Review

Salivary gland gene therapy: Personal reflections

Bruce J. Baum n

Tokyo Dental College, Japan

a r t i c l e i n f o

Article history:Received 20 August 2013Received in revised form2 October 2013Accepted 4 October 2013Available online 18 November 2013

Keywords:Oral biologyGene transferXerostomiaHistory

a b s t r a c t

The purpose of this paper is to describe, in somewhat familiar terms, the personal experiences I hadtrying to develop a clinically useful gene therapy for dry mouth over an approximately 20-year period.That research journey, which reached fruition in a recently completed Phase I clinical trial, was nurturedby my long career spent at the US National Institutes of Health and, in particular, by working within itshospital, The Clinical Center, the largest research hospital in the world. Through this paper, I wish totransmit several important lessons that I learned on my journey, which I believe will be applicablebroadly across oral biology.

& 2013 Japanese Association for Oral Biology. Published by Elsevier B.V. All rights reserved.

Contents

Conflict of interest. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41Acknowledgement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

For many years, I have been interested in what I consider thefailure of dental education to recognize the importance of biologyand medicine to dental practice, as well as to embrace both fully inplanning for the future of the dental profession [1,2]. Indeed, Ihave long believed that biology would provide dentistry, and thusdental patients, with the profession's next “fluoride-like” advance[3]. While the latter remains to be achieved, I still find thatdentistry and oral biology, in much of the world, is separatedfrom mainstream biomedical science.

Why has this “artificial” separation of dentistry from medicineand biomedical science been important to me? The answer is that Ithink dentistry would profit enormously from a full reengagementwith both. Unlike most of my colleagues in dentistry and oralbiology, I spent my professional career embedded in mainstreambiomedical science. I was especially fortunate to work fromJanuary 1982–October 2011 at the US National Institute of Dentaland Craniofacial Research (NIDCR), and within the NationalInstitutes of Health (NIH)'s Clinical Center, the largest researchhospital in the world. In that environment it was impossible not to

be stimulated by the endemic level of biological innovations andapplications being translated to the patient's bedside. As a youngdental scientist I was most impressed by leading physician–scientists at the Clinical Center who broke from traditional clinicalapproaches, i.e., who thought “outside the box” of convention, andused biology to develop new ways to help their patients.

I hope through this paper that I can transmit some of thelessons I learned, and applied to my work in gene therapy, to thereadership of the Journal of Oral Biosciences. I am particularlygrateful to the Journal, as well as its Editor, Prof. Hayato Oshima,and the Japanese Association for Oral Biology, for the opportunityto do this, because I believe the general lessons I transmit will bewidely applicable, or at least generally useful to consider, fortranslational investigators in each of the specific scientific dis-ciplines encompassing oral biology.

The concept of transferring genes for therapeutic purposes wasfirst discussed in the 1960s [see comments in 4]. However, it wasthen a technically impossible goal to achieve, i.e., molecularbiology and its abundance of experimentally useful tools did notexist. As a real life example, consider that in 1972 I took a graduatecourse in nucleic acid biochemistry while studying for my PhD. Inthat course there was no mention of restriction endonucleases orreverse transcriptase, because both were only just beginning to be

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/job

Journal of Oral Biosciences

1349-0079/$ - see front matter & 2013 Japanese Association for Oral Biology. Published by Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.job.2013.10.001

n Correspondence author.E-mail address: bjbaum@mac.com

Journal of Oral Biosciences 56 (2014) 38–42

studied, while the concept of the polymerase chain reaction wascompletely unknown. By 1990, armed with such tools, genetherapy had become a reality, in humans [5], after being demon-strated in various animal disease models [e.g., 6–9].

