Rapamycin Control of Exocrine Protein Levels in Saliva After Adenoviral Vector-Mediated Gene Transfer

Home , Parasympathomimetic drug

Gene Therapy (2004) 11, 729–733 & 2004 Nature Publishing Group All rights reserved 0969-7128/04 $25.00 www.nature.com/gt BRIEF COMMUNICATION Rapamycin control of exocrine protein levels in saliva after adenoviral vector-mediated gene transfer

J Wang, A Voutetakis, C Zheng and BJ Baum Gene Therapy and Therapeutics Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, DHHS, Bethesda, MD, USA

Transgene-encoded therapeutic secretory proteins can saliva were determined. Salivary hGH levels were rapamycin be efﬁciently secreted from salivary glands into saliva or concentration dependent. At a rapamycin dose of 10 mg/kg, the bloodstream after adenoviral (Ad)-mediated gene trans- total salivary hGH was 6937197 ng and the hGH concen- fer. Since transgene expression from conventional vectors tration in saliva was 4.671.3 mg/ml. Over a 16-day experi- is typically unregulated, we evaluated the rapamycin-based mental period, three separate administrations of rapamycin dimerizer regulation system for control of transgene (3 mg/kg) induced distinct elevations of salivary hGH expression in, and consequent exocrine protein secreted (B100–200 ng total hGH) that were entirely rapamycin from, rat salivary glands. We used human growth hormone dependent. This study demonstrates for the ﬁrst time (hGH) as a surrogate exocrine secretory protein. Two pharmacological control of transgenic exocrine protein Ad vectors, Ad C4ZF3, encoding activation and DNA production and presence in saliva after salivary gland binding domain fusion polypeptides, and Ad Z12-I-GH-2, gene transfer, and the potential for its application to the encoding hGH, were constructed and shown useful in vitro. management of oral, oropharyngeal and upper gastrointest- Thereafter, both vectors were delivered into submandibular inal tract disorders. glands by retroductal infusion. After 24 h, rapamycin (0, 1, 3 Gene Therapy (2004) 11, 729–733. doi:10.1038/sj.gt.3302225 or 10 mg/kg) was administered, and 20 h later hGH levels in Published online 22 January 2004

Keywords: salivary glands; saliva; adenoviral vectors; rapamycin; gene therapeutics; gene regulation; human growth hormone

