A Novel Vaccine Approach for Chagas Disease Using Rare Adenovirus Serotype 48 Vectors

Due to the increasing amount of people afflicted worldwide with Chagas disease and an increasing prevalence in the United States, there is a greater need to develop a safe and effective vaccine for this neglected disease. Adenovirus serotype 5 (Ad5) is the most common adenovirus vector used for gene therapy and vaccine approaches, but its efficacy is limited by preexisting vector immunity in humans resulting from natural infections. Therefore, we have employed rare serotype adenovirus 48 (Ad48) as an alternative choice for adenovirus/Chagas vaccine therapy. In this study, we modified Ad5 and Ad48 vectors to contain T. cruzi’s amastigote surface protein 2 (ASP-2) in the adenoviral early gene. We also modified Ad5 and Ad48 vectors to utilize the “Antigen Capsid-Incorporation” strategy by adding T. cruzi epitopes to protein IX (pIX). Mice that were immunized with the modified vectors were able to elicit T. cruzi-specific humoral and cellular responses. This study indicates that Ad48-modified vectors function comparable to or even premium to Ad5-modified vectors. This study provides novel data demonstrating that Ad48 can be used as a potential adenovirus vaccine vector against Chagas disease.


Introduction
Chagas disease (American trypanosomiasis) is one of the 17 neglected tropical diseases (NTDs) affecting the world today [1,2]. Trypanosoma cruzi (T. cruzi) is the intracellular parasite that causes Chagas disease (CD) [3]. T. cruzi is transmitted to the mammalian host at the site of a triatomine bug bite [4]. (CD) is an illness that once was only common in the Latin America region [4], now the disease is responsible for 10-50,000 deaths/year and infecting 12-20 million people worldwide [5]. Chagas disease has two clinical stages, acute and chronic stage [6]. The acute stage occurs after initial infection but often goes unobserved due to mild symptoms [7]. The acute stage can be deadly among children and immunocompromised adults [8]. There are pharmacological treatments for the acute stage however; the treatments are highly toxic [9]. Nifurtimox and Benznidazol are current treatments for this infection [9]. These anti-parasitic drugs are 80% successful in curing the acute phase with severe side effects. When the acute stage is untreated, the disease becomes chronic [7], anti-parasitic drugs are ineffective in curing the chronic phase.
The chronic stage can remain asymptomatic for many years. There is no treatment for the chronic stage. One-third of CD patients develop chronic chagasic cardiomyopathy (CCC) associated with protein as well as a transcriptional activator of Ad genes and it is able to reorganize host cell nuclear domains [41]. In addition, pIX demonstrates a capacity to tolerate significant genetic modifications, including the addition of large polypeptides [42][43][44][45] such as cell-type-specific targeting ligands [46][47][48] and antigens [49].
Based on the successful modification of Ad5 pIX, we speculated that pIX of Ad48 would serve as a suitable incorporation site for antigen incorporation. We reasoned that a vaccine vector including gp83 neutralizing epitope, amastigote surface protein-2 (ASP-2), or an epitope of the carboxyl region of amastigote surface protein-2 (ASP-C), respectively, would be beneficial for a T. cruzi vaccine vector. ASP-2 is a protein of yet unknown function in the biology of the parasite [50][51][52]; however, it has been important for pre-clinical T. cruzi vaccine efforts [53,54].

