Generation of transgenic fibroblasts expressing pancreas‑specific and doxycycline‑inducible ICER I γ for the establishment of a porcine model of human diabetes mellitus

Cell culture. The study was approved by the Ethics Committee of Chungbuk National University (Cheongju, Korea). The porcine fibroblasts were obtained from miniature pig fetuses (Yucatan pigs; Optifarm Solution Inc., Gyeonggi-do, Korea) on the thirtieth day of preg nancy and the cells were routinely maintained in Dulbecco's Modified Eagle's Medium (DMEM) containing 25 mM glucose, supplemented with 10% fetal bovine serum (FBS; WelGENE, Inc., Daejeon, South Korea), 100 U/ml penicillin and 100 µg/ml strep tomycin. The mouse β -cell line, MIN6 (American Type Culture Collection, Manassas, VA, USA), was cultured in DMEM containing 25 mM glucose, supplemented with 15% FBS, 55 µM 2-mercaptoethanol (Gibco, Grand island, NY, USA), 100 U/ml penicillin and 100 µg/ml strep tomycin. All cells were grown in a humidified 5% CO 2 atmosphere at 37˚C. Unless other-wise indicated, all cell culture materials were obtained from Invitrogen Life Technologies (Carlsbad, CA, USA). Genomic DNA extraction and PCR. Genomic DNA from the cells was isolated with a G-DEX™ IIc Genomic DNA Extraction kit (iNtRON Biotechnology, Inc., Seoul, South
Korea). Genomic DNA (0.1 µg) was amplified in a 20-µl PCR reaction containing 1 U i-Start Taq polymerase (iNtRON Biotechnology, Inc.), 2 mM dNTPs (Takara Bio, Inc., Shiga, Japan) and 10 pmol of each specific primer. The details of all primers are described in Table I. The PCR reactions were denatured at 94˚C for 30 sec, annealed at 62˚C for 30 sec and extended at 72˚C for 1-2 min. The PCR products were subjected to cloning processes and/or separated on a 1% agarose gel, stained with ethidium bromide and photographed under UV illumination. The image was scanned using GelDoc EQ (Bio-Rad, Hercules, CA, USA). Vector construction. All restriction enzymes were obtained from Takara Bio, Inc. The human insulin promoter region (from nucleotides (nt) -1,431 to +1 nt; +1 = the transcriptional start site) demonstrated the highest transcriptional activity in a previous study (14) and were prepared by long-range PCR using human genomic DNA (Clontech Laboratories, Mountain View, CA, USA) as the template, and specific primers containing restriction enzyme sites (SpeI at the 5' end or EcoRI at the 3' end). Amplified fragments were digested with SpeI and EcoRI and replaced with the CMV promoter of the transcriptional activator construct, pCMV-Tet3G, purchased from Clontech Laboratories. The ICER I γ cDNA was prepared by PCR using genomic DNA from pig pancreas as the template and was inserted into pTRE3G-ZsGreen1 which contains green fluo rescence protein, ZsGreen1 (Clontech Laboratories) through MluI at the 5' end or BamHI at the 3' end. Using the PCRmethod, the ICER I γ -expressing tTA-response region obtained from recombinant pTRE3G-ZsGreen1-pig ICER I γ construct was digested with Bst1107I or SpeI and combined with the recombinant pHINSP-Tet3G vector controlled by the human insulin promoter.

Establishment of transgenic cell lines. The fibroblasts were transfected with the linearized targeting vector using Lipofectamine ® 2000 (Invitrogen, Carlsbad, CA, USA).
Following 24 h of transfection, the medium was replaced with DMEM supplemented with 10% FBS and 250 µg/ml G-418 (Roche Diagnostics, Indianapolis, IN, USA) for four weeks. The antibiotic-resistant cells were further selected, subjected to PCR-based genotyping and stored until required for somatic cell nuclear transfer.

Transient transfection and dox treatment. Transient transfection was performed using Lipofectamine™ 2000 according to the manufacturer's instructions. Briefly, 3x10 5 cells were seeded in 6-well tissue culture plates one day prior to transfection. In total, 4 µg of the recombinant constructs per well was transfected into the cells under serum-free DMEM. Following incubation for 4 h, the medium was replaced with DMEM containing 10% FBS, 100 U/ml penicillin and 100 µg/ml strep tomycin. A total of 20 h later, various concentrations of dox were treated in transiently transfected cells for an addi-tional 24 h.