Salivary glands are secretory organs, capable of secreting delivery schedule.8,9 As a result, there has been much substantial amounts of protein both into saliva (exocrine) effort to control the time and extent of therapeutic and into the bloodstream (endocrine). These tissues have protein expression following gene transfer. One system proven to be useful target sites for delivery of genes that has shown promise for this purpose is termed encoding therapeutic proteins.1,2 Salivary glands can be dimerizer-mediated gene regulation, and is based on accessed noninvasively by intraoral cannulation of the transcriptional control of the transgene by rapamycin or main excretory ducts. Effective gene transfer into a derivative drug.8,9 With this system, the DNA-binding salivary cells has been achieved with both adenoviral and activation domains of a chimeric transcription factor (Ad)3 and adenoassociated viral4 vectors. are produced as separate polypeptides, each fused to the As in many other well-differentiated eukaryotic cells, drug-binding domain of the human proteins FKBP or secretory proteins expressed in salivary gland cells FRAP.8 In the presence of rapamycin, these two follow one of two main pathways for secretion; consti- components can form a heterodimer, thus reconstituting tutive or regulated.5,6 Salivary glands consist of polarized a functional transcription factor. In the ﬁrst generation of epithelial cells, from which transgene-encoded proteins this system, two vectors are used. One encodes both the using the constitutive pathway are primarily secreted activation domain and the DNA-binding domain, sepa- basolaterally into the bloodstream, while those taking the rated by an internal ribosome entry site, while the other regulated pathway are secreted apically, into saliva.1,2 encodes a transgene downstream of an engineered For novel treatments of oral, oropharyngeal and upper binding site for the DNA-binding domain and a minimal gastrointestinal tract disorders, salivary glands may be promoter.8–10 The rapamycin-based transcriptional con- particularly useful as gene transfer targets versus more trol system has proved useful in several diverse tissues, distal systemic gene delivery sites.7 including muscle,10,11 liver,12 and eye.13 We are unaware Transgene expression with conventional gene therapy of any reported applications of this dimerizer system to, vectors is unregulated and continuous. For clinical use, or any other system regulating gene expression of, this could result in both protein expression well beyond exocrine secretory proteins. The purpose of the present the therapeutic range, as well as an unsatisfactory study was to determine if the rapamycin-based system was useful to control the levels of a model-regulated Correspondence: Dr BJ Baum, GTTB, NIDCR, NIH, DHHS, Bldg. 10, pathway secretory protein found in saliva. Rm. 1N113, MSC-1190, Bethesda, MD 20892-1190, USA For this proof of concept study, we chose human Received 7 August 2003; accepted 20 November 2003 growth hormone (hGH) as a model exocrine protein. Rapamycin control of exocrine protein levels J Wang et al 730 When expressed in salivary glands, hGH primarily is published in vitro studies examining rapamycin (and its secreted through the regulated pathway in an exocrine analogues) control of gene expression in other cell lines.19 manner into saliva.2 Importantly, hGH has a short half Next, rats were administered both viruses (1:1; 2 Â 1011 life in rats (protein B45 min;2 mRNA B1.5 days),14 total viral particles/submandibular gland) via retrograde rendering it a reasonable candidate gene to be studied salivary duct delivery, and 24 h later the animals were using Ad vectors over a 2-week period. We constructed treated (intraperitoneal injection) with different doses of two recombinant, E1À, type 5 Ad vectors. AdC4ZF3 rapamycin (0, 1, 3, 10 mg/kg), based on previous in vivo contained the two components of the chimeric transcrip- studies with mice.9,10 Rapamycin was used for in vivo tion factor, while AdZ12-I-GH-2 encoded hGH. experiments (in place of AP21967) because of the amount Initially, we used these two Ad vectors in vitro to of drug required. After 20 h, we stimulated the secretion examine the effects of the dimerizer drug on hGH of saliva with the parasympathomimetic drug pilocar- expression. For these studies, we used A5 cells, a rat pine.2,4,7 As with the in vitro studies, we found that submandibular gland cell line.15 After infection with salivary hGH levels were dependent on the rapamycin AdC4ZF3 and AdZ12-I-GH-2 (1:1) at an MOI of either dose administered (Figure 2). At the highest dose 100 or 200 particles/cell, dimerizer-regulated hGH (10 mg/kg) of rapamycin used, the average total hGH expression was readily detected in culture media (Figure found in saliva was 693 ng. This corresponds to an 1). The expression of hGH was dependent on the average salivary hGH concentration of 4.671.3 mg/ml, presence of a rapamycin analogue (AP21967) over a and is signiﬁcantly greater than hGH levels at the two wide dosage range and was most effective at concentra- lower rapamycin doses used (3 mg/kg, P ¼ 0.029; 1 mg/ tions Z10 nM. On average, maximum hGH production kg, P ¼ 0.022). At 1 mg/kg, the average concentration of (48 mg/ml) was seen after 48 h with 100 nM dimerizer hGH in saliva was B1.5 mg/ml, while at 3 mg/kg the drug. Below 1 nM dimerizer drug, little secreted hGH average salivary hGH concentration was B2.1 mg/ml. was detected. In the absence of the dimerizer drug, hGH Interestingly, at this latter dose, the average total amount expression was less than the assay detection limit of hGH found in saliva was 323 ng and in serum was (0.02 ng/ml). These results are similar to previously B17 ng (a 19-fold difference; similar to results previously reported).2 Importantly, in the absence of the dimerizer drug, there was no detectable hGH in saliva or serum, that is, below the assay detection limit. Finally, we wished to determine if rapamycin’s presence could regulate hGH expression over time in

Figure 1 Rapamycin analogue-dependent hGH production in vitro in A5 cells. The EcoR I/BamH I fragment of the bicistronic transcript from the