Cell Culture
Human embryonic kidney (HEK293) cells were obtained from and cultured in the medium recommended by the American Type Culture Collection (Manassas, VA, USA). The cell line was incubated at 37˝C and 5% CO 2 under humidified conditions.
To rescue the Ad5-modified vectors, the recombinant adenoviral genomes were digested with PacI, and transfected with PolyJet (SignaGen Laboratories, Ijamsville, MD, USA) into the HEK293 cells. Multi-step large-scale propagations of recombinant Ad5 vectors were performed after the vectors were rescued. Viruses were purified by double cesium chloride (CsCl) gradient ultracentrifugation and dialyzed against 10% glycerol in phosphate buffered saline (PBS) without Mg 2+ or Ca 2+ and 10% glycerol. Viruses were stored at´80˝C until use.
To generate Ad48-CMV-ASP-2, the full length ASP-2 was amplified by PCR using two primers that introduced a KpnI site on the 5 1 end and an EcoRI site on the 3 1 end of the ASP-2 fragment. The digested fragment was cloned into the multiple cloning site of the Ad48 adaptor plasmid, pAdApt48 [29]. To rescue the modified vector, the recombinant plasmid, along with pWE.Ad48.∆E3.5orf6 [29] was co-transfected into HEK293 cells with PolyJet. The transfected cells were coated with agarose after 24 h and continued in culture at 37˝C in 5% CO 2 until plaques formed. The plaques were then collected and processed for large-scale upscale. Viruses were purified by double CsCl gradient ultracentrifugation and dialyzed against 10% glycerol in PBS without Mg 2+ or Ca 2+ and 10% glycerol. Viruses were stored at´80˝C until use. To generate Ad48-pIX-ASP-C, a short DNA sequence which included FLAG epitope, MfeI restriction site, and the c-terminal immunodominant region of ASP-2 (AA 542-560) was produced by annealing the following oligonucleotides: (top/bottom strand sequences): 5 1 -AATTGG ATTATAAGGATGACGATGACAAGAGTCGGGAGGATAACAGACAGTACAGCTTTGTGAACCACA GATTCACTCTTGTGTAGC-3 1 and 5 1 -AATTGCTACACAAGAGTGAATCTGTGGTTCACAAAGCTGT ACTGTCTGTTATCCTCCCGACTCTTGTCATCGTCATCCTTATAATCC-3 1 . The annealed product was ligated into a modified pAdApt48 (Mfe1 was introduced to the adaptor plasmid). The recombinant vector was rescued, upscaled, and purified as previously described for Ad48-CMV-ASP-2. The same cloning procedure was used to create Ad48-pIX-gp83 vector. The following oligonucleotides were annealed: 5 1 -AATTGAAAATTTATTGGAAACAGCCAGTGGAAGGGACGAAGAGTTGGACGCTG TCGAAGCATCACCATCACCATCACTAGC-3 1 and 5 1 -AATTGCTAGTGATGGTGATGGTGATGCTTC GACAGCGTCCAACTCTTCGTCCCTTCCACTGGCTGTTTCCAATAAATTTTC-3 1 . The annealed product was ligated into the modified pAdApt48. The recombination, rescue and purification steps were the same as described for Ad48-pIX-ASP-C. All incorporations were verified by DNA sequencing.
The physical titer of each purified vector was determined by measuring an absorbance at 260 nm and expressed as viral particles (VPs) per mL. The infectious particles (IPs) per mL were determined by tissue culture infectious dose (TCID 50 ) assay [56].

Western Blot Analysis
Protein expression was analyzed from lysates of HEK293 cells infected with Ad5-CMV-ASP-2 and Ad48-CMV-ASP-2. The other modified vectors, Ad5-pIX-ASP-C, Ad5-pIX-gp83, Ad48-pIX-ASP-C, and Ad48-pIX-gp83 (5ˆ10 9 VPs/vector) were analyzed for the protein of interest incorporated within the capsid. The lysates and vectors were denatured by boiling and resolved on SDS-PAGE gels, followed by transfer onto polyvinylidene difluoride (PVDF) membranes, which were then blocked with 5% dry non-fat milk in PBST for 1 h. Thereafter, the membranes were incubated overnight at 4˝C with His 6 MAb (1:5000 dilution in blocking buffer; GenScript), or FLAG-HRP antibody. After overnight incubation, the membranes were washed, and the His 6 membrane was incubated with HRP-conjugated goat anti-mouse antibody (1:5000; Millipore, Temecula, MA, USA). The proteins were detected by using 3 1  In order to investigate the exposure-display of T. cruzi epitopes on the surface of the pIX modified vectors, whole virus ELISAs were performed. In brief, the ELISA plates were coated with serial dilutions of the Ad5-pIX-ASP-C, Ad5-pIX-gp83, Ad48-pIX-ASP-C, Ad48-pIX-gp83, or the controls (Ad5 and Ad48). The immobilized vectors were blocked on 96-well plates. The immobilized vector was incubated with His 6 MAb (1:2000; GenScript) or FLAG-HRP antibody. The FLAG-incubated plate was washed and set up for the detection step. The His 6 -incubated plate was washed and incubated with the HRP-conjugated goat anti-mouse antibody (1:5000; Millipore). ELISAs were developed with the SIGMAFAST OPD peroxidase substrate (Sigma-Aldrich) and measured at OD 450 nm.
ELISA was performed to determine the ability of the pIX-modified vectors to induce humoral immune responses. Sera from the immunized mice were collected two weeks after prime and 2 weeks after boost. ELISA plates were coated with either 10 µM of gp83-18 peptide (KIYWKQPVEGTKSWTLSK) or ASP-C peptide (SREDNRQYSFVNHRFTLV) (GenScript) in 100 µL of 50 mM carbonate buffer per well as previously described in [16]. Unbound peptide was removed by washing with PBST buffer (1ˆPBS and 0.05% Tween 20). The plate was then blocked with 5% non-fat milk/PBST for 1 h at room temperature and 1 µL of serum samples (diluted 1:100) from control and immunized mice were applied to the plate and incubated for 2 h at room temperature. The plate was then extensively washed and blocked again followed by incubation with the HRP-conjugated goat anti-mouse antibody (1:5000; Millipore). ELISAs were developed with the SIGMAFAST OPD peroxidase substrate (Sigma-Aldrich) and measured at OD 450 nm. The amount of IgG in the sera was calculated based on a standard curve of mouse IgG protein.

