Targeting antigen to MHC Class I and Class II antigen presentation pathways for malaria DNA vaccines

Immunology Letters 111 (2007) 92–102

Targeting antigen to MHC Class I and Class II antigen presentation pathways for malaria DNA vaccines Carlota Doba˜no a,b,∗ , William O. Rogers a,1 , Kalpana Gowda a , Denise L. Doolan a,c,2 a

Malaria Program, Naval Medical Research Center, Silver Spring, MD 20910-7500, United States b Henry M. Jackson Foundation, Rockville, MD 20852, United States c Department of Molecular Microbiology and Immunology, School of Hygiene and Public Health, Johns Hopkins University, Baltimore, MD 21205-2179, United States Received 5 March 2007; received in revised form 4 May 2007; accepted 24 May 2007 Available online 18 June 2007

Abstract An effective malaria vaccine which protects against all stages of Plasmodium infection may need to elicit robust CD8+ and CD4+ T cell and antibody responses. To achieve this, we have investigated strategies designed to improve the immunogenicity of DNA vaccines encoding the Plasmodium yoelii pre-erythrocytic stage antigens PyCSP and PyHEP17, by targeting the encoded proteins to the MHC Classes I and II processing and presentation pathways. For enhancement of CD8+ T cell responses, we targeted the antigens for degradation by the ubiquitin (Ub)/proteosome pathway following the N-terminal rule. We constructed plasmids containing PyCSP or PyHEP17 genes fused to the Ub gene: plasmids where the N-terminal antigen residues were mutated from the stabilizing amino acid methionine to destabilizing arginine, plasmids where the C-terminal residues of Ub were mutated from glycine to alanine, and plasmids in which the potential hydrophobic leader sequences of the antigens were deleted. For enhancement of CD4+ T cell and antibody responses, we targeted the antigens for degradation by the endosomal/lysosomal pathway by linking the antigen to the lysosome-associated membrane protein (LAMP). We found that immunization with DNA vaccine encoding PyHEP17 fused to Ub and bearing arginine induced higher IFN-␥, cytotoxic and proliferative T cell responses than unmodified vaccines. However, no effect was seen for PyCSP using the same targeting strategies. Regarding Class II antigen targeting, fusion to LAMP did not enhance antibody responses to either PyHEP17 or PyCSP, and resulted in a marginal increase in lymphoproliferative CD4+ T cell responses. Our data highlight the antigen dependence of immune enhancement strategies that target antigen to the MHC Class I and II pathways for vaccine development. © 2007 Elsevier B.V. All rights reserved. Keywords: Plasmodium yoelii; Malaria; DNA vaccines; PyHEP17; PyCSP; Ubiquitin; LAMP; MHC; Antigen targeting; Antibody responses; T cell responses

1. Introduction The immunogenicity and protective capacity of plasmid DNA vaccines in animal models has been established in a number of systems, including malaria [1–4]. However, these first generation DNA vaccines are not optimal, and considerable efforts are now being directed at immune enhancement strategies [5–7]. ∗

Corresponding author. Current address: Centre de Recerca en Salut Internacional de Barcelona (CRESIB), Hospital Cl´ınic/Institut d’Investigacions Biom`edicas August P´ı i Sunyer (IDIBAPS), Universitat de Barcelona, Barcelona E-08036, Spain. Tel.: +34 932275706; fax: +34 932279853. E-mail address: [email protected] (C. Doba˜no). 1 Current address: Naval Medical Research Unit 2, Jakarta, Indonesia. 2 Current address: Queensland Institute of Medical Research, The Bancroft Centre, 300 Herston Road, PO Royal Brisbane Hospital, Brisbane, QLD 4029, Australia. 0165-2478/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.imlet.2007.05.007

In mice, immunization with DNA vaccines encoding Plasmodium yoelii circumsporozoite protein (PyCSP) and P. yoelii hepatocyte erythrocyte protein 17 kDa (PyHEP17) confers sterile protection against P. yoelii sporozoite challenge [1,2]. This protection is dependent on IFN-␥ and CD8+ T cells [1,2] and therefore requires efficient processing and presentation of the DNA-encoded antigen via the MHC Class I pathway [8]. The generation of minimal CD8+ T cell epitopes mostly depends on the degradation of the target antigen by the proteosome complex in the cytoplasm of the antigen-presenting cell (APC) [9]. Attachment of ubiquitin (Ub) [10] to lysine side chains of a protein has been reported to target that protein for rapid cytoplasmic degradation [11,12]. According to the N-end rule [13,14], proteins bearing destabilizing amino acids, such as arginine (R), at the amino terminus are rapidly ubiquitinated and degraded by the proteosome [15,16].

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

In addition to efficient antigen processing and presentation via the MHC Class I pathway, effective CD8+ T cell responses are dependent upon the help provided by CD4+ T cells [17,18]. Effective CD4+ T cell help is also required for the induction of high titer specific IgG antibodies, which in the malaria rodent model are associated with protection against blood stage parasitemia [19] and in humans are major contributors to the development of naturally acquired immunity [20]. Nevertheless, current generation DNA vaccines have failed to induce antibodies in the clinic [21–24]. The induction of CD4+ T cell effector responses requires most of the times proteolytic degradation of the antigen via the endosomal/lysosomal processing compartment and presentation in the cell surface by MHC Class II molecules. A family of lysosome-associated membrane proteins (LAMP) located in lysosomes and late endosomes and which contain a C-terminal YQTI sequence has been shown to target proteins through a vesicular pathway to lysosomes [25]. An optimal vaccine to protect against all stages of Plasmodium infection would ideally elicit CD8+ and CD4+ T cell and antibody responses and thus need to efficiently deliver antigen to both the MHC Class I and Class II processing and presentation pathways. Accordingly, this study investigates strategies for targeting the P. yoelii antigens encoded by DNA vaccines to MHC Class I or II pathways in an attempt to improve their immunogenicity and protective efficacy. To target the encoded antigen to the MHC Class I pathway we took advantage of the fact that the carboxy terminus of Ub is a substrate for a site-specific protease [26], so that linear fusions of Ub with a protein of interest are cleaved just distal to the last amino acid of Ub. This exposes the first amino acid of the fusion partner, making it possible to engineer proteins which will bear either stabilizing or destabilizing amino acids at the N-terminus following cleavage from the Ub fusion [11,15]. We speculated that ubiquitination should increase the availability of antigenic peptides for binding to MHC Class I molecules and that subsequent recognition by CD8+ T cells should be reflected in the magnitude of Ag-specific CD8+ T cellmediated immune responses. This strategy has been successful at targeting viral proteins for cytoplasmatic degradation and increased CTL responses in some systems: vaccinia-expressed influenza hemagglutinin [27], HIV envelope protein [28], or HIV nef protein [29]; but not in others: influenza nucleoprotein [30] or Hep C virus core protein [31], expressed in DNA vaccines. Another related strategy has been to take advantage of the fact that alanine (A) at the C-terminus of Ub prevents the cleavage of the Ub-specific protease and provides a substrate for polyubiquitination [32]. Thus, substitution of glycine by A at the C-terminus of Ub fused to a DNA vaccineencoded protein can result in increased degradation, CTL and protection, as observed in the LCMV nucleoprotein system [33,34]. To target the DNA-encoded Ag to the MHC Class II pathway, we speculated that fusion of the gene of interest to LAMP would target the encoded Ag for processing through the endosomal pathway and presentation via Class II molecules to CD4+ T cells. This approach has resulted in enhanced MHC Class II mediated