In the late 1980s, in addition to my laboratory studies onsalivary gland signal transduction, I was involved in clinical studiesfocused on the salivary gland hypofunction (dry mouth) thatresults from therapeutic radiation for head and neck cancers andwhich occurs as a sequela of Sjögren's syndrome. While mycolleagues, most notably Philip Fox, and I had experienced somesuccess in developing a pharmacological treatment for dry mouth,i.e., the per os use of pilocarpine [10,11], such a treatment was onlyuseful for patients with a sufficient amount of functional salivaryacinar (fluid secreting) tissue remaining. This represented perhapsone-third, or less, of patients with either condition. As a result, Iincreasingly grew frustrated at my inability to offer our researchpatients any relief or, more important personally, any experimen-tal ideas that might eventually lead to their future relief. Conse-quently, I searched the biomedical science literature for possiblenovel approaches for the treatment of dry mouth. I became awareof the potential of gene therapy, i.e., the transfer of an oligonu-cleotide (at the time either a true gene or a cDNA) for therapeuticpurposes, in particular because of the research then being done onusing gene transfer to the lung to correct the principle pulmonarydefect in cystic fibrosis [12,13]. As I knew from a past post-doctoralfellowship, the lungs are an epithelial tissue with many biologicalsimilarities to salivary glands [14,15]. In 1990, essentially allproposed uses of clinical gene therapy were either for single genemutations, e.g., in-born errors such as cystic fibrosis, for whichthere was no existing treatment, or for cancers that were refrac-tory to conventional therapy, such as malignant melanoma [16].However, I reasoned that the idea of using gene therapy was aworthwhile consideration “even for a quality of life” disorder suchas dry mouth, since there were so many patients who were unableto benefit from the available pharmacological therapies. Accord-ingly, I began to plan a strategy for accomplishing it [17].

One major personal difficulty was that I had no training ineither molecular biology or virology, and I recognized thatelements of both disciplines were likely going to be essential todeveloping a gene therapy for dry mouth. To overcome thishurdle, I did four things, each of which proved to be instrumentalin achieving the end result. First, I took a 4-day course inmolecular biology, so that I at least would become more familiarwith its language and techniques. Secondly, I took a 6-monthsabbatical in the laboratory of a good friend at NIH, so I couldpractice those techniques and become comfortable with theirapplications. Next, I sought a major collaborator with bona fideexpertise in gene therapy, and I was very lucky, because one suchperson, Dr. Ronald G. Crystal, then at the National Heart, Lung andBlood Institute at NIH, also had been one of my post-doctoralmentors [14,15]. Finally, I was successful in recruiting a verytalented, dentist/then new PhD (Brian O’Connell; now a professorin the School of Dental Science at Trinity College Dublin) to jointhis as yet unproven endeavor as a post-doctoral fellow. The toolsof molecular biology were part of his PhD thesis and his con-tributions while in our department were absolutely critical todemonstrating the feasibility of gene transfer to salivary glands[18]. His entire first year at NIH was spent in Dr. Crystal’slaboratory and, thereafter, he brought all of the required mole-cular biology and virology tools back to our department. Indeed,his initial success allowed me to recruit another new PhD post-doctoral fellow (Christine Delporte; now a professor in themedical school of the Free University of Brussels). She specificallybegan testing the gene therapy strategy that I had formulated forradiation (IR)-induced dry mouth, as well as demonstrated proofof concept in irradiated rats [19].

There were several key questions that I thought were necessaryto answer in the development of a salivary gland gene therapy fora dry mouth: (i) who were the most appropriate patients to treat;(ii) what would be the most efficient method to target the gene tothe greatest number of epithelial cells in a damaged salivary gland;(iii) how would a gene be delivered; and (iv), most importantly,what gene would be used? The answers to those questions cameas a result my past clinical experience, some reasonable hypothe-sizing, and a good deal of luck. The answer to question (i), asimplied above, was to focus on patients experiencing IR-induceddry mouth. It actually was a simple decision for, unlike patientswith Sjögren's syndrome, who experience a chronic disease withthe continued lymphocytic infiltration of their salivary glands,once an IR regimen is completed patients suffer no further insultto their gland tissue. The answer to question (ii) also was fairlystraightforward; cannulation of a targeted parotid gland andretrograde infusion of the gene suspended in an appropriatebuffer, much the same as would be done for a contrast X-ray ofthe gland (sialogram). The answer to question (iii) followed frommy reading of the existing gene therapy literature. There were twobasic means to transfer a gene, with or without a viral vector, andthe literature at that time was clear that use of a viral vector wasmarkedly more efficient at gene transfer than non-viral methods.The answer to question (iv) also came from the literature, and itstiming was pure good fortune. As I was putting the gene therapystrategy together, I was at a loss over what gene could be used; Iwas unaware of any existing gene with the therapeutic potentialfor “repairing” a dry mouth. In late 1991, a good friend andcolleague told me that Preston and Agre has just published apaper in the Proceedings of the National Academy of Science USA

Fig. 1. Schematic diagram of hypothesized process by which AdhAQP1 facilitatesfluid secretion from irradiated salivary glands. Surviving duct cells are presented ina simplified form, with only ion channels and ion transporters depicted (top). Thelumen is to the left of the cell shown, and the interstitium is to the right. The cell isshown as water impermeable. After transduction of this cell with AdhAQP1(bottom), the water channel aquaporin-1 is inserted into the apical and basalmembranes providing a pathway by which water can flow in response to anosmotic gradient. We have hypothesized that this gradient would be generated bymovement of Kþ and HCO3