Mice Immunizations
Mice immunizations with vectors (Ad5-CMV-ASP-2, Ad5-pIX-ASP-C, Ad48-CMV-ASP-2, Ad48-pIX-ASP-C, Ad5-pIX-gp83, and Ad48-pIX-gp83) were performed to determine the T. cruzi-specific immunogenicity. Groups of seven C57BL/6 mice (6-8 weeks old female) were immunized intramuscularly with the corresponding vectors (1ˆ10 10 or 3ˆ10 10 VP/mouse) at each time-point, with a two-week interval between prime, boost, and reboost. The animals were euthanized two weeks after reboost; subsequently spleens were collected for further analysis. The University of Alabama at Birmingham Institutional Animal Use and Care Committee approved the use of mice as described herein under the approved protocol number 141209997.

Intracellular Cytokine Staining (ICCS) Assay
Splenocytes collected from immunized C57BL/6 mice (four mice per group) were treated with ACK Lysing buffer (Life Technologies, Carlsbad, CA, USA) and the cell concentration was adjusted to 2ˆ10 6 cells/mL in 500 µL of cell culture medium containing CD107a-FITC (2 mg/mL), and Golgi stop (Monensin) (10 µg/mL), (BD Biosciences, San Jose, CA, USA). The ASP-C peptide, SREDNRQYSFVNHRFTLV (10 µM) was added to the experimental tubes; PMA (50 ng/mL) and ionomycin (1 µg/mL) (Sigma-Aldrich) were added to the positive control tubes. All tubes were incubated for 6 h at 37˝C in the presence of 5% CO 2 . The cells were washed with PBS+ 1% FBS and labeled with surface antibodies. The samples were stained with CD3-Pacific Blue hamster-anti-mouse (BD Biosciences), CD4-APC-efluor 780 anti-mouse (eBiosciences, San Diego, CA, USA), and CD8-PE rat-anti-mouse (BD Biosciences) to determine the surface phenotype. The cells were then washed twice with PBS+ 1% FBS and permeabilized with the Cytofix/cytoperm reagent (BD Biosciences) for 20 min at 4˝C in the dark. After being washed twice, cells were stained for intracellular markers TNFα-PE-Cy7 rat-anti-mouse and IFNγ-Alexa Fluor 700-rat-anti-mouse (BD Biosciences), for 40 min at 4˝C in the dark. Finally, cells were washed twice and fixed in 1% formalin. At least 100,000 CD3+ events were acquired from each sample using a Becton Dickinson LSR II flow cytometer (BD Biosciences) and data was analyzed using FlowJo Version 10 software (TreeStar). Lymphocytes were analyzed based on forward and side scatter profiles. The gates were set based on the media control and these gates were applied to all samples from the same individual for each time point. Cytokine production was measured from the CD3 + CD4 + or the CD3 + CD8 + gates relative to the media control values.

Statistical Analyses
Descriptive statistics, such as means and standard deviations, or standard error of the mean, were computed to study variables of interest. Statistical analyses were performed by the nonpaired two-tailed Student t-test, assuming equal variance. Statistical significance was defined as p < 0.05.

Construction of Ad5 and Ad48 Vectors
For this study, we have developed a series of recombinant Ad5 and Ad48 vectors, which are illustrated in Figure 1. We constructed Ad5 and Ad48 vectors expressing T. cruzi's amastigote surface protein 2 (ASP-2) along with His 6 under the transcriptional control of CMV ( Figure 1B,F). We developed Ad5 and Ad48 vectors that have an immunodominant CD8 T-cell epitope (VNHRFTLV) from the carboxyl-terminal region of ASP-2 as well as a FLAG (DYKDDDDK) epitope incorporated into the minor Ad capsid protein IX (pIX) ( Figure 1C,G). In addition, we also developed Ad5 and Ad48 vectors in which a T. cruzi gp83 trans-sialidase epitope [15] was incorporated into pIX ( Figure 1D,H). A FLAG epitope was also incorporated onto pIX in the Ad5-pIX-modifed vector. The modified Ad genomes were partially sequenced to confirm that the correct genes were incorporated. Subsequent transfection of HEK293 cells with the sequenced verified recombinant Ad genomes resulted in rescue of the following vectors: Ad5, Ad5-CMV-ASP-2, Ad5-pIX-ASP-C, Ad5-pIX-gp83, Ad48, Ad48-CMV-ASP-2, Ad48-pIX-ASP-C, and Ad48-pIX-gp83. correct genes were incorporated. Subsequent transfection of HEK293 cells with the sequenced verified recombinant Ad genomes resulted in rescue of the following vectors: Ad5, Ad5-CMV-ASP-2, Ad5-pIX-ASP-C, Ad5-pIX-gp83, Ad48, Ad48-CMV-ASP-2, Ad48-pIX-ASP-C, and Ad48-pIX-gp83.  (C) Ad5-pIX-ASP-C, Ad5 replication-defective genome containing an immunodominant region from the c-terminal of ASP-2 and a FLAG epitope within the pIX locale; (D) Ad5-gp83, Ad5 replication-defective genome containing an incorporated neutralizing T. cruzi trypomastigote gp83 epitope as well as His 6 and FLAG epitopes within the pIX locale; (E) Ad48, a replication-defective adenovirus with wildtype pIX; (F) Ad48-CMV-ASP-2, replication-defective Ad48 with full length ASP-2 plus His 6 under the control of CMV promoter; (G) Ad48-pIX-ASP-C, Ad48 replication-defective genome containing an immunodominant region from the c-terminal of ASP-2 and a FLAG epitope within the pIX locale; (H) Ad48-gp83, Ad48 replication-defective genome containing an incorporated neutralizing T. cruzi trypomastigote gp83 epitope within the pIX locale.