93

responses in a number of systems such as HPV-16 E7 [35], HIV envelope protein [36], Epstein–Barr and influenza viruses [37] and CMV protein pp65 [38] but not in the case of HCV core protein [31]. Accordingly, herein, we evaluated the potential of these Ag targeting strategies for the enhancement of vaccine-induced immunity to malaria. Specifically, we compared the immunogenicity induced by plasmid DNA vaccines encoding the pre-erythrocytic antigens PyCSP and PyHEP17 fused to Ub in accordance with the N-end rule, with or without their signal sequence, or fused to LAMP, to responses induced by immunization with unmodified DNA vaccines. 2. Materials and methods 2.1. Plasmid DNA vaccines Unmodified plasmid DNA vaccines, based on the VR1012 backbone [39], were obtained from Vical Inc. (San Diego, CA): VR2513 encoded PyHEP17 [2], and VR2507 encoded PyCSP [1]. 2.2. Cloning of Ub and LAMP fusion constructs A series of plasmid DNA vaccines encoding PyCSP and PyHEP17, with or without their native signal sequences, fused to the Ub or the LAMP mouse genes, were constructed. Ub fusion constructs were designed to bear either stabilizing or destabilizing amino acids at the amino terminus. This was achieved by mutating the codons encoding the first residue of the PyCSP or PyHEP17 antigen from the stabilizing amino acid M to the destabilizing R. In addition, in plasmids designed to lack the potential hydrophobic leader sequence of the antigen (signal sequence, –ss), the first amino acid distal to the Ub fusion was designed to be either M or R. Another targeting approach, not based on the N-end rule, was to fuse the malaria proteins to a mutant form of Ub with A instead of glycine in the last position (C-terminal). This mutation links Ub to the protein by a covalent bond which is relatively resistant to cleavage, thus targeting the stably ubiquitinated protein into the polyubiquitination pathway and hence to the proteasome. The mouse Ub gene, encoding a 76 amino acid protein highly conserved among eukaryotes [40], was amplified by PCR from genomic DNA. The Ub monomer was gel purified and cloned into pCR-script plasmid (Stratagene, La Jolla, CA). Recombinant PCR, using a series of four primers per reaction, was used to fuse the Ub gene in-frame to the P. yoelii genes PyCSP and PyHEP17. Amino acid modifications were introduced with a set of PCR oligonucleotides including the appropriate nucleotides at the junction of the two chimeric genes. Alternatively, site-directed mutagenesis of the relevant amino acids was carried out using the QuikChangeTM kit (Stratagene, La Jolla, CA) as described by the manufacturer. PCR products were digested with BamHI, purified by GeneClean II (Bio 101, Q-BIOgene, Morgan Irvine, CA), and ligated to linear dephosphorylated VR1012 plasmid at the BamHI restriction site. In some instances, PCR products were subcloned into pCR-script plasmid prior to cloning into VR1012. DH10␤ Escherichia coli

94

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

Table 1 DNA vaccine constructs PyCSP constructs

PyHEP17 construct

Description

VR2507 Ub-M-PyCSP/VR1012a Ub-R-PyCSP/VR1012 Ub-A-PyCSP/VR1012 Ub-M-PyCSP-ss/VR1012 Ub-R-PyCSP-ss/VR1012 PyCSP–LAMP/VR1012

VR2513 Ub-M-PyHEP17/VR1012 Ub-R-PyHEP17/VR1012 Ub-A-PyHEP17/VR1012 Ub-M-PyHEP17-ss/VR1012 Ub-R-PyHEP17-ss/VR1012 PyHEP17–LAMP/VR1012

DNA plasmid containing native antigens Ubiquitin fusion bearing stabilizing Met at the N-terminal of the P. yoelii antigen Ubiquitin fusion bearing destabilizing Arg at the N-terminal of the P. yoelii antigen Ubiquitin fusion bearing Ala at the C-terminal of the ubiquitin protein Ubiquitin fusion bearing stabilizing Met without antigen signal sequence Ubiquitin fusion bearing destabilizing Arg without antigen signal sequence LAMP fusion

a VR1012 is the empty DNA plasmid vector in which antigen genes and ubiquitin or LAMP fusions were cloned; it also served as negative control in immunization experiments.

cells were transformed with the ligated products by electroporation, and positive clones were selected in LB-Kan and screened by PCR using original 5 and 3 primers. Recombinant plasmids were purified by Minipreps (QIAGEN Inc., Chatsworth, CA). Additionally, a plasmid DNA vaccine vector that allowed in frame insertion of PyCSP and PyHEP17 as amino terminal fusions to the lysosome targeting signal from LAMP was constructed. These plasmids were prepared and purified as described above. The various DNA plasmid constructs used in this study are detailed in Table 1. 2.3. Characterization of recombinant plasmid DNA constructs Each recombinant plasmid was analyzed by digestion with restriction enzymes BamHI, BglII and DraI to determine the correct orientation of the insert. The DNA sequence of each construct was determined from both strands using the ABI PRISMTM Ready Reaction Dye DeoxyTM Terminator Cycle Sequencing kit (Perkin Elmer Corporation, Norwalk, CT), under conditions described by the manufacturer. Sequencing primers mapped to the vector DNA sequence adjacent to the polycloning site. Large-scale preparations of the DNA constructs were purified by standard alkaline lysis followed by cesium chloride gradient centrifugation [41]. Expression of the encoded gene was confirmed in vitro by transient transfection of UM449 cells (American Type Culture Collection, Rockville, MD or Vical Inc., San Diego, CA) with 2.5 ␮g of QIAGEN purified plasmid DNA and 10 ␮l lipofectin (Gibco BRL), as described by the manufacturer. Subsequent analysis of the UM449 cell lysate was done by Western blot using the PyCSP specific MAb NYS1 [42] or the PyHEP17-specific mAb NYLS3 [43]. 2.4. Mice and DNA vaccine immunizations Female 5–8-week-old BALB/cByJ (H-2/d) mice were obtained from The Jackson Laboratory, Bar Harbor, ME. Mice were immunized three times at 3 weeks intervals intramuscularly in each tibialis anterior muscle with 5 or 50 ␮g of each plasmid DNA construct in a total volume of 50 ␮l saline, or unmodified VR1012 empty plasmid. Two or three weeks after the third immunization, mice were sacrificed for T cell studies (six mice/group). The experiments reported herein were

conducted in compliance with the Animal Welfare Act and in accordance with the principles set forth in the “Guide for the Care and Use of Laboratory Animals”, Institute of Laboratory Animal Resources, National Research Council, National Academy Press, 1996. 2.5. Synthetic peptides Synthetic peptides based on PyCSP and PyHEP17 amino acid sequences used for in vitro stimulation in T cell assays or as capture antigens in enzyme-linked immunosorbent assay (ELISA) were synthesized commercially (Research Genetics, Huntsville, AL) at >95% purity (Table 2). Different sets of peptides representing CD4+ and/or CD8+ T cell epitopes were used in the different T cell assays, depending on their capacity to induce cytokine, proliferative or cytotoxic responses, as previously characterized in our laboratory (see figures below). Table 2 Synthetic peptide sequences PyCSP peptides [53–56] CD4+ T helper cell epitopes with overlapping CD8+ T cell epitopes Residues 280–296 SYVPSAEQILEFVKQI Dominant Residues 57–70 KIYNRNIVNRLLGD Subdominant Residues 59–79 YNRNIVNRLLGDALNGKPEEK Subdominant CD8+ T cell epitopes Residues 280–288 Residues 58–67