plasmid pC4N2-RHS3H/ZF3 (ARIAD Pharmaceuticals, Cambridge, MA, USA) was inserted directionally into pACCMV-pLpA.16 A cassette containing the hGH cDNA was cut with Mlu I and BamH I from pZ12-I-hGH-2 (ARIAD Pharmaceuticals), blunted and then inserted into pACCMV-pLpA after it had been cut with Not I and blunted. Plasmid structures were confirmed by restriction enzyme digestion and by Figure 2 Rapamycin-dependent hGH production in rats. Male Wistar rats AP21967-induced hGH production after transient transfection of A5 (250–280 g; Harlan–Sprague–Dawley, Walkersville, MD, USA) were cells. Recombinant, E1À Ads were generated and purified by standard anesthetized by an intraperitoneal (i.p.) injection of ketamine chloride methods.16,17 The vectors were screened for replication-competent Ad by a (60 mg/kg) and xylazine (8 mg/kg) (Phoenix, St Joseph, MO, USA), and quantitative PCR (QPCR) assay, essentially as described.18 The assay the orifices of the submandibular glands were cannulated as described.1 amplified the Ad E1 gene using the plasmid pBHG10 (Microbix Atropine (100 ml, 0.5 mg/kg; Sigma, St Louis, MO, USA) was Biosystems, Toronto, Canada) as negative control. The primers used in administered 10 min prior to retroductal infusion of viral vectors. this QPCR assay were as follows: forward primer, CTT GAG TGGCAG AdC4ZF3 and AdZ12-I-GH-2 (1:1; 2 Â 1011 total particles/gland) in CGA GTA GAG TT, and reverse primer, TAT GTC TCA TTT TCA GCA 150 ml virus dilution buffer (pH 7.4–7.5, 5 mM MgCl2,10mM Tris, 20% GTC CCG GT. The Ad vectors used herein gave results with this assay glycerol) were infused. Rapamycin was initially dissolved in N, N- that were indistinguishable from the background obtained with pBHG10 dimethylacetamide (DMA) to yield a stock solution, and shortly before use (17.377.8 copies/107 molecules; t ¼ 0.025; P ¼ 0.98). A5 cells were the stock was mixed with an equal volume of PEG-400 and Tween-80 (9:1). cultured in McCoy’s 5A media supplemented with 10% fetal bovine After 24 h, rats were injected i.p. with rapamycin at the indicated 5 serum, at 371C with 5% CO2. Cells (5 Â 10 /well, 12-well plates) were concentrations, or with vehicle, in a total volume of 600 ml. After 20 h, infected with a mixture of viruses (1:1; at the indicated doses). After 24 h pilocarpine (5 mg/kg, subcutaneous injection, Sigma, St Louis, MO, incubation, fresh media containing the rapamycin analogue AP21967 USA)-stimulated whole saliva samples were collected for 20 min, placed on (ARIAD Pharmaceuticals) at the indicated concentrations was added to dry ice, and stored at À801C until gravimetric and hGH analyses. For the cells. hGH expression in the media was subsequently detected after 24 and hGH assay, saliva samples were diluted as needed in TE buffer (pH 8.0). 48 h using an hGH chemiluminescence immunoassay kit (Nichols The background hGH level measured in naı¨ve animals was subtracted from Institute Diagnostics, San Juan Capistrano, CA, USA). Assays were all experimental values. The data shown represent the concentration of performed in triplicate, and the lower limitation of hGH detection was hGH found in saliva (means 7 s.e.m.; n ¼ 4). Statistically significant 0.02 ng/ml. Values of hGH shown are means ¼ (n ¼ 3). differences are shown; (a) P ¼ 0.022; (b) P ¼ 0.029.