Expression of ASP-2 and T. cruzi Epitopes
After successful incorporation of ASP-2 and T. cruzi epitopes, we next sought to verify expression of our transgenes and capsid incorporations at the protein level by Western blot analysis. Protein expression was analyzed from lysates generated from HEK293 cells infected with Ad5-CMV-ASP-2 and Ad48-CMV-ASP-2 as well as modified vectors Ad5-pIX-ASP-C, Ad5-pIX-gp83, Ad48-pIX-ASP-C, and Ad48-pIX-gp83. The lysates and vectors were denatured by boiling and resolved on SDS-PAGE gels. The probing of separated lysates and viral proteins of the modified vectors using His 6

Expression of ASP-2 and T. cruzi Epitopes
After successful incorporation of ASP-2 and T. cruzi epitopes, we next sought to verify expression of our transgenes and capsid incorporations at the protein level by Western blot analysis. Protein expression was analyzed from lysates generated from HEK293 cells infected with Ad5-CMV-ASP-2 and Ad48-CMV-ASP-2 as well as modified vectors Ad5-pIX-ASP-C, Ad5-pIX-gp83, Ad48-pIX-ASP-C, and Ad48-pIX-gp83. The lysates and vectors were denatured by boiling and resolved on SDS-PAGE gels. The probing of separated lysates and viral proteins of the modified vectors using His6 antibody detected the presence of protein bands with molecular weights of 78 kDa for Ad5-CMV-ASP-2 ( Figure 2A

T. cruzi Antigens, Incorporated within pIX, Are Exposed on the Virion Surface
It is known that Ads can assemble and function with a defected/deleted pIX capsid protein [57]. For the modified pIX capsid incorporated vectors, we performed ELISA assays to verify that pIXs were intact and the T. cruzi antigens were accessible on the vectors' surface ( Figure 3). Serial diluted vectors were immobilized in the wells of an ELISA plate and incubated with either His6 or FLAG-HRP antibody. In reference to FLAG antibody, Ad5-pIX-ASP-C, Ad48-pIX-ASP-C, and Ad5-pIX-gp83 vectors showed substantial dose-dependent binding of FLAG antibody whereas no binding was seen in response to the controls, Ad5 or Ad48. There was also no binding with Ad48-pIX-gp83 as expected ( Figure 3A). In reference to His6 antibody, there was substantial

T. cruzi Antigens, Incorporated within pIX, Are Exposed on the Virion Surface
It is known that Ads can assemble and function with a defected/deleted pIX capsid protein [57]. For the modified pIX capsid incorporated vectors, we performed ELISA assays to verify that pIXs were intact and the T. cruzi antigens were accessible on the vectors' surface ( Figure 3). Serial diluted vectors were immobilized in the wells of an ELISA plate and incubated with either His 6 or FLAG-HRP antibody. In reference to FLAG antibody, Ad5-pIX-ASP-C, Ad48-pIX-ASP-C, and Ad5-pIX-gp83 vectors showed substantial dose-dependent binding of FLAG antibody whereas no binding was seen in response to the controls, Ad5 or Ad48. There was also no binding with Ad48-pIX-gp83 as expected ( Figure 3A). In reference to His 6 antibody, there was substantial dose-dependent binding of His 6 to Ad5-pIX-gp83, Ad48-pIX-ASP-C, and Ad48-pIX-gp83 whereas no binding was seen in response to the controls, Ad5 or Ad48 ( Figure 3B). The data demonstrates that all modified pIX vectors assembled with functional pIX and the T. cruzi epitope incorporation was accessible. dose-dependent binding of His6 to Ad5-pIX-gp83, Ad48-pIX-ASP-C, and Ad48-pIX-gp83 whereas no binding was seen in response to the controls, Ad5 or Ad48 ( Figure 3B). The data demonstrates that all modified pIX vectors assembled with functional pIX and the T. cruzi epitope incorporation was accessible.