SYVPSAEQI IYNRNIVNRL

Dominant Subdominant

PyHEP17 peptides [43,57] CD4+ T cell epitope with nested CD8+ T cell epitopes (15-mer) Residues 26–40 KYGKNGKYGSQNVIK Subdominant Residues 61–75 EEIVKLTKNKKSLRK Dominant Residues 66–80 LTKNKKSLRKINVAL Subdominant Residues 71–85 KSLRKINVALATAL Dominant Residues 126–140 SFPMNEESPLGFSPE Subdominant CD8+ T cell epitopes (9-mer) Residues 73–81 LRKINVALA Residues 74–82 RKINVALAT Residues 61–69 EEIVKLTKN Residues 70–78 KKSLRKINV Residues 76–84 INVALATAL Residues 84–92 LSVVSAILL

Subdominant Subdominant Subdominant Subdominant Subdominant Subdominant

B cell epitope (peptide MR68) Residues 126–140 SFPMNEESPLGFSPE

Dominant

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

2.6. Ex vivo interferon-γ elispot assay Multiscreen MAHAS 4510 plates (Millipore, Bedford, MA) were coated with 60 ␮l/well of sterile carbonate/bicarbonate buffer containing 10 ␮g/ml of anti-murine IFN-␥ (R4, Pharmingen, San Diego, CA) and incubated overnight at room temperature. Plates were washed twice with 200 ␮l/well RPMI medium and twice with complete RPMI (cRPMI) medium containing penicillin/streptomycin, l-glutamine and 10% FBS, and incubated with 200 ␮l/well of cRPMI medium in 5% CO2 at 37 ◦ C for at least 3 h. After blocking, the plates were washed once more with cRPMI before the addition of target and effector cells. A20.2J (ATCC clone HB-98) target cells were washed once with cRPMI, incubated at 5 × 106 cells/ml with or without PyCSP peptide (10 ␮g/ml) or PyHEP17 peptide (20 ␮g/ml) for 1 h at 37 ◦ C in 5% CO2 , and irradiated in a 137 Cs gamma irradiator at 16,000 rads. Next, target cells were washed three times with cRPMI, diluted to 1.5 × 106 cells/ml in cRPMI. To obtain effectors, immunized single cell suspensions prepared from pooled spleens were washed three times, counted and diluted to 5 × 106 cells/ml and 2.5 × 106 cells/ml. Effector and target cells preparations were added to the IFN-␥ coated wells in quadruplicate at 100 ␮l/well, and incubated in 5% CO2 at 37 ◦ C for 36 h. Plates were washed three times with PBS followed by four washes with PBS-T (PBS 0.05% Tween20). Biotinylated anti-IFN-␥ (100 ␮l/well) (XMG1.2, Pharmingen, San Diego, CA) at 2 ␮g/ml in PBS-T were added to the plates and incubated overnight at 4 ◦ C. Plates were washed 6 times with PBS-T and 100 ␮l/well peroxidase-conjugated streptavidin (Kirkegaard & Perry, Gaithersburg, MD) was added at 1:800 dilution in PBS-T. After 1 h incubation at room temperature, plates were washed six times with PBS-T followed by three times with PBS alone, and developed with DAB reagent (Kirkegaard & Perry, Gaithersburg, MD) according to manufacturer’s instructions. After 15 min, the plates were rinsed extensively with dH2 O to stop the colorimetric reaction, dried and stored in the dark. Spots were counted with a KS ELIspot reader (Carl Zeiss Vision, Germany). ELIspot data is presented as IFN-␥ spot forming cells (SFC)/million splenocytes, expressed as mean ± standard deviation (S.D.). 2.7. Intracellular cytokine staining assay (ICS) A20.2J cells were pulsed with or without PyCSP or PyHEP17 antigens (same concentrations as for ELIspot) for 1 h at 37 ◦ C in 5% CO2 , and irradiated as above. Then, 100 ␮l/well of effector cells (5 × 106 cells/ml) and 100 ␮l/well A20.2J target cells (1.5 × 106 cells/ml) pulsed with or without peptides were incubated in duplicate in U-bottom 96-well plates (Costar) in the presence of 1 ␮M Brefeldin A (GolgiPlugTM , Pharmingen, San Diego, CA) in 5% CO2 at 37 ◦ C for 16 h. Plates were spun at 1200 rpm for 5 min, the supernatant flicked, and the cell pellet resuspended by gentle vortexing. Cell surface markers were stained with a combination of 0.3-0.5 ␮l/well of anti-CD8-APC, anti-CD4-PERCP, anti-DX5-FITC or antiCD62L-FITC Abs (Pharmingen, San Diego, CA) in a final volume of 100 ␮l of FACS wash, on ice in the dark for 20 min.

95

After the surface staining, cells were washed with FACS wash twice, gently resuspended, and incubated with 90 ␮l of Perm/Fix buffer (Pharmingen, San Diego, CA) for 20 min on ice in the dark. Next, cells were washed with 100 ␮l of Perm/Wash buffer and intracellular IFN-␥ stained with 0.5 ␮l/well of anti-IFN␥-PE Abs or anti-TNF-␣-PE Abs (Pharmingen, San Diego, CA) in a final volume of 100 ␮l in Perm/Wash buffer. After 20 min incubation on ice in the dark, cells were washed twice with Perm/Wash, once with FACS wash, resuspended in 100 ␮l of FACS wash and stored at 4 ◦ C prior to analysis. Antigen-specific IFN-␥ or TNF-␣ intracellular production was determined by performing four-color fluorescent activated cell sorting using the FACSCaliburTM (Becton Dickinson Immunocytometry Systems, San Jose, CA) with CellQuest software. ICS and FACS data is presented as % T cell population, expressed as mean ± S.D. 2.8. Cytotoxic T lymphocyte (CTL) assay Spleen cells from BALB/c mice collected 3 weeks after the third immunization with PyCSP or PyHEP17 DNA plasmids were incubated at a concentration of 5 × 106 cells in 2 ml of RPMI 1640 medium supplemented with 10% heat inactivated FBS (Sigma), 10 mM HEPES, 2 mM l-glutamine (Gibco), 50 ␮M 2-ME, 50 U/ml penicillin, and 50 U/ml streptomycin (Gibco) in a 24-well plate in the presence of synthetic peptide PyCSP 280-296 containing the previously defined 10-mer CTL epitope (2.5 ␮M), or with a pool of PyHEP17 synthetic peptides (2.5 ␮g/ml each), or with irradiated MHC-matched (H-2d) A20.2J cells transiently transfected 48 h previously with 2.5 ␮g of PyHEP17 plasmid DNA, at a responder: stimulator ratio of 10:1. Rat T-stim (Collaborative Biomedical Products) (2.5%) was added at 48 h as a source of IL-2. After 6 days of incubation, cells were used as effectors in a standard 6 h chromium release assay. Target cells were MHC-matched P815 cells (ATCC, TIB 64) or MHC-mismatched (H-2b) EL4 lymphoma cells (ATCC TIB 39) transiently transfected with PyHEP17 plasmid DNA 48 h previously, or pulsed overnight with the minimal 9-mer CTL epitope PyCSP 280-288, or with a PyHEP17 peptide pool, or with no peptide, and labeled with 100 ␮Ci 51 Cr (sodium chromate solution) (Dupont NEN). Percent lysis was determined as (experimental release − medium control release)/(maximum release − medium control release) × 100. Percent-specific lysis (net) was determined as % lysis (test) − % lysis (control). Responses were classified as positive if percent-specific lysis was greater than 10%. 2.9. Lymphocyte proliferation (LPA) Spleen cells from immunized mice were cultured in quadruplicate at a concentration of 1.25–5 × 105 cells in 0.2 ml of cRPMI medium in a flat-bottom 96-well tissue culture plate in the presence of peptide (at 2.5, 5, 10 or 20 ␮g/ml), without peptide, or with mitogen (conA at 5 ␮g/ml) for 5 days. Wells were then pulsed with 1.0 ␮Ci 3H-methyl thymidine (Dupont NEN) overnight, and uptake assessed by liquid scintillation spectroscopy (Beckman LS6800). Results were expressed as a

