FRMD7 and exogenously expressed FRMD7 proteins, we used expression vectors to express hFRMD7-FL or hFRMD7-S differentially tagged with green (EGFP) or red (DsRed) fluorescent protein, respectively. Following transient co-transfection with the fusion constructs encoding pEGFP-hFRMD7-FL and pDsRed-hFRMD7-S, untreated NT2 cells (without ATRA) and NT2 cells treated with ATRA for 7 days were subjected to fluorescence microscopy. In these experiments, EGFP-hFRMD7-FL and DsRed-hFRMD7-S exhibited a high degree of co-localization (Figure 4); both isoforms were primarily localized in cytoplasm and perinuclear regions in undifferentiated NT2 cells (Figure 4A). Similarly, in NT2 cells induced with ATRA for 7 days, EGFP-hFRMD7-FL and DsRed-hFRMD7-S co-localized extensively in the cytoplasm, perinuclear regions, and neurite processes. Notably, the distribution of EGFP-hFRMD7-FL was more abundant in the distal terminals of the neurites relative to that of DsRed-hFRMD7-S (Figure 4B).

To test for protein–protein interactions, we performed co-IP experiments. We co-expressed HA and Myc NH₂-terminally tagged versions of hFRMD7-FL (hFRMD7-FL-HA) and hFRMD7-S (hFRMD7-S-Myc), respectively, in NT2 cells. At 48 h post-transfection, total cell lysates were extracted, and hFRMD7-FL-HA was immunoprecipitated using anti-HA antibodies, followed by immunoblot analysis with anti-Myc antibodies. The results showed that hFRMD7-FL-HA co-immunoprecipitated with hFRMD7-S-Myc (Figure 4C). This finding was confirmed by reciprocal immunoprecipitation experiments, which showed that hFRMD7-S-Myc co-immunoprecipitated with hFRMD7-FL-HA (Figure 4D). The results of these experiments may suggest that hFRMD7-FL and hFRMD7-S either interact directly or are associated with the same protein complex in NT2 cells.

hFRMD7-FL protein stimulates transcription of hFRMD7-S

Alternative splicing often impacts the transcription and subsequent translation of the major protein. Therefore, we performed experiments to determine whether overexpression of one FRMD7 splice variant was associated with altered transcription of the other splice variant. Overexpression of hFRMD7-FL or hFRMD7-S was confirmed by RT–PCR and immunoblot analyses (data not shown). We found that the transcription levels of hFRMD7-S exhibited a time-dependent increase in cells overexpressing hFRMD7-FL relative to control cells, with an average of 4.8 fold higher hFRMD7-S expression levels at 72 h (Figure 5A); in contrast, hFRMD7-FL transcription levels did not change significantly in NT2 cells overexpressing hFRMD7-S (Figure 5B).