Cellular Immune Response of Mice Immunized with Modified Ad5 and Ad48 Vectors
It is well documented that the CD8 + T-cell population plays an important role against resistance in parasitic infections [58][59][60], so it is imperative that modified Ad48 vectors elicit a strong cell-mediated immune response. To examine the T-cell profile generated by the Ad-modified vectors, C57BL/6 mice were immunized intramuscularly with equal amounts (1 × 10 10 VP/mL) of Ad5, Ad48, Ad5-CMV-ASP, or Ad48-CMV-ASP vectors according to the immunization schedule depicted in Figure 4A. Splenocytes from immunized mice were subjected to flow cytometry analysis. Splenocytes were stimulated with mitogen, gp83 peptide, or ASP-C peptide, gated for CD4 + and CD8 + phenotype, and analyzed for dual effector molecule secretion of T-cell degranulation marker, CD107a, and intracellular effector cytokines IFNγ and TNFα. Representative histograms of the double-positive gating for IFNγ+CD107a + cells are shown in Figure 4B. There was no significant differences in the frequencies of peptide-specific CD4 + T-cells that mobilized CD107a to their surface and expressed IFNγ or TNFα compared to the control groups (Ad5 or Ad48) ( Figure  4C). However, there was a significant difference with CD8 + T-cells. The Ad5-CMV-ASP-2

Cellular Immune Response of Mice Immunized with Modified Ad5 and Ad48 Vectors
It is well documented that the CD8 + T-cell population plays an important role against resistance in parasitic infections [58][59][60], so it is imperative that modified Ad48 vectors elicit a strong cell-mediated immune response. To examine the T-cell profile generated by the Ad-modified vectors, C57BL/6 mice were immunized intramuscularly with equal amounts (1ˆ10 10 VP/mL) of Ad5, Ad48, Ad5-CMV-ASP, or Ad48-CMV-ASP vectors according to the immunization schedule depicted in Figure 4A. Splenocytes from immunized mice were subjected to flow cytometry analysis. Splenocytes were stimulated with mitogen, gp83 peptide, or ASP-C peptide, gated for CD4 + and CD8 + phenotype, and analyzed for dual effector molecule secretion of T-cell degranulation marker, CD107a, and intracellular effector cytokines IFNγ and TNFα. Representative histograms of the double-positive gating for IFNγ+CD107a + cells are shown in Figure 4B. There was no significant differences in the frequencies of peptide-specific CD4 + T-cells that mobilized CD107a to their surface and expressed IFNγ or TNFα compared to the control groups (Ad5 or Ad48) ( Figure 4C). However, there was a significant difference with CD8 + T-cells. The Ad5-CMV-ASP-2 immunized mice showed a significant difference in the frequencies of peptide-specific CD8 + T-cells that mobilized CD107a to the cell surface and expressed IFNγ (p < 0.05) or TNFα (p < 0.05) compared to the Ad5 immunized mice ( Figure 4D). We made a similar observation with the Ad48-CMV-ASP-2 immunized mice compared to Ad48 immunized mice. Figure 4D also indicated a significant difference in the frequencies of peptide-specific CD8 + T-cells that mobilized CD107a to the cell surface and expressed IFNγ (p < 0.05), but only an increasing trend in CD8 + T-cells that mobilized CD107a to the cell surface and expressed TNFα compared to the Ad48 immunized group. immunized mice showed a significant difference in the frequencies of peptide-specific CD8 + T-cells that mobilized CD107a to the cell surface and expressed IFN· (p < 0.05) or TNF΅ (p < 0.05) compared to the Ad5 immunized mice ( Figure 4D). We made a similar observation with the Ad48-CMV-ASP-2 immunized mice compared to Ad48 immunized mice. Figure 4D also indicated a significant difference in the frequencies of peptide-specific CD8 + T-cells that mobilized CD107a to the cell surface and expressed IFN· (p < 0.05), but only an increasing trend in CD8 + T-cells that mobilized CD107a to the cell surface and expressed TNF΅ compared to the Ad48 immunized group.   We preformed identical experiments with the pIX-ASP-C vectors and followed the schedule depicted in Figure 5A. Representative histograms of the double-positive gating for IFN·+CD107a + cells are shown in Figure 5B. There were no differences in the percentages of CD4 + +CD107a + IFN· + or CD4 + +CD107a + +TNF΅ + cells in Ad5-pIX-ASP-C immunized mice compared to the Ad5 immunized mice ( Figure 5C). There was not a significant difference of CD4 + +CD107a + IFN· + cells in Ad48-pIX-ASP-C immunized mice when compared to Ad48 immunized mice. Further analysis showed an increase of CD4 + +CD107a + TNF΅ + cells in Ad48-pIX-ASP-C immunized mice compared to Ad48 immunized mice (p < 0.05) ( Figure 5C). Ad5-pIX-ASP-C immunized mice showed an increasing trend of the frequency of CD8 + +CD107a + IFN· + and CD8 + +CD107a + TNF΅ + cells when compared to Ad5 immunized mice. Ad48-pIX-ASP-C immunized mice showed an increasing trend of the frequency of CD8 + +CD107a + IFN· + when compared to Ad48 immunized mice. In addition, of important note, Ad48-pIX-ASP-C immunized mice showed a significant increase in the frequency of CD8 + +CD107a + TNF΅ + cells (p < 0.01) when compared to the Ad48 immunized mice ( Figure 5D). Overall, all modified Ad vectors were able to improve the specific T. cruzi T-cell response when compare to the controls. When cross comparing the vectors, there was a significant increase in the frequency of CD8 + +CD107a + IFN· + T-cells (p < 0.05) and CD8 + +CD107a + TNF΅ + T-cells (p < 0.05) from the Ad48-CMV-ASP-2 compared to Ad5-CMV-ASP-2. There was no significant T-cell response between Ad5-pIX-ASP-C and Ad48-pIX-ASP-C. We preformed identical experiments with the pIX-ASP-C vectors and followed the schedule depicted in Figure 5A. Representative histograms of the double-positive gating for IFNγ+CD107a + cells are shown in Figure 5B. There were no differences in the percentages of CD4 + +CD107a + IFNγ + or CD4 + +CD107a + +TNFα + cells in Ad5-pIX-ASP-C immunized mice compared to the Ad5 immunized mice ( Figure 5C). There was not a significant difference of CD4 + +CD107a + IFNγ + cells in Ad48-pIX-ASP-C immunized mice when compared to Ad48 immunized mice. Further analysis showed an increase of CD4 + +CD107a + TNFα + cells in Ad48-pIX-ASP-C immunized mice compared to Ad48 immunized mice (p < 0.05) ( Figure 5C). Ad5-pIX-ASP-C immunized mice showed an increasing trend of the frequency of CD8 + +CD107a + IFNγ + and CD8 + +CD107a + TNFα + cells when compared to Ad5 immunized mice. Ad48-pIX-ASP-C immunized mice showed an increasing trend of the frequency of CD8 + +CD107a + IFNγ + when compared to Ad48 immunized mice. In addition, of important note, Ad48-pIX-ASP-C immunized mice showed a significant increase in the frequency of CD8 + +CD107a + TNFα + cells (p < 0.01) when compared to the Ad48 immunized mice ( Figure 5D). Overall, all modified Ad vectors were able to improve the specific T. cruzi T-cell response when compare to the controls. When cross comparing the vectors, there was a significant increase in the frequency of CD8 + +CD107a + IFNγ + T-cells (p < 0.05) and CD8 + +CD107a + TNFα + T-cells (p < 0.05) from the Ad48-CMV-ASP-2 compared to Ad5-CMV-ASP-2. There was no significant T-cell response between Ad5-pIX-ASP-C and Ad48-pIX-ASP-C.