96

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

stimulation index (S.I.) (c.p.m. sample/c.p.m. control without peptide).

cleavage site, were found to express the desired P. yoelii proteins at the expected molecular sizes (data not presented).

2.10. Antibody assays

3.2. Effect of ubiquitination and N-end rule on antigen-specific T cell responses

Mice were bled from tail vein for serum approximately 2–3 weeks after each immunization. Antibodies were measured by ELISA against recombinant PyCSP protein, PyHEP17 MR68 peptide [43] or recombinant PyHEP17 protein, as previously described [44]. ELISA data is presented as antibody titers at OD405 . In addition, antibodies were assessed by the indirect fluorescent antibody test (IFAT) against air-dried P. yoelii sporozoites for PyCSP or air-dried P. yoelii parasitized erythrocytes for PyHEP17, as previously described [44]. IFAT data is presented as geometric mean of antibody titers. 3. Results 3.1. In vitro expression of plasmid DNA constructs The in vitro expression of the plasmid DNA constructs (Table 1) was confirmed by analysis of transiently transfected UM449 cells by Western blot. PyCSP and PyHEP17 constructs expressing linear fusions of LAMP or Ub, designed with either stabilizing or destabilizing amino acids distal to the Ub protease

Fig. 1. Antigen-specific IFN-␥ responses induced by immunization with unmodified or ubiquitinated PyHEP17 or PyCSP plasmid DNA vaccines. Ex vivo IFN-␥ ELIspot was carried out using splenocytes from BALB/c mice immunized with unmodified or ubiquitinated PyHEP17 (A) or PyCSP (B) plasmid DNA vaccines harvested 2 weeks after the third immunization as effector cells, stimulated in vitro with MHC-matched A20.2J APCs pulsed with synthetic peptides representing defined CD8+ and CD8+ /CD4+ T cell epitopes, or without peptides. Data are presented as mean ± S.D. (error bars) of spot forming cells (SFC) per million splenocytes (n = 3 mice per group, pooled) of quaduplicate wells. Ub = ubiquitin, A = alanine, R = arginine, M = methionine.

The T cell immunogenicity of plasmid DNA vaccines encoding PyCSP or PyHEP17 fused in-frame to Ub, with unmodified amino acid (M) at the N-terminus of the antigen, or incorporating R or A mutations (Table 1), was compared in mice at 5 and 50 ␮g doses. We found that immunization with DNA vaccines encoding PyHEP17 fused to Ub incorporating a M to R mutation induced higher antigen-specific IFN-␥ responses to the peptide PyHEP17 71–85 – which contains an immunodominant CD4+ T cell epitope with nested CD8+ T cell epitope (Table 2) – as compared with immunization with unmodified PyHEP17 plasmid vaccines or with any of the other Ub fusion constructs, as assessed by ex vivo IFN-␥ ELIspot (Fig. 1A). A less marked effect was observed for responses to peptides PyHEP17 61–75 and 136–150, and there was no effect on responses to peptide PyHEP17 26–40. The enhancement effect was evident both at the 50 and 5 ␮g doses; specifically, the magnitude of the IFN-␥

Fig. 2. Lymphoproliferative responses induced by immunization with unmodified or ubiquitinated PyHEP17 or PyCSP plasmid DNA vaccines. BALB/c mice were immunized with unmodified or ubiquitinated PyHEP17 (A) or PyCSP (B) plasmid DNA vaccines, splenocytes harvested 2 weeks after the third immunization, and assayed by lymphoproliferation against synthetic peptides representing defined CD4+ /CD8+ or CD4+ T cell epitopes, or in media alone. Data are presented as stimulation index (S.I.) (c.p.m. sample/c.p.m. control without peptide), at a peptide concentration of 10 ␮g/ml and cell concentration of 500,000 cells per well for n = 3 mice per group, pooled. These results were reproducible in three out of four LPA experiments. Similar results were obtained at other cell and peptide concentrations (data not shown). Ub = ubiquitin. A = alanine, R = arginine, M = methionine.

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

response induced by immunization with 5 ␮g of Ub-R-PyHEP17 was equivalent to that induced by 50 ␮g of unmodified PyHEP17 plasmid, thus reducing by 10-fold the required dose to produce the same effect (Fig. 1A). IFN-␥ responses induced by DNA vaccines encoding PyHEP17 fused to Ub incorporating a G to A mutation at the C-terminus were higher than those induced by DNA vaccines encoding Ub-M-PyHEP17 fusion, but only

97

marginally superior to those induced by unmodified PyHEP17 DNA vaccine (Fig. 1A). A similar pattern of responses as shown by the ex vivo ELIspot assay was noted for the LPA assay, 6 days after in vitro culture. Thus, the plasmid construct encoding Ub-R-PyHEP17 fusion also induced stronger lymphoproliferation than the unmodified PyHEP17 plasmid or than the other ubiquitinated constructs (Fig. 2A). This effect was also seen

Fig. 3. Antigen-specific, MHC-restricted CTL activity induced by immunization with PyHEP17 (A) or PyCSP (B) plasmid DNA vaccines fused to ubiquitin or LAMP, with or without signal sequence. Effector cells were splenocytes from BALB/c harvested 2 or 3 weeks after the third immunization (n = 3 per group, pooled), and cultured in vitro for 6 days in the presence of synthetic peptides representing defined CTL epitopes (pool of PyHEP17 residues 61–75, 66–88 and 71–85, or PyCSP 280–296). Effectors were then assayed against MHC-matched P815 or MHC-mismatched EL4 (not shown) target cells pulsed with a pool of PyHEP17 9-mer or 15-mer peptides, or with peptide PyCSP 280–288 or 280–296, or without peptide, in a conventional chromium release assay. Percent lysis was determined as (experimental release − medium control release)/(maximum release − medium control release) × 100. Percent-specific lysis (net) was determined as % lysis (test) − % lysis (control). Ub = ubiquitin, LAMP = lysosome-associated membrane protein, –ss = no signal sequence, A = alanine, R = arginine, M = methionine.

