Modulation of transgene expression in mesothelial cells by activation ...

Nephrol Dial Transplant (1998) 13: 1420–1429

Nephrology Dialysis Transplantation

Original Article

Modulation of transgene expression in mesothelial cells by activation of an inducible promoter Catherine M. Hoff1, Jennifer L. Cusick2, Elizabeth M. Masse2, Robert W. Jackman2, Janice A. Nagy2 and Ty R. Shockley1 1Renal Division Scientific Affairs, Baxter Healthcare Corporation, McGaw Park, IL, and 2Departments of Pathology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA, USA

Abstract Background. The efficacy of peritoneal dialysis and its success as a long-term treatment depends on the preservation of the integrity of the peritoneal membrane. With increasing time on dialysis, the membrane may become compromised resulting in decreased dialysing capacity. We have pursued an innovative strategy, i.e. genetic modification of the mesothelial cell to change the properties of the membrane to potentially improve its dialysing capacity and longevity, and have demonstrated the feasibility of this approach in a rat model of ex vivo gene transfer. The potential to regulate transgene expression in this model is examined here. Methods. Rat peritoneal mesothelial cells (MCs) were stably modified to express human growth hormone (hGH ) under control of the heavy metal ion and glucocorticoid-regulatable murine metallothionein-1 promoter. The effect of zinc and the synthetic glucocorticoid dexamethasone on hGH expression was analysed in MC clones maintained in continuous passage or stationary phase, and in our rat model of ex vivo gene transfer. Results. Exposure of these clones to zinc and dexamethasone, either singly or in combination, resulted in significant (i.e. 2–200-fold ) increases in hGH production. Zinc-induced modulation of hGH production was demonstrated in cells in continuous passage and stationary culture. Regulation was also demonstrated after ex vivo gene transfer by both the intraperitoneal administration of zinc ions or the systemic administration of dexamethasone. Conclusions. Our results demonstrate the modulation of transgene expression in MCs in vitro and in vivo, and suggest the potential for the regulation of gene expression in a genetically modified mesothelium that may ultimately be used for the delivery of therapeutic proteins to maintain peritoneal membrane viability in the peritoneal dialysis patient.

Correspondence and offprint requests to: Catherine M. Hoff PhD, Renal Division Scientific Affairs–MPGR–R2, Baxter Healthcare Corporation, 1620 Waukegan Rd, McGaw Park, IL 60085, USA.

Key words: dexamethasone; ex vivo gene transfer; gene therapy; heavy metal ions; human growth hormone; inducible promoter; intraperitoneal; mesothelial cells; mMT-1; peritoneal cavity; peritoneal dialysis; peritoneal mesothelium; systemic delivery

Introduction In peritoneal dialysis (PD), the peritoneal membrane functions as a semi-permeable membrane that regulates the selective transport of water and solutes between the systemic circulation and the peritoneal cavity [1]. The efficacy of PD and its success as a long-term treatment depend, therefore, on the maintenance of the dialysing capacity and integrity of the peritoneal membrane. Ironically, PD solutions are considered to be largely bioincompatible [2] and may have adverse effects on the mesothelium [3–7] and on host defence mechanisms [8–10] which in turn may contribute to eventual peritoneal membrane failure [11]. New and innovative strategies to improve or enhance the durability of the membrane in order to prevent injury during peritoneal dialysis must be investigated. To this end we have developed a molecular genetic approach, i.e. to genetically modify peritoneal mesothelial cells (MCs) to express potentially therapeutic proteins, and through this modification, change the properties of the membrane. We have previously established a rat model of ex vivo MC-mediated gene therapy. Specifically, we have shown that MCs can be isolated from rat parietal peritoneum, genetically modified, and then implanted onto the denuded parietal peritoneal surface of syngeneic recipients [12]. We have demonstrated MC -mediated production of an intracellular marker protein, b-galactosidase [12]; the delivery of a MC-secreted protein, human growth hormone (hGH ), to the systemic circulation [13]; and the expression of a membrane-bound recombinant protein, thrombomodulin, on the MC surface [14]. These studies confirm that genetically modified MCs are effective for both the