Humoral Immune Response to Capsid-Modified Vectors
As stated in the introduction, pIX is an Ad minor capsid protein and has 1/3 the amount of copy numbers that the major capsid protein hexon contains [61]. For these set of experiments, the amount of vector used for immunization was increased from 1ˆ10 10 VP/mL to 3ˆ10 10 VP/mL. To determine the T. cruzi-specific antibody responses elicited by the gp83 vector, C57BL/6 mice were immunized intramuscularly with 3ˆ10 10 VP/mL of Ad5, Ad48, Ad5-pIX-gp83, or Ad48-pIX-gp83 vectors according to the immunization schedule depicted in Figure 6A. Sera were collected from the immunized mice and evaluated for antibodies against T. cruzi by ELISA. The peptide specific for gp83, identical to what is in the capsid locale of the respective vectors was bound to ELISA plates. The plates were then incubated with the immunized mice sera. The binding was detected with HRP-conjugated secondary antibody. The total amount of IgG in sera was calculated based on a mouse IgG dilution standard curve. There was no significant difference of IgG levels between mice immunized with Ad5-pIX-gp83 and mice immunized with Ad5 ( Figure 6B). There was also no significant difference of IgG levels between mice immunized with Ad48-pIX-gp83 and mice immunized with Ad48 ( Figure 6B); however, there was a significant increase in the total IgG between prime and boost of the Ad48-pIX-gp83 immunized mice (p < 0.05) ( Figure 6B). When cross-comparing the Ad5 and Ad48 groups, there was a significant increase of total IgG in the Ad48-pIX-gp83 immunized group compared to Ad5-pIX-gp83 immunized group (p < 0.05) ( Figure 6B).