Gene Therapy Rapamycin control of exocrine protein levels J Wang et al 731 salivary glands in vivo. Vectors were delivered to rat vectors, but no dimerizer drug, displayed no detectable submandibular glands as above, and at various times salivary hGH. Furthermore, rapamycin was able to thereafter rapamycin was administered intraperitoneally. induce similar elevations of hGH in saliva at three For these studies, we used the intermediate dose of separate times over a 16-day experiment. rapamycin (3 mg/kg). This dose is slightly less than that This latter finding suggests that most transduced used previously to regulate hGH expression in murine salivary cells were able to survive during this time muscle (5 mg/kg).10 Saliva was subsequently collected period and produce substantial amounts of hGH, a on several days and the total amount of salivary hGH somewhat unexpected result based on our previous was determined. As shown in Figure 3, we were able to studies with first-generation Ad vectors.20 Rapamycin is induce three distinct elevations of hGH expression with a potent immunosuppressive agent, inhibiting a wide the dimerizer drug over the experimental time course. At spectrum of T- and B-cell activities.21 These activities, in 20 h following the first administration of rapamycin, the conjunction with the administration of dexamethasone to total salivary hGH was 94736 ng. Thereafter, salivary rats, which we used to blunt the adenoviral immune hGH returned towards background levels. Two subse- response,1 may have contributed to the prolonged Ad- quent administrations of rapamycin led to two addi- mediated transgene expression seen herein. In an earlier tional and distinct elevations of hGH in saliva, 177758 study examining rapamycin-based regulation of hGH and 100749 ng, respectively. All three elevations in expression in muscle of immunocompetent Balb/c mice, salivary hGH clearly corresponded with the previous utilizing Ad vectors similar to those employed herein, administration of the dimerizer drug. Importantly, the Rivera et al10 were unable to elicit hGH expression control animals, that is, receiving the same dose of the following a second administration of the dimerizer drug. two viral vectors per gland but no rapamycin, displayed The present results may be a reflection of the immuno- no detectable salivary hGH. These results are similar logically different tissue environment in which gene to those previously observed for rapamycin induction transfer occurred. of erythropoietin expression in serum in mice.11 As we have previously reported, following transfer of Our study demonstrates the potential utility of the hGH gene into salivary glands, hGH protein is rapamycin-based regulation of transgene expression in primarily secreted into saliva, consistent with it being salivary glands of immunocompetent rats with a model- secreted in the regulated secretory pathway.2 The total regulated secretory pathway protein, hGH. The expres- amount of hGH secreted into saliva is B10–25-fold sion and appearance of hGH in saliva after stimulation greater than the total amount of hGH found in serum.2,22 with the secretogogue pilocarpine was responsive to, and Thus, despite its physiological role as an endocrine dependent on, rapamycin administration. Rapamycin hormone, hGH serves as an excellent model for an elicited hGH production in vitro and in vivo in a dose- exocrine secretory protein when expressed in a salivary dependent manner. Control animals receiving both Ad gland.2,22 The concentrations of hGH measured in saliva were quite high, averaging B4.5 mg/ml following administration of the 10 mg/kg rapamycin dose. In general, the pattern of hGH expression in rat salivary glands is quite similar to that reported for immunodeficient mouse muscle by Rivera et al,10 and likely reflect post- transcriptional events, such as mRNA decay rates rather than the clearance of rapamycin.9 Many factors can affect the duration of transgene expression, including the type of vector used, the host response to the vector and the nature of the transgene. Ad vectors have been extensively used in preclinical, in vivo salivary gland gene transfer studies, because of their ease of production and because they can efficiently infect both dividing and nondividing cells.1,22–24 However, Ad vector expression in salivary glands, as in other tissues,25 is typically short lived due to their induction of a potent immune response.1,20 Conversely, AAV-mediated transgene expression in salivary glands is long lived.4,26 The use of a dimerizer regulation system in the context of AAV vectors should permit extended control of trans- Figure 3 hGH appearance in saliva in rats after rapamycin administra- gene expression in salivary glands as it has in other tion. Rat submandibular glands were cannulated and the Ad vectors tissues.11–13 AdC4ZF3 and AdZ12-I-GH-2 (1:1) were infused into Wharton’s duct at a total dose of 2 Â 1011 particles/gland. Control animals received both Ad Our finding that rapamycin can regulate significant vectors, but no rapamycin, that is, they were injected with the DMA, exocrine protein expression in salivary glands over a 16- PEG400 and Tween-80 vehicle. Rats also received a daily intramuscular day period suggests that dimerizer drug regulation may injection of dexamethasone (1 mg/rat) during the first 4 days of this have applications for gene therapeutics in managing experiment. Arrows indicate the times of rapamycin administration. specific oral, oropharyngeal and upper gastrointestinal Pilocarpine-stimulated whole saliva samples were collected into pre- tract disorders. Indeed, the concentrations of the model weighed 1.5 ml Eppendorf tubes at the days indicated (data points) and put exocrine transgenic protein (hGH), achieved in saliva in dry ice immediately until assayed exactly as in Figure 2. The values of B total salivary hGH shown are means7s.e.m. (n ¼ 6), and were calculated ( 1–5 mg/ml) herein, are consistent with two poten- as previously described.2 No detectable hGH was found in the saliva from tially valuable applications; for treating oral bacteria control animals (not shown). and fungal species that are resistant to conventional