© 1998 European Renal Association–European Dialysis and Transplant Association

Regulation of mesothelial-cell-mediated transgene expression

localized and the systemic delivery of recombinant proteins, and suggest that such a strategy might be used to modify and/or improve the function of the mesothelium, particularly in the PD patient. A successful gene therapy strategy must consist not only of a system by which the DNA encoding the therapeutic protein is delivered to the target cell population, but also of a means by which the expression of the transgene can be controlled. Regulation of gene expression can be accomplished by a number of mechanisms, including transcriptional regulation through use of promoters that respond to specific physiological signals, antibiotics, and synthetic steroids (reviewed in [15]). To demonstrate the feasibility of regulating transgene expression in our MC-mediated model of gene therapy, we chose a model construct containing the gene for hGH driven by the murine metallothionein (mMT-1) promoter. This construct promotes hGH gene expression that can be modulated by heavy metal ions or by glucocorticoids. Although regulation of hGH expression using this construct has been demonstrated previously both in cell culture [16–19] and in in vivo studies of gene expression [20–22], the results presented here are the first to demonstrate regulation of transgene expression in the peritoneal MC. In the present study we monitored the enhancement of mMT-1 promoter-driven hGH expression in genetically modified MCs by the heavy metal ion zinc and by the synthetic glucocorticoid dexamethasone. We compared hGH transgene expression in continuously passaged MCs and in stationary confluent culture in the presence and absence of added inducer molecules. In addition, we tested the responsiveness of these hGHproducing MCs to both zinc and dexamethasone following their implantation onto the peritoneal surface. Our results in this model system demonstrate that genetically modified MCs are responsive to the administration of regulator molecules both in cell culture and in vivo, and importantly, that transgene expression in the genetically modified peritoneal mesothelium can be upregulated by either local (i.e. intraperitoneal ) or systemic administration of inducer molecules. These results support our previous conclusion that genetic modification of the peritoneal mesothelium is a viable strategy for enhancing the function of the peritoneal membrane, and suggest the potential for regulating MC-mediated transgene expression.

Subjects and methods Mesothelial cell isolation and culture Rat primary MCs were isolated by enzymatic digestion of the parietal peritoneal wall as previously described [12]. This strategy yielded a phenotypically pure population of cells that stained positively for a number of MC-specific cytokeratins and adopted a cobblestone appearance at confluence. The cells used in this study, Clone B, were grown in DME/F12 (151) media (Sigma Chemical Co., St Louis, MO) supplemented with 10% Fetal Bovine Serum (FBS,

1421

Gibco BRL), 2X antibiotic/antimycotic (Sigma) and 2 mM L-glutamine (Sigma). Clone B cells maintained in continuous passage were split every 6–7 days; cells maintained in stationary phase were grown to confluence and the medium changed every 3–4 days.

Plasmids pXGH5 (Nichols Institute Diagnostics, San Juan Capistrano, CA) is a pUC12-derived expression vector containing a 817 bp coding sequence for human growth hormone (hGH ) under the control of the murine metallothionein (mMT-1) promoter [17]. pcDNA3 (Invitrogen, San Diego, CA) contains the gene for neomycin resistance under the control of the immediate early CMV enhancer-promoter sequences.

Stable transfection of MCs with pXGH5 and pcDNA3 Clone B cells were transfected by electroporation with plasmids pXGH5 and pcDNA3 to obtain MCs stably secreting hGH. Plasmids pXGH5 and pcDNA3 were mixed in a 951 molar ratio, and 50 mg total DNA was added to 3×106 cells; the mixture was pulsed at 400 V, 960 mF (Gene Pulser, Bio-Rad, Hercules CA) and immediately chilled for 10 min. The cells were plated in regular growth medium which was replaced at 48 h with medium containing 100 mg/ml geneticin (G418, Sigma). G418-resistant colonies were picked and screened for hGH production.

Measurement of hGH concentration The concentration of hGH present in conditioned medium from transfected MCs or in rat serum samples was measured using a solid-phase two-site RIA (Nichols Institute Diagnostics) according to the manufacturer’s instructions. Production of hGH in cell culture was calculated as mg hGH/106 cells/day based on the cell number at the time of sampling and the volume of culture medium. Untransfected Clone B cells were consistently at or below the level of detection for this assay (0.5 ng/ml ), indicating little or no cross-reactivity between rat and human GH in this assay.