Humoral Immune Response to Capsid-Modified Vectors
As stated in the introduction, pIX is an Ad minor capsid protein and has 1/3 the amount of copy numbers that the major capsid protein hexon contains [61]. For these set of experiments, the amount of vector used for immunization was increased from 1 × 10 10 VP/mL to 3 × 10 10 VP/mL. To determine the T. cruzi-specific antibody responses elicited by the gp83 vector, C57BL/6 mice were immunized intramuscularly with 3 × 10 10 VP/mL of Ad5, Ad48, Ad5-pIX-gp83, or Ad48-pIX-gp83 vectors according to the immunization schedule depicted in Figure 6A. Sera were collected from the immunized mice and evaluated for antibodies against T. cruzi by ELISA. The peptide specific for gp83, identical to what is in the capsid locale of the respective vectors was bound to ELISA plates. The plates were then incubated with the immunized mice sera. The binding was detected with HRP-conjugated secondary antibody. The total amount of IgG in sera was calculated based on a mouse IgG dilution standard curve. There was no significant difference of IgG levels between mice immunized with Ad5-pIX-gp83 and mice immunized with Ad5 ( Figure 6B). There was also no significant difference of IgG levels between mice immunized with Ad48-pIX-gp83 and mice immunized with Ad48 ( Figure 6B); however, there was a significant increase in the total IgG between prime and boost of the Ad48-pIX-gp83 immunized mice (p < 0.05) ( Figure 6B). When cross-comparing the Ad5 and Ad48 groups, there was a significant increase of total IgG in the Ad48-pIX-gp83 immunized group compared to Ad5-pIX-gp83 immunized group (p < 0.05) ( Figure  6B).   To determine if "Antigen Capsid-Incorporation" strategy would increase the humoral response to the T-cell specific Ad-modified vectors, C57BL/6 mice were immunized intramuscularly with 3 × 10 10 VP/mL of Ad5-pIX-ASP-C, Ad48-pIX-ASP-C, or the controls (Ad5 and Ad48) and followed the same protocol for ELISA as previously described. There was no significant difference between Ad5 immunized mice and Ad5-pIX-ASP-C immunized mice or between Ad48 immunized mice and Ad48-pIX-ASP-C immunized mice at prime or boost ( Figure 6C). When comparing the IgG levels between prime and boost within the Ad5 groups, there was no significant increase ( Figure 6C); however, there was a significant increase of anti-T. cruzi IgG levels detected at boost when comparing Ad48-pIX-ASP-C immunized mice to the IgG levels at prime (p < 0.05) ( Figure 6C).

Discussion
We have developed novel Ad vectors that have the potential to optimize Ad vaccine approaches. In this manuscript we constructed Ad vectors for the development of a Chagas vaccine. We evaluated a series of Ad vectors of two serotypes, Ad5 and Ad48, whereby antigens were expressed (transgene antigen approach) or antigens were presented via the antigen capsid-incorporation strategy.
This manuscript highlights a two pronged approach: (1) the use of a rare serotype vector Ad48 in combination with the (2) Antigen Capsid-Incorporation strategy. The Ad48 vector was utilized because this serotype vector possesses many attributes that would make it an attractive alternative platform vector as compared to Ad5-based vector [62]. Some of these attribute included, namely its low seroprevalence in humans [63], failure to interact with factor X and transduce the liver [64,65], and its predictable cytokine profile, a reflection of its high-level accumulation in the spleen [62]. In this study, we utilized the antigen capsid-incorporation strategy, incorporating T. cruzi gp83 or T. cruzi ASP-C into the pIX locales of Ad5 and Ad48, respectively. To our knowledge this is the first Figure 6. Antigen capsid-incorporation vector elicits an in vivo T. cruzi humoral immune response. (A) Study design: C57BL/6 mice (n = 7) were primed and boosted with 3ˆ10 10 VP of Ad5, Ad5-pIX-gp83, Ad5-pIX-ASP-C, Ad48, Ad48-pIX-gp83, or Ad48-pIX-gp83. Two weeks after the last immunization serum was collected for ELISA binding assays. Either 10 µM of gp83 peptide or the ASP-C peptide was bound to ELISA plates; (B) Post-prime and post-boost serum from mice immunized with the T. cruzi gp83 modified vectors; (C) Post-prime and post-boost serum from mice immunized with the T. cruzi ASP-C pIX-modified vectors. The amount of anti-gp83 and anti-ASP-C antibodies in the sera was expressed as the mean˘SEM. (*) = p ď 0.05.
To determine if "Antigen Capsid-Incorporation" strategy would increase the humoral response to the T-cell specific Ad-modified vectors, C57BL/6 mice were immunized intramuscularly with 3ˆ10 10 VP/mL of Ad5-pIX-ASP-C, Ad48-pIX-ASP-C, or the controls (Ad5 and Ad48) and followed the same protocol for ELISA as previously described. There was no significant difference between Ad5 immunized mice and Ad5-pIX-ASP-C immunized mice or between Ad48 immunized mice and Ad48-pIX-ASP-C immunized mice at prime or boost ( Figure 6C). When comparing the IgG levels between prime and boost within the Ad5 groups, there was no significant increase ( Figure 6C); however, there was a significant increase of anti-T. cruzi IgG levels detected at boost when comparing Ad48-pIX-ASP-C immunized mice to the IgG levels at prime (p < 0.05) ( Figure 6C).