98

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

for peptide PyHEP17 71–85, as with ELIspot, and for peptide PyHEP17 126–140. These results were reproducible in three out of four LPA experiments. With regard to the CTL assays, cytotoxic responses to Ub-R PyHEP17 plasmid were better than to unmodified plasmid or to the other DNA vaccine constructs (see sections below), in one of two experiments, although the levels of % specific lysis were low (<20%) (Fig. 3A). Consistent with our previous studies, CTL responses to PyHEP17 are generally difficult to induce and usually of low magnitude. When the same targeting strategy was applied to a different malaria pre-erythrocytic antigen, we found different results. In contrast to the findings with PyHEP17 described above, T cell responses induced by PyCSP Ub fusion constructs were not superior to those induced by unmodified PyCSP plasmid, as assessed by IFN-␥ ELIspot (Fig. 1B), LPA (Fig. 2B), or CTL assays (Fig. 3B). ICS and FACS analysis showed that IFN-␥ and TNF-␣ induced by immunization with PyCSP Ub constructs were primarily produced by CD8+ T cells and not CD4+ T cells, and that the profile of CD8+ IFN-␥ responses as detected by ICS was similar to the profile obtained by the ex vivo ELIspot (data not shown). In summary, targeting Ag for proteosome degradation by ubiquitination and N-end rule moderately enhanced cellular immunity against PyHEP17, as evaluated by IFN-␥ ELIspot, CTL or lymphoproliferation, but there was no apparent effect on antigen-specific cellular immunity for PyCSP plasmids. 3.3. Effect of deleting the native signal sequence in ubiquitinated DNA vaccines on T cell responses In an attempt to improve the enhancement effect of T cell responses after immunization with Ub-R-PyHEP17 DNA vaccine, and to further explore whether the ubiquitination and N-end rule targeting strategy would also be beneficial for PyCSP DNA vaccines, we investigated whether the native signal sequence of the antigen was interfering with the Ub/proteasome MHC Class I targeting. One potential explanation for the results obtained above was that sites for post-translational proteolytic processing are present in the P. yoelii proteins and that if cleavage at those sites occurs before cleavage of the N-terminal Ub, the alterations in the N-terminus of the encoded protein will have no effect on proteasomal targeting. We therefore designed new Ub-fusion constructs which included a deletion of putative hydrophobic leaders (potential sites for post translational proteolytic processing) from the P. yoelii proteins. We then compared the immunogenicity of plasmid DNA vaccines encoding truncated (no signal sequence; –ss) PyCSP or PyHEP17 antigens fused in-frame to Ub, with unmodified (M) or mutated (R) amino acid at the N-terminus of the antigen (Table 1), at 50 ␮g doses. The constructs with the G to A mutation at the Cterminus of the Ub were not further evaluated due to the absence of a clear immune enhancement effect. We found that deleting the signal sequence in PyCSP or PyHEP17 DNA vaccines fused to Ub and bearing destabilizing amino acid R did not enhance the magnitude of antigen-specific T cell responses compared to unmodified plasmid or to Ub-R-PyHEP17 plasmid. Rather, deletion of the signal sequence in either R or M constructs had

Fig. 4. Antigen-specific IFN-␥ responses induced by immunization with PyHEP17 (A) or PyCSP (B) DNA fused to LAMP, or without signal sequence. Ex vivo IFN-␥ ELIspot was carried out using splenocytes from BALB/c mice harvested 3 weeks after the third immunization as effector cells, stimulated in vitro with MHC-matched A20.2J target cells pulsed with synthetic peptides representing defined CD8+ and CD8+ /CD4+ T cell epitopes. Data are presented as mean + S.D. (error bars) of spot forming cells (SFC) per million splenocytes (n = 3 mice per group, pooled) of quaduplicate wells. Ub = ubiquitin, LAMP = lysosome-associated membrane protein, –ss = no signal sequence, R = arginine, M = methionine.

a negative effect, as assessed by CTL (Fig. 3A), IFN-␥ ELIspot (Fig. 4), or LPA (Fig. 5). 3.4. Effect of targeting Ag with LAMP on T cell responses We also compared the T cell immunogenicity of plasmid DNA vaccines encoding PyCSP or PyHEP17 fused in-frame to LAMP, at 50 ␮g doses each, to that of ubiquitinated and truncated fusion constructs, or unmodified plasmids. We found that immunization with DNA vaccines encoding PyHEP17 (Fig. 4A) or PyCSP (Fig. 4B) fused to LAMP induced similar antigenspecific IFN-␥ responses to those induced by immunization with unmodified PyHEP17 or PyCSP plasmid vaccines, as assessed by ex vivo ELIspot. These experiments also confirmed data from the PyHEP17 ELIspot assays reported above, i.e. that plasmids encoding Ub-R-PyHEP17 were the most immunogenic vaccines with regard to IFN-␥ responses (Fig. 4A). With reference to lymphoproliferation, LAMP targeting to the MHC Class II processing and presentation pathway resulted in a modest enhancement of CD4+ T cell-mediated responses. since immunization with PyHEP17–LAMP constructs induced

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

99

Ub-R-PyCSP plasmids induced the lowest antibody titers (Fig. 6A). Similar results were assessed by IFAT against P. yoelii sporozoites (Fig. 6B). PyHEP17-specific antibody responses to MR-68 peptides as detected by ELISA were negligible for most constructs, with only low IgG titers being detected in serum from mice immunized with 50 ␮g of unmodified PyHEP17 plasmid after three doses (data not shown). In contrast to the PyCSP data described above, IFAT against P. yoelii-infected erythrocytes revealed that antibody titers induced by the M- or R-Ub-fusion constructs were similar to that of the unmodified plasmid, and the lowest IgG titers corresponded to the Ub-A-PyHEP17 plasmids (Fig. 6C). 3.6. Effect of deleting signal sequence in ubiquitinated DNA vaccines on antibody responses Consistent with the T cell data, both PyHEP17 and PyCSP Ub DNA vaccines in which the signal sequence of the antigen had

Fig. 5. Lymphoproliferative responses induced by immunization with PyHEP17 (A) or PyCSP (B) DNA fused to LAMP, or without signal sequence. Splenocytes from BALB/c mice harvested 2 or 3 weeks after the third immunization (n = 3 per group, pooled), and stimulated in vitro with synthetic peptides representing defined CD4+ /CD8+ or CD4+ T cell epitopes. Data are presented at a peptide concentration of 2.5–5 ␮g/ml and cell concentration of 250,000 cells per well. Ub = ubiquitin, LAMP = lysosome-associated membrane protein, –ss = no signal sequence, R = arginine, M = methionine.

higher proliferative responses to an immunodominant CD4+ T cell epitope of PyHEP17 (residues 71–85) as compared with immunization with unmodified plasmid (Fig. 5A); there was a very slight trend to increased responses for a second dominant CD4+ T cell epitope (residues 61–75). For PyCSP DNA vaccines, the effect of LAMP on proliferative responses was mixed; overall responses were low, and responses to PyCSP–LAMP compared to unmodified plasmid were marginally better for epitope 280–296 but worse for epitope 57–70 (Fig. 5B). It is noteworthy that, as observed with ELIspot, these separate sets of experiments reproduced the earlier findings of increased immunogenicity of the Ub-R-PyHEP17 DNA vaccines also with regards to LPA responses (Fig. 2A). 3.5. Effect of ubiquitination and N-end rule on antibody responses Consistent with our previous studies, robust antibody responses as assessed by ELISA were only detected for mice immunized with PyCSP DNA vaccines. The highest antibody titers against full-length recombinant protein after two or three doses were induced in mice immunized with native PyCSP plasmid, followed by mice immunized with Ub-A-PyCSP or Ub-M-PyCSP plasmids; immunization with destabilizing

Fig. 6. Antibody responses induced by immunization with unmodified and ubiquitinated PyCSP or PyHEP17 plasmid DNA vaccines, detected by ELISA or IFAT. (A) PyCSP ELISA against recombinant full-length PyCSP protein after two immunizations. (B) PyCSP IFAT against P. yoelii sporozoites. (C) PyHEP17 IFAT against P. yoelii parasitized erythrocytes. Data show results with pooled sera (n = 6) collected 2 weeks after the 2nd or 3rd immunization. Ub = ubiquitin, A = alanine, R = arginine, M = methionine.