RNA preparation and semi-quantitative (q)PCR. Total RNA from MIN6 cells was extracted using TRIzol reagent (Invitrogen Life Technologies) according to the manufacturer's instructions. The concentration of the total RNA was determined by measuring the absorbance at 260 nm. First-strand cDNA was prepared by subjecting total RNA (1 µg) to reverse transcription using Moloney murine leukemia virus (MMLV) reverse transcriptase (Invitrogen Life Technologies) and random primers (9-mers; Takara Bio, Inc.). To determine the optimal conditions for logarithmic phase PCR amplification for target cDNA, aliquots of total cDNA (1 µg) were amplified using different numbers of cycles. The mouse β -actin gene was amplified as the internal control to rule out the possibility of RNA degradation and to control for variations in mRNA
concentrations. A linear correlation between PCR product band visibility and the number of amplification cycles was observed for the target mRNA. The mouse β -actin and target genes were quantified using 28 and 30 cycles, respectively. The PCR reactions were denatured at 94˚C for 30 sec, annealed at 58˚C for 30 sec and extended at 72˚C for 30 sec. The PCR products were on a 2.3% agarose gel, stained with ethidium bromide and photographed under UV illumination. The image was scanned using GelDoc EQ (Bio-Rad).

Western blot analysis. Lysates were prepared in RIPA buffer (50 mM Tris, pH 7.4, 150 mM sodium chloride (NaCl), 1% Triton X-100, 0.5% sodium deoxycholate, 1 mM EDTA and 1 mM PMSF) and protease inhibitor cocktail (Roche Diagnostics). Protein content was determined using the Pierce BCA Micro Protein Assay kit (Thermo Fisher Scientific, Waltham, MA, USA). Total proteins obtained by centrifugation (13,000 x g for 20 min at 4˚C) were denatured at 95˚C for 5 min. A total of 20-50 µg of protein were size-separated by electrophoresis on 13% SDS-polyacrylamide gels and electrophoretically transferred to a nitrocellulose membrane. Non-specific binding was blocked with TBST (20 mM Tris, pH 7.6, 137 mM NaCl and 0.05% Tween-20) containing 5% non-fat milk for 2 h at room temperature (RT). The membrane was probed with the following primary antibodies for 2 h at RT: rabbit polyclonal antibody to Insulin (H-86, sc-9168), diluted 1:1,000; rabbit polyclonal antibody to CREM (X-12,sc-440), diluted 1:1,000 and rabbit polyclonal antibody to β -actin (N-21, sc-130656), diluted 1:1,000. All antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa
Cruz, CA, USA). The membrane was subsequently exposed to a horseradish peroxidase-conjugated secondary antibody (1:5,000 dilution in TBST) for 1 h at RT.

Establishment of a unitary tet-on ICER I γ expression vector. The unitary tet-on system was composed of an activator cassette and responder cassette (Fig. 1A). The activator cassette had a tTA under the control of the human insulin promoter (-1,432 to +1 nt) which is associated with maximum promoter activity in
mouse MIN6 β- cells line according to our previous study (14). The responder cassette contained the porcine ICER I γ target gene and cDNA encoding ZsGreen1 green fluorescence protein. The expression of both genes was controlled by a TRE promoter. tTA was specifically transcribed in pancreas cells under the control of the human insulin promoter, underwent a conformational change induced by the administration of dox and then bound to the TRE promoter of the responder cassette. Finally, the key gene downstream of the TRE promoter, ICER I γ , was specifically transcribed in the pancreas cells. Additionally, expression of the green fluorescence marker, ZsGreen1, was observed.

Generation and characterization of fibroblast cell lines containing the dox-inducible porcine ICER I γ construct. The dox-inducible porcine ICER I γ vector was linearized and used to transfect miniature pig fibroblasts using a liposomal-mediated gene delivery system. The transfected fibroblasts were selected and maintained with a medium containing G418 (250 µg/ml) for four weeks. Chromosomal integration of the targeting vector was confirmed by a PCR-based method using primer sets specific for the vector (Table I). Genomic DNA extracted from G418-resistant colonies was amplified with primers a and a' (product size, 992 bp; Fig. 1B) or primers b and b' (product size,
1,585 bp; Fig. 1C). In total, 12 positive colonies were obtained following two rounds of transfection (Table II). Fibroblasts from the positive colonies may serve as a cellular source for somatic cell nuclear transfer to generate a porcine model that overexpresses ICER I γ in a dox-inducible, pancreas-specific manner.

Observation of dox-inducible fluorescence in MIN6 cells and porcine fibroblasts. To observe tissue-specific expres- sion, we transiently transfected mouse pancreatic β -cells(MIN6) and porcine non-pancreatic fibroblasts. The cells were Figure 1. Schematic structure of the targeting vector and PCR-based confirmation of transgenic

co-transfected with pCMV-TA and pTRE-ZsGreen1-ICER I γ constructs as a positive control to confirm the activity of the dox-inducible expression system. In the MIN6 cells expressing pHINSP-TA-TRE-ZsGreen-ICER I γ , we observed green fluorescence indicative of dox-inducible and pancreas-specific ICER I γ expression (Fig. 2). In this regard, green fluorescence was also confirmed following dox treatment in a dose-dependent manner (0, 0.1 and 1 µg/ml). However, no fluorescence was observed in
either the untreated or dox-treated porcine fibroblasts transiently transfected with pHINSP-TA-TRE-ZsGreen-ICER I γ due to the pancreas specific insulin promoter (Fig. 3).