Discussion
We have developed novel Ad vectors that have the potential to optimize Ad vaccine approaches. In this manuscript we constructed Ad vectors for the development of a Chagas vaccine. We evaluated a series of Ad vectors of two serotypes, Ad5 and Ad48, whereby antigens were expressed (transgene antigen approach) or antigens were presented via the antigen capsid-incorporation strategy.
This manuscript highlights a two pronged approach: (1) the use of a rare serotype vector Ad48 in combination with the (2) Antigen Capsid-Incorporation strategy. The Ad48 vector was utilized because this serotype vector possesses many attributes that would make it an attractive alternative platform vector as compared to Ad5-based vector [62]. Some of these attribute included, namely its low seroprevalence in humans [63], failure to interact with factor X and transduce the liver [64,65], and its predictable cytokine profile, a reflection of its high-level accumulation in the spleen [62]. In this study, we utilized the antigen capsid-incorporation strategy, incorporating T. cruzi gp83 or T. cruzi ASP-C into the pIX locales of Ad5 and Ad48, respectively. To our knowledge this is the first time ever an antigen has been incorporated with the pIX locus of Ad48. T. cruzi antigen incorporation was validated at the genomic level in these vectors by sequencing and PCR analysis [66]. Antigen was confirmed at the protein level and within the viral capsid by western blot analysis ( Figure 2) and ELISA analysis (Figure 3). This study herein, illustrated that the capsid incorporation of T. cruzi antigen is comparable between Ad5-modified vectors and Ad48-modified vectors (Figure 3). T cruzi antigens were also expressed within the deleted E1 region of Ad48 (Figure 2). To our knowledge this is the first time ever that Ad48 has been utilized for a Chagas vaccine vector. Expression of ASP-2 was also comparable between Ad5 and Ad48-modified vectors by means of protein quantitation compared to a standard.
Subsequently, these vectors were utilized in animal experiments to evaluate T. cruzi-specific humoral and T-cell responses. As shown in the data, we evaluated antigen-specific T-cell responses of conventional transgene expressing Ad5 and Ad48-modified vectors ( Figure 4). Our data illustrate that Ad48-CMV-ASP-2 vector generated higher magnitude responses with respect to CD8 + CD107a + (TNFα + and IFNγ + ) responses when compared to Ad5-CMV-ASP-2 vector ( Figure 4D). In addition, we evaluated cell-mediated responses generated after immunization with antigen capsid-modified vectors (Ad5-pIX-ASP-C or Ad48-pIX-ASP-C) ( Figure 5). After immunization with the Ad48-pIX-ASP-C vector, we observed significant T. cruzi-specific CD4 + T-cell and CD8 + T-cell responses ((*) = p ď 0.05, (**) = p < 0.01)). Whereas, immunization with Ad5-pIX-ASP-C did not show any T. cruzi-specific CD4 + and CD8 + responses ( Figure 5C,D). To our knowledge this is the first demonstration that Ad48 has been utilized to generate a T. cruzi-specific response. Furthermore, this is the first demonstration of T. cruzi-specific responses generated from the Ad48 vector in combination with the Antigen Capsid-Incorporation strategy.
Studies were performed to determine T. cruzi-specific antibody responses elicited by the gp83 or ASP-C vectors ( Figure 6). The data demonstrates that immunization with Ad48-pIX-gp83 elicits a superior and significant antibody response as compared to Ad5-pIX-gp83 at boost (*) = p ď 0.05 ( Figure 6B). Of note, additional data demonstrates that immunizations with Ad48-pIX-ASP-C elicits a significant antibody response after boosting; whereas, homologous boosting of Ad5-pIX-ASP-C does not increase antibody specific response. It is an interesting phenomenon that Ad48-pIX-ASP-C elicits a significant antibody response after boosting because the ASP-C epitope incorporated within the vector is a CD8 + T-cell epitope and there are no known B-cell epitopes in the ASP-C epitope. This validates that the antigen capsid-incorporation strategy can induce a robust antigen-specific humoral immune response ( Figure 6C). We speculate that the Ad48 vector provides better B cell help as compared to the blunted response observed in Figure 6C after priming and boosting with Ad5-pIX-ASP-C.
To our knowledge this is the first time ever an antigen has been incorporated within the pIX locus of Ad48. We have incorporated antigens within the pIX locale of Ad48 based on the plasticity of Ad5 pIX. In the future, we will explore the "Antigen Capsid-Incorporation" strategy in the context of additional Ad 48 minor (VI, VIII, IX, IIIa, and IVa2) and major (hexon, fiber, and penton) capsid proteins such as, hexon. Historically, the benefit of incorporating antigens within Ad5 hexon is that there is three times the amount of hexon monomers per Ad5 virion as compared to pIX. In theory, antigen incorporation within the hexon protein would yield a superior antigen-specific immune response as compared to pIX. However, incorporations within the pIX locale would be less detrimental to virion formation and stability as compared to antigen incorporation within hexon. As it relates to incorporations within the virion, pIX allows the incorporation of large epitopes/antigens [42][43][44]46,67], whereas hexon incorporation has a much smaller incorporation capacity [68]. However, a thorough examination of Ad48 hexon hypervariable regions will need to be assessed first to establish its utility for antigen incorporation.

Conclusions
Our data in this manuscript highlights a unique utility of rare serotype vector, Ad48. The Ad48 vector has not been utilized for T. cruzi immunizations prior to our study, herein.
The focus of this manuscript is primarily to address vector-based questions, exploring the "Antigen Capsid-Incorporation" strategy within the context of the Ad48 vector system. Further studies are necessary to determine efficacy of a multivalent Ad48-based T. cruzi vaccine in challenge models, as well as immunogenicity in human populations.