100

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

Fig. 7. Antibody responses induced by immunization with PyCSP fused to LAMP, or without signal sequence, detected by ELISA or IFAT. (A) ELISA against recombinant full-length PyCSP protein after two immunizations. (B) IFAT against P. yoelii sporozoites. Data show results with pooled sera (n = 6) collected 2 weeks after the 1st, 2nd or 3rd immunization. Ub = ubiquitin, LAMP = lysosome-associated membrane protein, –ss = no signal sequence, R = arginine, M = methionine.

been removed, elicited the lowest antibody responses compared to unmodified or to the other fusion constructs (Fig. 7). 3.7. Effect of targeting Ag with LAMP on antibody responses In our model, LAMP targeting to MHC Class II processing and presentation pathway did not result in the desired enhancement of CD4+ T cell help for antibody responses to neither of the antigens tested. Rather, immunization with PyCSP–LAMP constructs induced similar (by IFAT) or lower (by ELISA) antibody responses compared to immunization with unmodified plasmid (Fig. 7). In these experiments, PyHEP17 plasmid DNA constructs induced very low antibody titers, and none of the fusion constructs elicited higher antibody titers than unmodified PyHEP17 plasmid (data not shown). 4. Discussion The development of an effective vaccine against the complex human malaria parasites will most likely require the contribution of both arms of the immune system, with the induction of cellular (CD4+ and CD8+ T cells) as well as antibody responses. DNA vaccines provide an excellent platform for the inclusion of multiple antigens to combat diseases caused by multi-stage pathogens such as Plasmodium falciparum because of their simplicity, stability, and ease of large-scale production. Plasmid

DNA vaccines are particularly adequate for the induction of cellular immunity [22,1], but thus far they have been relatively inefficient in generating antibody responses [21–24]. Using the malaria rodent species P. yoelii as a model, we aimed to evaluate the immune enhancement potential of different antigen targeting strategies to selectively improve responses mediated by CD8+ and CD4+ T lymphocytes and by antibodies, induced after intramuscular immunization with DNA plasmids. In this study, we attempted to enhance the immunogenicity of first generation DNA vaccines by targeting the model P. yoelii pre-erythrocytic antigens PyCSP and PyHEP17 for (i) cytoplasmatic degradation by the proteasome by ubiquitination and the N-end rule and Class I presentation [45], and (ii) lysosomal degradation by fusing them to LAMP and Class II presentation [46]. In the first strategy for MHC Class I targeting, we constructed a series of DNA vaccines encoding PyCSP and PyHEP17 covalently attached to ubiquitin and bearing either stabilizing (M) or destabilizing (R) amino acids at the amino terminus. The capacity of these DNA vaccines to induce cellular immune responses (IFN-␥ ELIspot, ICS, CTL, LPA) and antibodies (ELISA, IFAT) was assessed. With regards to fusion constructs including a G to A mutation at the C-terminus of the ubiquitin, we did not find an increased CTL and decreased antibody responses, in contrast to previous studies using the LCMV model [33,34]. Nevertheless, we found that ubiquitination of the antigen bearing destabilizing R resulted in increased antigen-specific CD8+ T cell responses as assessed by IFN-␥ production and cytotoxicity. This is consistent with the increased cytoplasmatic degradation and CTL responses to influenza [27] or HIV [28,29] proteins reported in previous studies. However, in our model, the CD8+ T cell-mediated immune enhancement effect was observed for the PyHEP17 but not for the PyCSP antigen. Other studies that have applied this strategy with other viral antigens have reported different results. Thus, with other influenza [30] or HCV [31] proteins using a similar targeting strategy, there was no evidence of enhancement of protein half life or magnitude of CTL responses. Therefore, conflicting results have been reported in the literature with different antigens and infectious systems. Interestingly, the ubiquitination and N-end terminal rule strategy designed to improve MHC Class I-mediated cellular responses also resulted in improved cytokine and proliferative responses mediated by CD4+ T cells. It could be that increasing protein degradation by the proteasome also yielded peptides that could be taken up by MHC Class II molecules. However, and consistent with previous studies, targeting of antigens to the MHC Class I pathway resulted in a decrease in antibody responses, indicating that protein antigens were more degraded and that B cells received less efficient help from CD4+ T lymphocytes. The lack of effect of destabilizing strategies on PyCSP may be related to the inherent natural instability of this protein after sporozoite infection. It is possible that the N-terminus of PyCSP contains a protease site which is cleaved rapidly in mammalian cells. If this was the case, any modification attached to the N-terminus will be rapidly removed and will be unable to alter the course of antigen processing and presentation. Such effects might account for the variable results obtained with these

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

strategies between different antigens. The presence of glycosylphosphatidylinositol (GPI) motifs [47] may also influence antigen processing and presentation. Many plasmodia surface proteins, including the PyCSP, are anchored by a GPI motif that is highly conserved among different species of Plasmodium. GPI appears to be poorly processed in mammalian cells and may interfere with protein expression. It has been shown that the presence of the GPI anchor on PyCSP affects total protein production, cellular distribution (subcellular localization), antigen processing and secretion, leading to less effective antigen presentation and reduced immunogenicity and protective capacity in mice upon immunization with a recombinant adenovirus expressing PyCSP [48]. It was subsequently demonstrated that a significant enhancement of IFN-␥ producing T cells was obtained upon vaccination with a recombinant adenovirus expressing PyCSP with a disrupted GPI anchor. With regard to the effect of GPI anchor on the outcome of the destabilizing strategies evaluated in our study, we would expect a PyCSP construct with deleted GPI anchor to be more immunogenic, but that the overall results of our targeting strategies would be the same as for the GPI containing PyCSP. In the second strategy for MHC Class II targeting, we constructed two DNA vaccines encoding PyCSP and PyHEP17 fused to the lysosomal targeting sequence of LAMP. In contrast to what has been reported in other studies using LAMP-fusion viral proteins [35–38], this approach did not result in enhanced antibody responses to the expressed antigens; we could only detect a marginal increase in CD4+ T cell proliferative responses. Other previous studies have also failed to show an increase in either proliferative or antibody responses [31]. Overall, our results and those reported in the literature indicate that there is a strong antigen dependence of the effects of Class I and Class II antigen targeting strategies. For example, it has been shown that ubiquitination is important in antigen processing but that the process is tightly linked to the nature of the antigen [49]. Other studies suggest that mammalian codon optimization of the antigen gene might be relevant when fusing Ub to some proteins [50]. Thus, in an attempt to elucidate why Ub fusion does not improve immune responses for some antigens, a study investigated the “rules” for proteasome targeting [51]. The authors concluded that the efficacy of ubiquitination to enhance CD8+ responses, especially against subdominant antigen epitopes, was dependent upon mammalian codon optimization of the DNA constructs [51]. Alternatively, the failure of our DNA constructs to induce the desired immuno-enhancement effect in vivo could potentially be explained by a failure of the antigens to reach the desired MHC pathway, as targeting of the antigen to the proteasome or to the endosomes after fusion to ubiquitin or LAMP was not definitely proven in vitro. In conclusion, the effect of targeting antigens for processing and presentation by ubiquitination, N-end rule and LAMP fusion may depend upon the infectious disease model and the target antigen. Strategies aimed at targeting the ubiquitin–proteasome pathway in MHC Class I antigen processing and presentation are considered to have greater potential for viral and anticancer T cell vaccines [52], and strategies aimed at targeting the endosomal/lysosomal pathway in MHC Class II antigen processing and