Gene Therapy Rapamycin control of exocrine protein levels J Wang et al 732 antibiotics and antifungals, respectively, and for enhan- 4 Yamano S et al. Recombinant adeno-associated virus serotype cing the healing of oral ulcers. 2 vectors mediate stable interleukin 10 secretion from For example, there is now considerable concern about salivary glands into the bloodstream. Hum Gene Ther 2002; 13: the significant morbidity resulting from emerging anti- 287–298. biotic-resistant oral and oropharyngeal bacteria.27–30 5 Kelly RB. Pathways of protein secretion in eukaryotes. Science Antimicrobial peptides encoded by single genes, such 1985; 230: 25–32. as the b-defensins,31,32 can be introduced into salivary 6 Castle D, Castle A. Intracellular transport and secretion of glands by the gene transfer approach used here7 and are salivary proteins. Crit Rev Oral Biol Med 1998; 9: 4–22. effective in the 1–5 mg/ml concentration range.31 Simi- 7 O’Connell BC et al. Transfer of a gene encoding the anticandidal larly, oral candidiasis is relatively common in immuno- protein histatin 3 to salivary glands. Hum Gene Ther 1996; 7: suppressed (AIDS, transplant) patients,33,34 and Candidal 2255–2261. 8 Rivera VM et al. A humanized system for pharmcological control species are becoming increasingly resistant to azole-type of gene expression. Nat Med 1996; 2: 1028–1032. drugs.34,35 Genes encoding anticandidal peptides, such as 7,34 9 Rivera VM. Controlling gene expression using synthetic ligands. the histatins, also can be readily introduced into Methods 1998; 14: 421–429. salivary glands. Importantly, a typical course of anti- B 10 Rivera VM et al. Long term regulated expression of growth microbial therapy for oral infections would be for 10– hormone in mice after intramuscular gene transfer. Proc Natl 14 days, that is, similar to that shown possible herein Acad Sci USA 1999; 96: 8657–8662. with Ad vectors. 11 Ye X et al. Regulated delivery of therapeutic proteins after in vivo Oral ulcerations, both those occurring iatrogenically somatic cell gene transfer. Science 1999; 283: 88–91. (eg, following cytotoxic chemotherapy and therapeutic 12 Auricchio A et al. Constitutive and regulated expression of irradiation during the treatment for cancers), and those processed insulin following in vivo hepatic gene transfer. Gene resulting from disease (eg, Bechet’s syndrome), are Therapy 2002; 9: 963–971. common, painful and difficult to treat. Recently, pre- 13 Auricchio A et al. Pharmacological regulation of protein clinical and clinical studies have indicated that certain expression from adeno-associated viral vectors in the eye. Mol growth factors (eg, keratinocyte growth factor,36 basic Ther 2002; 6: 238–242. fibroblast growth factor;37 epidermal growth factor37) 14 Helms SR, Rottman FM. Characterization of an inducible and cytokines (eg, interleukin-11;38 interferon-a 2a39)may promoter system to investigate decay of stable mRNA be beneficial for treating these ulcerative conditions. The molecules. Nucleic Acids Res 1990; 18: 255–259. recombinant growth factors or cytokines were delivered 15 Brown AM, Rusnock EJ, Sciubba JJ, Baum BJ. Establishment and in these studies either by frequent topical application or characterization of an epithelial cell line from the rat injection. The therapeutically required concentrations of submandibular gland. J Oral Pathol Med 1989; 18: 206–213. these proteins are much lower than those achieved in the 16 Becker TC et al. Use of recombinant adenovirus for metabolic present study (eg, ng/ml). Controlled expression of the engineering of mammalian cells. In: Roth MG (ed). Methods in Cell Biology. Academic Press: San Diego, 1994, pp 161–189. genes for such proteins would require a time course 17 Mastrangeli A et al. Direct in vivo adenovirus-mediated gene similar to that shown herein, likely be much less transfer to salivary glands. Am J Physiol 1994; 266: G1146–G1155. expensive, and provide much higher local bioavailability 18 Zheng C, Wang J, Baum BJ. Integration efficiency of a hybrid than achieved by the conventional protein delivery adenoretroviral vector. Biochem Biophys Res Commun 2003; 300: methods. 115–120. In conclusion, our data show that effective and 19 Pollock R et al. Delivery of a stringent dimerizer-regulated gene repeatable rapamycin-regulated hGH expression, and expression system in a single retroviral vector. Proc Natl Acad Sci consequent appearance in pilocarpine-stimulated saliva, USA 2000; 97: 13221–13226. can be achieved using Ad vector delivery of the required 20 Kagami H et al. Repetitive adenovirus administration to the regulatory components to salivary glands in vivo. The parotid gland: role of immunological barriers and induction of expression of hGH is dependent on the time and dose of oral tolerance. Hum Gene Ther 1998; 9: 305–313. rapamycin administration, and no hGH is detected in 21 Chen Y et al. A putative sirolimus (rapamycin) effector protein. saliva in the absence of the dimerizer drug. Biochem Biophys Res Commun 1994; 203: 1–7. 22 Hoque AT, Baccaglini L, Baum BJ. Hydroxychloroquine enhances the endocrine secretion of adenovirus-directed Acknowledgements growth hormone from rat submandibular glands in vivo. Hum We thank Dr Victor Rivera and ARIAD Pharmaceuticals Gene Ther 2001; 12: 1333–1341. (www.ariad.com/regulationkits) for providing the plas- 23 Kagami H, O’Connell BC, Baum BJ. Evidence for the systemic mids, rapamycin and AP21967 used in these experi- delivery of a transgene product from salivary glands. Hum Gene ments. We also thank Dr Rivera, and Drs J Chiorini, A Ther 1996; 7: 2177–2184. 24 Delporte C et al. Increased fluid secretion after adenoviral- Cotrim, M Schmidt and R Wellner for their comments on mediated transfer of the aquaporin-1 cDNA to irradiated rat an earlier draft of this manuscript. salivary glands. Proc Natl Acad Sci USA 1997; 94: 3268–3273. 25 Yang Y et al. Cellular immunity to viral antigens limits E1- References deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA 1994; 91: 4407–4411. 1 Baum BJ, Wellner RB, Zheng C. Gene transfer to salivary glands. 26 Voutetakis A et al. Long term functional erythropoietin Int Rev Cytol 2002; 213: 93–146. production from salivary glands after rAAV-mediated gene 2 Baum BJ et al. Polarized secretion of transgene products from transfer. Mol Ther 2003; 7: S185–S186; (Abstr. 474). salivary glands in vivo. Hum Gene Ther 1999; 10: 2789–2797. 27 Andersson DI. Persistence of antibiotic resistant bacteria. Curr 3HeXet al. Systemic action of human growth hormone following Opin Microbiol 2003; 6: 452–456. adenovirus-mediated gene transfer to rat submandibular glands. 28 Bryskier A. Viridans group streptococci: a reservoir of resistant Gene Therapy 1998; 5: 537–541. bacteria in oral cavities. Clin Microbiol Infect 2002; 8: 65–69.