Southern blot analysis DNA was isolated from monolayer cultures of transfected and control MCs using DNAzolTM (Life Technologies, Gaithersburg, MD) as described by the manufacturer. Ten micrograms of each DNA sample was digested with Mse-I (New England Biolabs, Beverly, MA), electrophoresed on a 0.8% agarose gel in 0.5X TBE buffer, and transferred to a MS Nytran nylon membrane (Schleicher & Schuell, Keene, NH ) using the S&S TurboBlotterTM Rapid Downward Transfer System. A probe specific to 769 bases within the hGH sequences of pXGH5 was generated by PCR using the following primers: upstream (5∞CAGGGAGGAAACACAACAGAAATC3∞) and downstream (5∞TGGGGAGGGGTCACAGG3∞), and labelled with [a32P]-dCTP (Amersham Corporation, Arlington Heights, IL) using the RadPrime DNA Labeling System (Life Technologies). Hybridization was performed as described previously [23] with a final wash of 0.1X SSC+0.5% SDS, and the membrane was exposed to Kodak AR X-ray film at −80°C for 24 h. Quantification of 32P was performed by cutting out the individual bands, eluting the bound probe with 0.1 N NaOH, and counting in a LS-5000 scintillation counter (Beckman Instruments, Schaumburg, IL).

1422

Modulation of hGH production in vitro MCs were cultured in regular growth medium supplemented with concentrations of zinc sulphate (ZnSO .7H O) ranging 4 2 from 50 mM to 500 mM, or with concentrations of dexamethasone ranging from 0.5 mM to 20 mM. In some assays, cultures were supplemented with medium containing both 250 mM zinc and 1 mM dexamethasone. Cell culture medium was sampled at various timepoints, and stored at 4°C until analysed for hGH content. Cell number was determined at the end of each time period by trypsinization of the culture and counting in a haemacytometer.

Animals Fischer rats (female, 150–170 g, 10–15 weeks, Charles River Laboratories, Wilmington MA) were housed one per cage and fed a regular diet. All rats received subcutaneous injections (5 mg/kg/day) of cyclosporin-A (Sandimmun concentrate for injections; Sandoz Pharmaceuticals, East Hanover, NJ ) for immunosuppression starting 24 h prior to MC transplantation surgery and continuing for the duration of the study.

Reseeding of genetically modified MCs onto the parietal peritoneal surface Genetically modified MCs were transplanted into the peritoneal cavity of cyclosporin-A treated syngeneic recipients as previously described [12]. Briefly, animals were anaesthetized and a 16 cm2 portion of the peritoneal surface was removed by Gelfilm denudation. The abdominal cavity was then sutured closed and each animal received an i.p. injection of 1×107 genetically modified MCs. At various times after surgery, animals were anaesthetized with Metofane (PitmanMoore, Mundelein, IL) and a blood sample taken by retroorbital puncture into a known volume of heparin. Serum samples, prepared by centrifugation (14 000 g, 10 min.), were analysed to determine the concentration of hGH.

Modulation of hGH expression by genetically modified MCs in vivo To determine the effect of local administration of zinc on hGH expression in rats implanted with genetically modified MCs, 2 ml of 250 mM ZnSO in PBS was delivered i.p. 34 4 days after MC reimplantation surgery. Blood samples were taken by retro-orbital puncture and serum prepared at various times (9 h, then 1, 2, 3, 5, 7 days and then weekly) after i.p. ZnSO injection, and hGH serum concentrations 4 determined. To study the effect of glucocorticoids on hGH expression in rats implanted with genetically modified MCs, rats were given drinking water supplemented with 1.2 mg/litre dexamethasone. Animals were bled by retroorbital puncture and serum samples analysed for hGH concentration.

Statistical methods Evaluation of serum hGH concentration in experimental and control animals was made using the non-parametric Mann–Whitney test to calculate a two-tailed P value. Linear regression was used to determine if hGH production in vitro

C. M. Hoff et al.

was dependent on time in culture. The level of significance used for all analyses was P