101

presentation could help increasing the potency of the above vaccines by providing T cell help and to improve antibody responses to neutralize extracellular pathogens. Although we have shown some potential for MHC Class I antigen targeting in our Plasmodium model, the applicability of these strategies for vaccine development against malaria and other infectious agents will have to be tested and validated in a case-by-case basis. Acknowledgements We thank Ms. Norma Graber and Dr. Gary T. Brice for assistance with T cell assays. Work was supported by funds allocated to the Naval Medical Research Center by the U.S. Army Medical Research and Materiel Command, work unit 62787A.870.F. A0080. The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, nor the U.S. Government. References [1] Sedegah M, Hedstrom R, Hobart P, Hoffman SL. Protection against malaria by immunization with plasmid DNA encoding circumsporozoite protein. Proc Natl Acad Sci USA 1994;91:9866–70. [2] Doolan DL, Hedstrom RC, Rogers WO, Charoenvit Y, Rogers M, De la Vega P, et al. Identification and characterization of the protective hepatocyte erythrocyte protein 17 kDa gene of Plasmodium yoelii, homolog of Plasmodium falciparum exported protein 1. J Biol Chem 1996;271:17861–8. [3] Kirman JR, Seder RA. DNA vaccination: the answer to stable, protective T-cell memory? Curr Opin Immunol 2003;15:471–6. [4] Donnelly JJ, Wahren B, Liu MA. DNA vaccines: progress and challenges. J Immunol 2005;175:633–9. [5] McKenzie BS, Corbett AJ, Brady JL, Dyer CM, Strugnell RA, Kent SJ, et al. Nucleic acid vaccines: tasks and tactics. Immunol Res 2001;24: 225–44. [6] Doria-Rose NA, Haigwood NL. DNA vaccine strategies: candidates for immune modulation and immunization regimens. Methods 2003;31:207–16. [7] Hung CF, Wu TC. Improving DNA vaccine potency via modification of professional antigen presenting cells. Curr Opin Mol Ther 2003;5:20–4. [8] Groettrup M, Soza A, Kuckelkorn U, Kloetzel PM. Peptide antigen production by the proteasome: complexity provides efficiency. Immunol Today 1996;17:429–35. [9] Rock KL, York IA, Saric T, Goldberg AL. Protein degradation and the generation of MHC Class I-presented peptides. Adv Immunol 2002;80:1–70. [10] Hershko A, Ciechanover A. The ubiquitin system. Annu Rev Biochem 1998;67:425–79. [11] Chau V, Tobias JW, Bachmair A, Marriott D, Ecker DJ, Gonda DK, et al. A multiubiquitin chain is confined to specific lysine in a targeted short-lived protein. Science 1989;243:1576–83. [12] Hershko A. Ubiquitin-mediated protein degradation. J Biol Chem 1988;263:15237–40. [13] Bartel B, Wunning I, Varshavsky A. The recognition component of the N-end rule pathway. Embo J 1990;9:3179–89. [14] Varshavsky A. The N-end rule. Cell 1992;69:725–35. [15] Bachmair A, Finley D, Varshavsky A. In vivo half-life of a protein is a function of its amino-terminal residue. Science 1986;234:179–86. [16] Bachmair A, Varshavsky A. The degradation signal in a short-lived protein. Cell 1989;56:1019–32. [17] Bourgeois C, Tanchot C. Mini-review CD4 T cells are required for CD8 T cell memory generation. Eur J Immunol 2003;33:3225–31. [18] Rocha B, Tanchot C. Towards a cellular definition of CD8+ T-cell memory: the role of CD4+ T-cell help in CD8+ T-cell responses. Curr Opin Immunol 2004;16:259–63.