� into the lumen, i.e. lumen4 interstitium. The figure isbased on the experiments presented in Delporte et al. [19]. .The figure and legend are reprinted with permission from [21]

B.J. Baum / Journal of Oral Biosciences 56 (2014) 38–42 39

Fig. 2. Summary of clinical response data. Clinical responses following vector delivery as measured by (a) absolute parotid salivary flow rate from the targeted gland and (b) theproportional increase in peak parotid salivary flow shown as the percent of baseline. Statistical significance was determined using the Wilcoxon matched pair rank test for the changein absolute values. The Wilcoxon signed rank test was used to test if the peak proportional increase in parotid salivary flow was significantly different from the baseline (100%).Individual changes in parotid salivary flow are shown in (c) for absolute salivary flow rates and in (d) for proportional changes compared to baseline. Coding for individual subjects isshown as indicated in the panel (c) insert. All subjects shown in black were considered non-responders (o50% increase in salivary flow rate). All subjects shown in colors wereconsidered responders (at least 50% increase in parotid salivary flow rate following AdhAQP1 administration). The days indicated to the right of each peak data point correspond to thedays onwhich that peak parotid flow rate was observed. Visual analogue scale (VAS) results from all subjects, at baseline and peak time of parotid salivary flow, are shown for both theamount of saliva perceived (e; rate how much saliva is in your mouth) and dryness of their mouth (f; rate the dryness in your mouth). Note that lower VAS results indicate animprovement in symptoms. The colors and symbols used to identify individual subjects are identical to those shown in panel c. (a) Parotid salivary flow rate from the targeted gland,(b) proportional increase in peak salivary flow, (c) absolute change in saliva flow, (d) proportional change in saliva flow, (e) amount saliva and (f) dry mouth. (For interpretation of thereferences to color in this figure legend, the reader is referred to the web version of this article.)Reprinted with permission from [30]

B.J. Baum / Journal of Oral Biosciences 56 (2014) 38–4240

[20] reporting a gene, then called CHIP 28 and later Aquaporin-1(AQP1), which could fit into my therapeutic scheme. That paperwas the cornerstone of what would become Peter Agre’s Nobelprize-winning work on the discovery of water channels.

The goal of the therapeutic strategy became delivery of thehuman (h) AQP1 cDNA to surviving epithelial cells in an irradiatedsalivary gland [17,19]. The logic was as follows. IR was believed tolead to the loss of most acinar cells from salivary glands, and thatthe overwhelming majority of surviving salivary epithelial cellswould be non-fluid secreting cells, which were ductal in origin. Asdescribed previously [21–23], I reasoned, with the help of my longtime colleague James Turner, that duct cells could, in the absenceof any significant acinar cell primary fluid secretion, generate anosmotic gradient (lumen4 interstitium), and the only thing pre-venting duct cells from secreting fluid was the absence in thesecells of a facilitated water permeability pathway, i.e., a waterchannel (Fig. 1). While this hypothesis has still not been unequi-vocally proven, after developing reasonable expertise in transfer-ring genes in healthy rats [24,25] my colleagues and I began to testit in vitro and then in an irradiated rat model [19]. The latter studyshowed proof of concept, i.e., feasibility. The vector we employedwas a first generation recombinant adenoviral vector encodinghAQP1 (AdhAQP1), which was constructed by Dr. Delporte, and itsdelivery to irradiated rat submandibular glands led to near normalsalivary fluid secretion in rats four months following IR [19].

We next took a two-pronged approach toward translating thispotential therapy into a real clinical trial: (i) determining if thegene therapy could scale-up to an animal significantly larger thana rat and (ii) evaluating the safety of AdhAQP1 delivery to thesalivary glands, i.e., toxicology and biodistribution studies. Initially,we conducted a small study in non-human primates [26], but thevector’s efficacy was equivocal because of the small number ofanimals involved (n¼2/dose with 2 doses, with only 1 controlanimal); macaques were very expensive to study. We then tried asimilar experiment using a different animal model, the miniaturepig, which had a more reasonable cost, i.e., we could use a muchlarger number of animals. These studies were done in collabora-tion with a former post-doctoral fellow in our department, SonglinWang, now a professor at Capital Medical University in Beijing[27,28]. Impressively, AdhAQP1 was able to essentially restoresalivary fluid flow in IR-damaged miniature pig parotid glands(“n”¼18 animals) and, as it happened, at a lower dose of AdhAQP1than required in the rat [28]. The second prong, the safety study,was conducted in collaboration with colleagues (Rick Irwin andMolly Vallant) at the National Toxicology Program of the NIH’sNational Institute of Environmental Health Sciences. Those studiesshowed that the AdhAQP1 vector resulted in no significant adverseeffects in rats [29]. Based on the findings from these two keystudies, we decided to submit a clinical protocol to test thestrategy in humans.