Gene Therapy Rapamycin control of exocrine protein levels J Wang et al 733 29 Martin JM et al. Erythromycin-resistant group A streptococci in 35 De Lucca AJ, Walsh TJ. Antifungal peptides: novel therapeutic schoolchildren in Pittsburgh. N Engl J Med 2002; 346: 1200–1206. compounds against emerging pathogens. Antimicrob Agents 30 Erickson PR, Herzberg MC. Emergence of antibiotic resistant Chemother 1999; 43: 1–11. Streptococcus sanguis in dental plaque of children after frequent 36 Dorr W et al. Modification of oral mucositis by keratinocyte antibiotic therapy. Pediatr Dent 1999; 21: 181–185. growth factor: single radiation exposure. Int J Radiat Biol 2001; 31 Maisetta G et al. Activity of human b-defensin 3 alone or 77: 341–347. combined with other antimicrobial agents against oral bacteria. 37 Fujisawa K, Miyamoto Y, Nagayama M. Basic fibroblast Antimicrob Agents Chemother 2003; 47: 3349–3351. growth factor and epidermal growth factor reverse impaired 32 Huang GT-J et al. A model for antimicrobial gene therapy: ulcer healing of the rabbit oral mucosa. J Oral Pathol Med 2003; demonstration of human b-defensin 2 antimicrobial activities in 32: 358–366. vivo. Hum Gene Ther 2002; 13: 2017–2025. 38 Sonis ST et al. Defining the mechanisms of action of interleukin- 33 Pinjon E et al. Molecular mechanisms of itraconazole resistance in 11 on the progression of radiation-induced oral mucositis in Candida dubliniensis. Antimicrob Agents Chemother 2003; 47: 2424–2437. hamsters. Oral Oncol 2000; 36: 373–381. 34 Rothstein DM et al. Anticandida activity is retained in P-113, a 12 39 Alpsoy E et al. Interferon alfa-2a in the treatment of Bechet’s amino acid fragment of histatin 5. Antimicrob Agents Chemother disease: a randomized placebo-controlled and double-blind 2001; 45: 1367–1373. study. Arch Dermatol 2002; 138: 467–471.

Gene Therapy