Salivary secretions maintain the health of the oral cavity. Building on our past studies of saliva formation and its alteration during pathology, we are developing novel approaches to treat salivary gland (SG) dysfunction primarily using principles of gene therapy and tissue engineering. During this reporting period we have made considerable progress in many facets of our work. Our studies are directed at fundamental questions necessary to move such therapy into the clinic for Phase I trials in three areas: irradiation (IR)-induced salivary hypofunction, systemic single protein deficiency disorders (SSPDDs), and Sjogren?s syndrome (SS).? ? The treatment of most head and neck cancer patients includes IR. SGs in the IR field suffer irreversible damage. Over the last 10+ years, we have studied the value of transferring the human aquaporin-1 (hAQP1) gene to restore salivary flow in such patients. In the past year, we conducted an extensive safety (toxicology and biodistribution) study of AdhAQP1 in rats (100 of each gender over 92 days). This study revealed no significant vector-associated effects, other than dose-related inflammatory changes in the targeted gland, after SG administration of AdhAQP1. Based on these and previously reported findings, we developed a clinical gene therapy protocol using the AdhAQP1 vector. The protocol, ?Open-label, dose-escalation study evaluating the safety of a single administration of an adenoviral vector encoding human aquaporin-1 to one parotid salivary gland in individuals with irradiation-induced parotid salivary hypofunction?, has been reviewed and approved by the NIDCR-Institutional Review Board, the NIH-Institutional Biosafety Committee, the NIH Radiation Safety Committee and the Recombinant DNA Advisory Committee. Our application to the FDA to conduct this protocol has just been submitted. Specifically, we proposed that AdhAQP1 be used in a Phase 1/2 clinical trial in adults with a history of therapeutic irradiation for a head and neck cancer that has resulted in severe parotid hypofunction and significant evidence of related oropharyngeal morbidities. There are two objectives: (i) to test the safety of AdhAQP1 when administered to a single parotid gland in a dose escalation design, and (ii) to determine if transfer of the hAQP1 transgene to irradiation-damaged parotid glands will result in increased parotid saliva secretion and a lessening of xerostomia (dry mouth) complaints in patients. The vector will be administered in ~1mL buffer following direct cannulation of Stensen?s duct with a flexible catheter. Up to 5 dose levels (n=3 patients/dose) will be used (one dose/patient). The doses will range from 1.4 x 10e8 particle units (pu)/gland to 1 x 10e11 pu/gland. ? ? We also have examined the usefulness of a non-gene therapy approach to preventing IR-induced salivary hypofunction in mice that involves administration of the stable nitroxide 4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl (Tempol). During this reporting period, we have made an important step forward with Tempol, testing its efficacy in a fractionated radiation dose model. Tumor-bearing mice (SCCVII tumor, a squamous cell carcinoma similar to oral cavity tumors, 8 mm diameter) were treated with a fractionated IR regimen (5 x 3 Gy daily fractions; patients typically receive this regimen each week for 4-6 weeks) with and without Tempol treatment and tumor regrowth was assessed. Importantly, Tempol treatment did not protect the tumor from IR, while it can protect normal SG tissue. The mechanism for this appears related to a differential reduction of Tempol in normal versus tumor tissue. ? ? Our past studies in rats and mice clearly have shown that SGs are a potentially useful target site for treating SSPDDs, particularly when the transgene encodes a constitutive pathway secretory protein, which is readily secreted into the bloodstream. During this year, we extended these studies to non-human primates. This is a critical step in developing a gene therapy. Specifically, we compared the efficacy of two widely used adenoassociated virus (AAV) serotypes (2 and 5) in rhesus parotid glands. Both the AAV2 and AAV5 vectors expressed rhesus erythropoietin (RhEPO) under the control of the Rous Sarcoma Virus promoter. A total of four groups (n=3/dosage group) of macaques received either 10e10 (low dose) or 3x10e11 (high dose) particles of either rAAV2RhEPO or rAAV5RhEPO via intraoral cannulation of the right parotid gland. The higher dose used was comparable on a particle per kg basis to that used in our previous murine experiments. Prior to vector administration, serum RhEPO levels were minimal (0.41 0.2 mU/ml; 6 weeks prior to vector delivery for all 12 macaques used) with no animal measuring >1.6 mU/ml, and salivary RhEPO levels were below the lowest standard used (0.8 mU/ml). Serum RhEPO levels increased to 3.36?1.2 and 3.8?0.75 mU/ml in the high dose groups for both AAV2 and AAV5 administered animals, respectively, at week 8. Salivary RhEPO levels increased similarly. The kinetics of transgene expression were different between the two vector serotypes, and were not predicted by the kinetics observed in mice. RhEPO produced in the targeted gland was preferentially secreted into the bloodstream. However, Hct values did not increase appreciably (maximum 5% compared to pre-administration levels), probably due to the relatively low overall levels of RhEPO achieved and tighter control of hematocrit (Hct) in macaques versus mice. The immune response (as measured by the presence of circulating anti-AAV antibody titers) was higher after AAV5 than after AAV2 vector administration. ? ? We also continued to use the female non-obese diabetic (NOD) mouse as a model for the autoimmune sialadenitis occurring in SS. A big advantage of this model is that it is commercially available, and we have used this model extensively to study the efficacy of local AAV2 vector-mediated transfer of immunomodulatory transgenes to alter the SS-like disease. However, our data collected over the past two-year period indicated a changing SS phenotype in NOD mice purchased from Jackson Laboratories. Data from six different experimental studies over two years were retrospectively analyzed and compared. Salivary flow rate, focus score, and SMG cytokines IL-2, IL-4, IL-6, IL-10, IL-12 (p70), TNF-alpha and IFN-gamma all showed significant change over time. Based on these findings we stopped purchasing NOD mice commercially and now secure animals through a university-based collaboration.? ? We have long worked to develop an artificial SG for use in patients without any salivary epithelial cells present and who are not suitable for gene transfer approaches. During the past year we continued our focus on establishing an autologous, polarized, non-human primate epithelial cell system to use in the proposed device and facilitate in vivo functional testing. We obtained three major findings this year. First, primary cultures of rhesus parotid gland (RPG) cells attained a polarized-orientation, with Na/K-ATPase, ZO-1, and claudin-1 distributed in appropriate domains. Second, RPG cells exhibited two essential epithelial functions required for an artificial SG, i.e., they provide an effective barrier to paracellular water flow and they generate a moderate transepithelial electrical resistance. Third, RPG cells can express functional water channels, capable of mediating unidirectional transcellular fluid movement, after transduction by recombinant adenoviral and AAV2 vectors. These results demonstrate that it is feasible to individually prepare RPG cells for eventual use in a prototype artificial SG to test in macaques.
Showing the most recent 10 out of 101 publications