102

C. Doba˜no et al. / Immunology Letters 111 (2007) 92–102

[19] Hirunpetcharat C, Tian JH, Kaslow DC, van Rooijen N, Kumar S, Berzofsky JA, et al. Complete protective immunity induced in mice by immunization with the 19-kilodalton carboxyl-terminal fragment of the merozoite surface protein-1 (MSP1[19]) of Plasmodium yoelii expressed in Saccharomyces cerevisiae: correlation of protection with antigenspecific antibody titer, but not with effector CD4+ T cells. J Immunol 1997;159:3400–11. [20] Cohen S, Mc GI, Carrington S. Gamma-globulin and acquired immunity to human malaria. Nature 1961;192:733–7. [21] Epstein JE, Gorak EJ, Charoenvit Y, Wang R, Freydberg N, Osinowo O, et al. Safety, tolerability, and lack of antibody responses after administration of a PfCSP DNA malaria vaccine via needle or needle-free jet injection, and comparison of intramuscular and combination intramuscular/intradermal routes. Hum Gene Ther 2002;13:1551–60. [22] Wang R, Doolan DL, Le TP, Hedstrom RC, Coonan KM, Charoenvit Y, et al. Induction of antigen-specific cytotoxic T lymphocytes in humans by a malaria DNA vaccine. Science 1998;282:476–80. [23] Wang R, Epstein J, Baraceros FM, Gorak EJ, Charoenvit Y, Carucci DJ, et al. Induction of CD4(+) T cell-dependent CD8(+) type 1 responses in humans by a malaria DNA vaccine. Proc Natl Acad Sci USA 2001;98:10817–22 (Epub 2001, August 28). [24] Wang R, Richie TL, Baraceros MF, Rahardjo N, Gay T, Banania JG, et al. Boosting of DNA vaccine-elicited gamma interferon responses in humans by exposure to malaria parasites. Infect Immun 2005;73:2863–72. [25] Guarnieri FG, Arterburn LM, Penno MB, Cha Y, August JT. The motif Tyr-X-X-hydrophobic residue mediates lysosomal membrane targeting of lysosome-associated membrane protein 1. J Biol Chem 1993;268:1941–6. [26] Holowaty MN, Sheng Y, Nguyen T, Arrowsmith C, Frappier L. Protein interaction domains of the ubiquitin-specific protease. J Biol Chem 2003;278:47753–61 (Epub 2003, September 23). [27] Townsend A, Bastin J, Gould K, Brownlee G, Andrew M, Coupar B, et al. Defective presentation to class I-restricted cytotoxic T lymphocytes in vaccinia-infected cells is overcome by enhanced degradation of antigen. J Exp Med 1988;168:1211–24. [28] Tobery TW, Siliciano RF. Targeting of HIV-1 antigens for rapid intracellular degradation enhances cytotoxic T lymphocyte (CTL) recognition and the induction of de novo CTL responses in vivo after immunization. J Exp Med 1997;185:909–20. [29] Tobery T, Siliciano RF. Cutting edge: induction of enhanced CTLdependent protective immunity in vivo by N-end rule targeting of a model tumor antigen. J Immunol 1999;162:639–42. [30] Fu TM, Guan L, Friedman A, Ulmer JB, Liu MA, Donnelly JJ. Induction of MHC Class I-restricted CTL response by DNA immunization with ubiquitin-influenza virus nucleoprotein fusion antigens. Vaccine 1998;16:1711–7. [31] Vidalin O, Tanaka E, Spengler U, Trepo C, Inchauspe G. Targeting of hepatitis C virus core protein for MHC I or MHC II presentation does not enhance induction of immune responses to DNA vaccination. DNA Cell Biol 1999;18:611–21. [32] Varshavsky A. The N-end rule: functions, mysteries, uses. Proc Natl Acad Sci USA 1996;93:12142–9. [33] Rodriguez F, Zhang J, Whitton JL. DNA immunization: ubiquitination of a viral protein enhances cytotoxic T-lymphocyte induction and antiviral protection but abrogates antibody induction. J Virol 1997;71:8497–503. [34] Rodriguez F, An LL, Harkins S, Zhang J, Yokoyama M, Widera G, et al. DNA immunization with minigenes: low frequency of memory cytotoxic T lymphocytes and inefficient antiviral protection are rectified by ubiquitination. J Virol 1998;72:5174–81. [35] Wu TC, Guarnieri FG, Staveley-O’Carroll KF, Viscidi RP, Levitsky HI, Hedrick L, et al. Engineering an intracellular pathway for major histocompatibility complex class II presentation of antigens. Proc Natl Acad Sci USA 1995;92:11671–5. [36] Rowell JF, Ruff AL, Guarnieri FG, Staveley-O’Carroll K, Lin X, Tang J, et al. Lysosome-associated membrane protein-1-mediated targeting of the HIV-1 envelope protein to an endosomal/lysosomal compartment enhances its presentation to MHC Class II-restricted T cells. J Immunol 1995;155:1818–28.

[37] Thomson SA, Burrows SR, Misko IS, Moss DJ, Coupar BE, Khanna R. Targeting a polyepitope protein incorporating multiple class II-restricted viral epitopes to the secretory/endocytic pathway facilitates immune recognition by CD4+ cytotoxic T lymphocytes: a novel approach to vaccine design. J Virol 1998;72:2246–52. [38] Bonini C, Lee SP, Riddell SR, Greenberg PD. Targeting antigen in mature dendritic cells for simultaneous stimulation of CD4+ and CD8+ T cells. J Immunol 2001;166:5250–7. [39] Hartikka J, Sawdey M, Cornefert-Jensen F, Margalith M, Barnhart K, Nolasco M, et al. An improved plasmid DNA expression vector for direct injection into skeletal muscle. Hum Gene Ther 1996;7:1205–17. [40] Finch JS, Bonham K, Krieg P, Bowden GT. Murine polyubiquitin mRNA sequence. Nucleic Acids Res 1990;18:1907. [41] Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989. [42] Charoenvit Y, Mellouk S, Cole C, Bechara R, Leef MF, Sedegah M, et al. Monoclonal, but not polyclonal antibodies protect against Plasmodium yoelli sporozoites. J Immunol 1991;146:1020–5. [43] Charoenvit Y, Majam Fallarme V, Corradin G, Sacci Jr JB, Wang R, Doolan DL, et al. CD4(+) T-cell- and gamma interferon-dependent protection against murine malaria by immunization with linear synthetic peptides from a Plasmodium yoelii 17-kilodalton hepatocyte erythrocyte protein. Infect Immun 1999;67:5604–14. [44] Charoenvit Y, Leef ML, Yuan LF, Sedegah M, Beaudoin RL. Characterization of Plasmodium yoelii monoclonal antibodies directed against stage-specific sporozoite antigens. Infect Immun 1987;55:604–8. [45] Weissman AM. Regulating protein degradation by ubiquitination. Immunol Today 1997;18:189–98. [46] Ruff AL, Guarnieri FG, Staveley-O’Carroll K, Siliciano RF, August JT. The enhanced immune response to the HIV gp160/LAMP chimeric gene product targeted to the lysosome membrane protein trafficking pathway. J Biol Chem 1997;272:8671–8. [47] Gowda DC, Davidson EA. Protein glycosylation in the malaria parasite. Parasitol Today 1999;15:147–52. [48] Bruna-Romero O, Rocha CD, Tsuji M, Gazzinelli RT. Enhanced protective immunity against malaria by vaccination with a recombinant adenovirus encoding the circumsporozoite protein of Plasmodium lacking the GPIanchoring motif. Vaccine 2004;22:3575–84. [49] Michalek MT, Grant EP, Gramm C, Goldberg AL, Rock KL. A role for the ubiquitin-dependent proteolytic pathway in MHC Class I-restricted antigen presentation. Nature 1993;363:552–4. [50] Ramakrishna L, Anand KK, Mahalingam M, Mohankumar KM, Ramani S, Siddappa NB, et al. Codon optimization and ubiquitin conjugation of human immunodeficiency virus-1 Tat lead to enhanced cell-mediated immune responses. Vaccine 2004;22:2586–98. [51] Andersson HA, Barry MA. Maximizing antigen targeting to the proteasome for gene-based vaccines. Mol Ther 2004;10:432–46. [52] Sijts A, Zaiss D, Kloetzel PM. The role of the ubiquitin–proteasome pathway in MHC Class I antigen processing: implications for vaccine design. Curr Mol Med 2001;1:665–76. [53] Del Giudice G, Grillot D, Renia L, Muller I, Corradin G, Louis JA, et al. Peptide-primed CD4+ cells and malaria sporozoites. Immunol Lett 1990;25:59–63. [54] Franke ED, Sette A, Sacci Jr J, Southwood S, Corradin G, Hoffman SL. A subdominant CD8(+) cytotoxic T lymphocyte (CTL) epitope from the Plasmodium yoelii circumsporozoite protein induces CTLs that eliminate infected hepatocytes from culture. Infect Immun 2000;68:3403–11. [55] Renia L, Marussig MS, Grillot D, Pied S, Corradin G, Miltgen F, et al. In vitro activity of CD4+ and CD8+ T lymphocytes from mice immunized with a synthetic malaria peptide. Proc Natl Acad Sci USA 1991;88:7963–7. [56] Weiss WR, Mellouk S, Houghten RA, Sedegah M, Kumar S, Good MF, et al. Cytotoxic T cells recognize a peptide from the circumsporozoite protein on malaria-infected hepatocytes. J Exp Med 1990;171:763–73. [57] Doba˜no C, Doolan DL. Identification of minimal CD8+ and CD4+ T cell epitopes in the Plasmodium yoelii hepatocyte erythrocyte protein 17 kDa. Mol Immunol 2007;44(11):3037–48.