The clinical protocol was submitted simultaneously for reviewto the NIDCR's Institutional Review Board (protocol 06-D-0206)and to the US Recombinant DNA Advisory Committee in Novemberof 2005. In addition to approval from both of those committees,the protocol received subsequent approvals from the NIH BiosafetyCommittee, the US Food and Drug Administration (IND BB-13,102)and an independent Data Safety and Monitoring Board, whichoversaw the conduct of the entire study. All approvals werereceived by February 2007 and, after establishing the necessaryinfrastructure (e.g., database, electronic case report forms, mon-itoring policies, etc), the first patient was enrolled and treated inthe summer of 2008.

While the study had been approved for 15 subjects, only 11could be treated prior to the expiration date of the vector [30]. Myassessment of the time required to establish the infrastructure,unfortunately, was unrealistically short. Coupled with the normal

difficulties associated with conducting a “first in human” study, wedid not have sufficient time to treat 15 subjects before the clinicalgrade AdhAQP1 vector “expired”. Since this was designed as adose-escalation study (n¼3/dose group), we were unable to treatpatients at the highest dose. However, this turned out to befortuitous. Of the 11 subjects treated, five experienced both aquantitative increase in parotid saliva output in the targeted glandand an improvement in their xerostomic symptoms (Fig. 2) [30].These five subjects were treated either with 4.8�107 (n¼1),2.9�108 (n¼2) or 1.3�109 (n¼2) AdhAQP1 vector particles (vp)to a single parotid gland. Of the six subjects who did not benefitfrom the gene therapy maneuver, three (one at 1.3�109 vp andtwo at 5.8�109 vp) showed evidence of a significant inflamma-tory response to AdhAQP1, i.e., the higher approved dose likelywould have led to additional inflammation. One subject had alatent adenoviral serotype 5 infection in the targeted gland,leading to targeted cell lysis but no adverse effects [31], andanother had a highly unusual pattern of 99mTcO4 uptake in thetargeted gland prior to treatment [30], indicative of major func-tional damage to the surviving epithelial cells. We still do notunderstand why the therapy was ineffective in the last of these sixnon-responding subjects. Presently, my former NIDCR colleaguesare planning a new clinical trial using the same hAQP1 cDNA, butemploying a much less inflammatory viral vector, based on theserotype 2 adeno-associated virus [32,33], for targeted genedelivery.

I began this paper by stating that I hoped my experiences indeveloping a clinical gene therapy strategy would be helpful to abroad range of oral biologists. So what are the general lessons thatI want this narrative to transmit? To me there are six lessons andthey are as follows: (i) read extensively outside of oral biology tolearn what is happening in the wider biomedical research com-munity that may be potentially applicable to your work; (ii) do notbe afraid to have a good imagination and “see” possibilities thatothers may not recognize; (iii) do not be afraid to changedirections in your research, even in the middle of a reasonablysuccessful career, as I was experiencing in 1991; (iv) do nothesitate to seek expert help when you try something new, as itis better to test ideas out quickly and efficiently versus stumblingalong trying to work things out on your own; (v) try to think longrange, but have definite short term goals that can be achievableand, thus, encouraging to you and your colleagues; and, finally, (vi)follow the wise advice of the French writer Voltaire, “don’t let theperfect be the enemy of the good” (originally, “le mieux estl’ennemi du bien”). Early on in this endeavor I recall most peopleoffering criticisms about one thing or another in my strategybecause it was imperfect, not “ideal”, e.g., using an early adeno-viral vector, which was known to elicit a significant inflammatoryresponse. In such situations Voltaire's advice was invaluable, sincefor patients who have no other options, it is reasonable to trysomething that may be good and helpful, as long as it is shown tobe safe, versus waiting until perfection comes along.

Conflict of interest

The author declares no conflicts of interest.