Dox-inducible porcine ICER I γ mRNA expression in mouse β -cells. To evaluate dox-inducible ICER I γ expression under the control of the human insulin promoter, reverse transcription (RT)-PCR was performed using mRNA obtained from transiently transfected MIN6 cells and porcine fibroblasts. Consistent to the fluorescence microscopy findings, dox-dose dependent and pancreas specific mRNA expression of ICER I γ was observed in MIN6 cells (Fig. 4A) following dox treatment but not in porcine fibroblasts (data not presented).

In rodent models, it has been reported that ICERs have an important role in T-cell mediated downregulated expression of interleukin (IL)-2, an essential growth factor for auto-aggres-sive T-effector cells (15,16). In pancreatic β -cells, ICER I γ suppresses not only insulin gene transcription but also cell replication via the reduction of cyclin A levels (7,17). Therefore, the prolonged or constitutive expression of ICER results in the development of pathological conditions (7) similar to severe
cases of diabetes mellitus, because ICER I γ overexpression significantly blocks CRE-mediated transcription by competing with CRE-binding activators (6,18). Therefore, in the present study, a tissue-specific ICER I γ vector was constructed, whose expression was controlled by dox, and a porcine fibroblast cell line with this construct was also established to generate transgenic piglets.

The insulin promoter directly regulates insulin gene transcription according to plasma glucose levels and enables β- cells to produce insulin (19,20). On this basis, the insulin promoter was utilized in the present study as a β- cell-specific regulator of gene transcription. The human insulin promoter region spanning -1,431 to +1 nt significantly increased ICER I γ gene expression in MIN6 cells in the presence of high glucose levels and dox (Fig. 4A). However, induction of ICER I γ expression in MIN6
cells slightly downregulated insulin production (Fig. 4B). It was hypothesized that the activity of the porcine ICER I γ protein would not be maximized in mouse β- cells because the porcine ICER I γ protein is documented as having 92.6-94.4% identity to mouse ICERs, using a BLAST tool. By contrast, it may be possible to reduce the interaction between porcine ICER I γ and mouse insulin promoter regulatory elements.

It was reported that the relative number and composition of principal insulin promoter regulatory elements in different species was determined through transcription factor binding site turnover and accretion (21-23). In particular, CRE is a key determinant of gene expression (24,25) and binds to the widest array of transcription factors in the insulin promoter (26). Unlike the human insulin promoter that contains four copies of CREs, the mouse insulin promoter only has a single copy (23). For this reason, the binding rate of overexpressed ICER I γ within the mouse insulin promoter may decrease and the repression of insulin expression by ICER I γ may be weaker in spite of ICER I γ overexpression via the human insulin promoter in mouse β- cells.

MIN6 cells were derived from a transgenic C57BL/6 mouse insulinoma expressing an insulin-promoter/T-antigen constructand possess characteristics of pancreatic β- cells, including the ability to secrete insulin in response to glucose (27,28). Although the amount of insulin secreted from MIN6 cells in the presence of a high (25 mM) glucose concentration is six-to seven-fold greater than that observed with a low (5 mM) glucose level, it has been noted that glucose-induced insulin secretion from the MIN6 cells may suddenly be lost during the course of passaging (28). This may result in a poor response to the downregulation of glucose-mediated insulin production
via dox-inducible ICER I γ expression. To generate a porcine model for diabetes or metabolic syndromes, pig fibroblast cell lines expressing dox-inducible ICER I γ were established. These cells should be a useful source for somatic cell nuclear transfer procedures. Porcine models provide various advantages for studying human meta-
bolic syndromes because pigs are mono-gastric omnivores and have anatomical/physiological characteristics similar to that of humans (29). The porcine model established using the created cells may offer information crucial for understanding the mechanisms underlying human diabetes mellitus. In the present study, the dox-inducible, tissue-specific expression of the transcriptional repressor ICER I γ was observed in a mouse pancreatic β- cell line. This result was compared with the expression of ICER I γ observed in transiently transfected primary porcine fibroblasts. The unitary tet-on ICER I γ induction system was pancreas-specific and dox-inducible. When generating transgenic animals, these cellular traits may reduce stillbirth rates during pregnancy and unwanted outcomes caused by uncontrollable gene expression.