Vaccine 21 (2003) 1640–1649 Modulating gene expression using DNA vaccines with different 3 ′ -UTRs influences antibody titer, seroconversion and cytokine profiles John W. Zinckgraf a,1 , Lawrence K. Silbart b,∗ a Department of Molecular and Cell Biology, University of Connecticut, Storrs, CT 06269, USA b Department of Animal Science, Center of Excellence for Vaccine Research, University of Connecticut, 1390 Storrs Road, Unit 4163, Storrs, CT 06269, USA Received 27 March 2002; received in revised form 23 October 2002; accepted 28 October 2002 Abstract To determine if modulating the amount of foreign antigen produced by a DNA vaccine can influence the overall intensity and cytokine polarization of the ensuing immune response, three different plasmids, each encoding the hepatitis B (HB) surface antigen, were constructed. In each construct, HBs gene expression was driven by the cytomegalovirus immediate early promoter, but differed in the 3 ′ -untranslated regions (3 ′ -UTR) containing the polyadenylation sequence. These 3 ′ -UTR sequences were derived from either the hepatitis B virus (HBVpA), bovine growth hormone (BGHpA), or rabbit -globin (pA). BALB/c mice were immunized intramuscularly with equimolar amounts of each plasmid and blood was collected bi-weekly. Following immunization, total IgG titers correlated with in vitro antigen production levels (from transfected CHO cells), as evidenced by the following response pattern: HBVpA > BGHpA ≫ pA. All groups demonstrated a heavy bias toward a Th1 immune response, as evidenced by high serum IgG2a/IgG1 ratios and the predominance of IFN- over IL-4 secretion from cultured splenocytes. In addition, the HBVpA construct resulted in a seroconversion rate of 100%, in comparison to 40–50% in the BGHpA, and 0% in the pA group. Surprisingly, splenocytes isolated from mice immunized with the pA construct secreted the highest levels of IFN-. Taken together, these findings suggest that altering the level of gene expression not only affects the overall titer and seroconversion rates of vaccinated animals, but also may play a role in modulating cytokine profiles. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: DNA vaccines; Hepatitis B virus; Gene regulation 1. Introduction Over the past decade, genetic (or DNA) vaccines have emerged as a viable alternative to traditional vaccine ap- proaches, such as live-attenuated, subunit, peptide or protein vaccines. DNA vaccines consist of bacterial plasmid DNA that encode antigens, usually under the control of strong viral promoters [1]. The ability of foreign DNA to induce an immune response to the expressed antigen in a mam- malian host was shown by several key studies performed in the early 1990s. In one study, intramuscular (i.m.) injection of mice with naked plasmid reporter constructs resulted in immune responses to the encoded antigens [2]. The feasi- bility of using naked DNA for vaccination purposes was shown in another study in which mice were immunized with a plasmid DNA vaccine encoding the influenza A ∗ Corresponding author. Tel.: +1-860-486-6073; fax: +1-860-486-5067. E-mail address: lsilbart@canr.uconn.edu (L.K. Silbart). 1 Present address: Epimmune Inc., 5820 Nancy Ridge Drive, San Diego, CA 92121, USA. nucleoprotein. Immunized mice were protected against in- fluenza A challenge, as evidenced by decreased viral titers in lungs and prolonged survival [3]. These and other pub- lications gave a better understanding of the mechanisms involved in genetic immunizations and further stimulated research to expand and improve the potential of DNA vaccines. DNA vaccines can be delivered by several different meth- ods, including needle injection, particle bombardment (gene gun), microspheres and liposomes. Following antigen ex- pression, processing and presentation of the antigen occurs in the context of both MHC I and MHC II molecules, thus stimulating cell-mediated and humoral immune responses. The route of immunization can also be used to influence the cytokine profile of the immune response. For instance, i.m. injections of naked plasmid DNA typically yield a Th1-biased immune response, as evidenced by high levels of IL-2, IFN-, IgG2a and strong cell-mediated immunity [4–6]. Conversely, gene gun immunization generally induces a Th2 response, as indicated by the Type 2 cytokine pro- file (IL-4, IL-5, IL-6) and the predominance of IgG1 over 0264-410X/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII:S0264-410X(02)00740-5