Acknowledgement

None of the above studies would have been possible if leftalone to me. I am most appreciative to my many wonderful andtalented colleagues and collaborators with whom I had theconsiderable good fortune to work throughout my career, to theIntramural Research Program of the National Institute of Dental

B.J. Baum / Journal of Oral Biosciences 56 (2014) 38–42 41

and Craniofacial Research, which supported my studies from 1982to 2011, and to the eleven research subjects, who participated inthe first in human, AdhAQP1 salivary gland gene therapy study.

References

[1] Baum BJ. Has modern biology entered the mouth? The clinical impact ofbiological research J Dent Educ 1991;55:299–303.

[2] Baum BJ. Inadequate training in the biological sciences and medicine fordental students: an impending crisis for dentistry. J Am Dent Assoc2007;138:16–20.

[3] Baum BJ. Crowning achievements in dentistry. Lancet 1999;354(Suppl: SIV12).[4] Cotrim AP, Baum BJ. Gene therapy: some history, applications, problems and

prospects. Toxicol Pathol 2008;36:97–103.[5] Rosenberg SA, Aebersold P, Cornetta K, Kasid A, Morgan RA, Moen R, et al.

Gene transfer into humans – immunotherapy of patients with advancedmelanoma, using tumor-infiltrating lymphocytes modified by retroviral genetransduction. N Engl J Med 1990;323:570–8.

[6] Anderson WF, Goldberg S, Kantroff P, Berg P, Eglitis M, Humphries RK.Attempts at gene therapy in β-thalassemic mice. Ann NY Acad Sci1985;445:445–51.

[7] Stratford-Perricaudet LD, Levrero M, Chasse JF, Perricaudet M, Briand P.Evaluation of the transfer and expression in mice of an enzyme encodinggene using a human adenovirus vector. Hum Gene Ther 1990;1:241–56.

[8] Culver KW, Morgan RA, Osborne WR, Lee RT, Lenschow D, Able C, Cornetta K,Anderson WF, Blaese RM. In vivo expression and survival of gene-modified Tlymphocytes in rhesus monkeys. Hum Gene Ther 1990;1:399–410.

[9] Wilson JM, Chowdhury NR, Grossman M, Wajsman R, Epstein A, Mulligan RC,Chowdhury JR. Temporary amelioration of hyperlipidemia in low densitylipoprotein receptor-deficient rabbits transplanted with genetically modifiedhepatocytes. Proc Natl Acad Sci (USA) 1990;87:8437–41.

[10] Fox PC, van der Van PF, Baum BJ, Mandel ID. Pilocarpine for the treatment ofxerostomia associated with salivary gland dysfunction. Oral Surg Oral MedOral Pathol 1986;61:243–8.

[11] Fox PC, Atkinson JC, Macynski AA, Wolff A, Kung DS, Valdez IH, Jackson W,Delapenha RA, Shiroky J, Baum BJ. Pilocarpine treatment of salivary glandhypofunction and dry mouth (xerostomia). Arch Int Med 1990;151:1149–52.

[12] Whitsett JA, Dey CR, Stripp BR, Wikenheiser KA, Clark JC, Wert SE, Gregory RJ,Smith AE, Cohn JA, Wilson JM, Engelhardt J. Human cystic fibrosis transmem-brane conductance regulator directed to respiratory epithelial cells in trans-genic mice. Nat Genet 1992;2:13–20.

[13] Yoshimura K, Rosenfeld MA, Nakamura H, Scherer EM, Pavirani A, Lecocq JP,Crystal RG. Expression of the human cystic fibrosis transmembrane conduc-tance regulator gene in the mouse lung after in vivo intratracheal plasmid-mediated gene transfer. Nucl Acids Res 1992;20:3233–40.

[14] Baum BJ, Moss J, Breul SD, Crystal RG. Association in normal human fibroblastsof newly elevated levels of adenosine 3′,5′-monophosphate with a selectivedecrease in collagen production. J Biol Chem 1978;253:3391–4.

[15] Bienkowski RS, Baum BJ, Crystal RG. Fibroblasts degrade newly synthesizedcollagen within the cell before secretion. Nature 1978;276:413–6.

[16] Anderson WF. Prospects for human gene therapy. Science 1990;226:401–9.[17] Baum BJ, O’Connell BC. The impact of gene therapy on dentistry. J Am Dent

Assoc 1995;126:179–89.

[18] Mastrangeli A, O'Connell BC, Aladib W, Fox PC, Baum BJ, Crystal RG. Directin vivo adenovirus-mediated gene transfer to salivary glands. Am J Physiol1994;266:G1146–55.

[19] Delporte C, O’Connell BC, He X, Lancaster HE, O’Connell AC, Agre P, Baum BJ.Increased fluid secretion after adenoviral-mediated transfer of the aquaporin-1 cDNA to irradiated rat salivary glands. Proc Natl Acad Sci (USA)1997;94:3268–73.

[20] Preston G, Agre P. Isolation of the cDNA for erythrocyte integral membraneprotein of 28 kilodaltons: member of an ancient channel family. Proc NatlAcad Sci (USA) 1991;88:11110–4.

[21] Vitolo JM, Baum BJ. The use of gene transfer for the protection and repair ofsalivary glands. Oral Dis 2002;8:183–91.

[22] Baum BJ, Zheng C, Alevizos I, Cotrim AP, Kiu S, McCullagh L, Goldsmith CM,McDermott N, Chiorini JA, Nikolov NP, Illei GG. Development of a genetransfer-based treatment for radiation-induced salivary hypofunction. OralOncol 2010;46:4–8.

[23] Samuni Y, Baum BJ. Gene delivery in salivary glands: from the bench to theclinic. Biochem Biophys Acta 2011;1812:1515–21.

[24] Adesanya MM, Redman RS, Baum BJ, O’Connell BC. Immediate inflammatoryresponses to adenovirus-mediated gene transfer in rat salivary glands. HumGene Ther 1996;7:1085–93.

[25] Kagami H, O’Connell BC, Baum BJ. Evidence for the systemic delivery of atransgene product from salivary glands. Hum Gene Ther 1996;7:2177–84.

[26] O’Connell AC, Baccaglini L, Fox PC, O’Connell BC, Kenshalo D, Oweisy H, HoqueAT, Sun D, Herscher LL, Braddon VR, Delporte C, Baum BJ. Safety and efficacy ofadenovirus-mediated transfer of the human aquaporin-1 cDNA to irradiatedparotid glands of non-human primates. Cancer Gene Ther 1999;6:505–13.

[27] Li J, Zheng C, Zhang X, Liu X, Zhang C, Goldsmith CM, Baum BJ, Wang S.Developing a convenient large animal model for gene transfer to salivaryglands in vivo. J Gene Med 2004;6:55–63.

[28] Shan Z, Li J, Zheng C, Liu X, Fan Z, Zhang C, Goldsmith CM, Wellner RB, BaumBJ, Wang S. Increased fluid secretion after adenoviral-mediated transfer of thehuman aquaporin-1 cDNA to irradiated miniature pig parotid glands. Mol Ther2005;11:444–51.

[29] Zheng C, Goldsmith CM, Mineshiba F, Chiorini JA, Kerr A, Wenk ML, Vallant M,Irwin RD, Baum BJ. Toxicity and biodistribution of a first generation recombi-nant adenoviral vector, encoding aquaporin-1, after retroductal delivery to asingle rat submandibular gland. Hum Gene Ther 2006;17:1122–33.

[30] Baum BJ, Alevizos A, Zheng C, Cotrim AP, Liu S, McCullagh L, Goldsmith CM,Burbelo PD, Citrin DE, Mitchell JB, Nottingham LK, Rudy SF, Van Waes C,Whatley MA, Brahim JS, Chiorini JA, Danielides S, Turner RJ, Patronas NJ, ChenCC, Nikolov NP, Illei GG. Early responses to adenoviral-mediated transfer ofthe aquaporin-1 cDNA for radiation-induced salivary hypofunction. Proc NatlAcad Sci (USA) 2012;109:19403–7.

[31] Zheng C, Nikolov NP, Alevizos I, Cotrim AP, Liu S, McCullagh L, Chiorini JA, IlleiGG, Baum BJ. Transient detection of E1-containing adenovirus in saliva afterdelivery of a first generation adenoviral vector to human parotid gland. J GeneMed 2010;12:3–10.

[32] Braddon VR, Chiorini JA, Wang S, Kotin RM, Baum BJ. Adenoassociated virus-mediated transfer of a functional water channel into salivary epithelial cellsin vitro and in vivo. Hum Gene Ther 1995;9:2777–85.

[33] Gao R, Yan X, Zheng C, Goldsmith CM, Afione S, Hai B, Xu J, Zhou J, Chang C,Chiorini JA, Baum BJ, Wang S. AAV2-mediated transfer of the humanaquaporin-1 cDNA restores fluid secretion from irradiated miniature pigparotid glands. Gene Ther 2011;18:38–42.

B.J. Baum / Journal of Oral Biosciences 56 (2014) 38–4242