HIV/SIV DNA vaccine combined with protein in a co-immunization protocol elicits highest humoral responses to envelope in mice and macaques

Abstract

Vaccination with HIV/SIV DNAs elicits potent T-cell responses. To improve humoral immune responses, we combined DNA and protein in a co-immunization protocol using in vivo electroporation in mice and macaques. DNA&protein co-immunization induced higher antibody responses than DNA or protein alone, or DNA prime/protein boost in mice. DNA&protein co-immunization induced similar levels of cellular responses as those obtained by DNA only vaccination. The inclusion of SIV or HIV Env gp120 protein did not impair the development of cellular immune responses elicited by DNA present in the vaccine regimen. In macaques, the DNA&protein co-immunization regimen also elicited higher levels of humoral responses with broader cross-neutralizing activity. Despite the improved immunogenicity of DNA&protein co-immunization, the protein formulation with the EM-005 (GLA-SE) adjuvant further increased the anti-Env humoral responses. Dissecting the contribution of EM-005, we found that its administration upregulated the expression of co-stimulatory molecules and stimulated cytokine production, especially IL-6, by dendritic cells in vivo. These terminally differentiated, mature, dendritic cells possibly promote higher levels of humoral responses, supporting the inclusion of the EM-005 adjuvant with the vaccine. Thus, DNA&protein co-immunization is a promising strategy to improve the rapidity of development, magnitude and potency of the humoral immune responses.

1. Introduction

Numerous HIV vaccine modalities have been tested in different animal models and 3 phase III efficacy trials have been completed. A combination vaccine consisting of recombinant Canarypox ALVAC^®-HIV together with gp120 Env protein (AIDSVAX^® B/E) used in RV144 clinical trial in Thailand [1] resulted in modest protection from HIV-1 infection, which correlated with anti-Env binding antibodies (bAb), especially IgG antibodies against the V1V2 region [2,3]. No evidence of vaccine-induced virus control was found in the individuals who became infected. Using the macaque model and repeated low dose SIV virus challenge, different vaccine regimens showed delay virus acquisition [4–9] and reduction of viremia [4,5,7,8]. Although correlates of immune protection against HIV infection are still unclear, an efficient vaccine should elicit both cellular and humoral responses. We and others have previously shown that vaccination of macaques with optimized SIV/HIV DNAs induces strong, long-lasting humoral and cellular immune responses [8,10–16], which are able to control SIV/SHIV viremia [11,12,14–16]. Although DNA vaccines induce robust immunity, responses have not provided sterilizing protection. To enhance immunogenicity, other vaccine strategies used a prime/boost regimen with plasmid DNA followed by viral vector or protein boost [reviewed in Refs. [17–19]]. DNA priming and Env protein boost provided protection from homologous challenge in macaque SHIV models [20,21] and was immunogenic in humans [22,23]. In the present work, we evaluated the immunogenicity of a HIV/SIV vaccine consisting of DNA or protein only immunization or of DNA and gp120 Env protein as a boost or upon co-immunization. We also tested the contribution of EM-005 (GLA-SE) [24] as adjuvant and showed that its inclusion further increased the humoral responses obtained with the DNA&protein co-immunization regimen in macaques.

2. Materials and methods

2.1. Vaccine

Expression-optimized [25,26] endotoxin-free DNA vectors for SIV Gag and Env, HIV-1 Env and IL-12 and the gp120 proteins HIV_BaL and SIV_mac251 (transmitted clone M766) are described (see Supplementary Methods).

2.2. Vaccination of mice

BALB/c mice were vaccinated with 10μg of DNA mixture containing Env and Gag plasmids in 50μl PBS injected IM followed by in vivo electroporation (EP) (Ichor Medical Systems, Inc., San Diego, CA). Two μg of gp120 Env proteins were formulated with 5μg EM-005, an oil-in-water emulsion containing a TLR-4 agonist, as adjuvant in a 50μl volume [24] and injected by conventional IM injection following the DNA electroporation, at the same sites as DNA.

2.3. Vaccination of rhesus macaques

Macaques were immunized by IM injection with 1mg plasmid DNA followed by EP (ICHOR Medical Systems, Inc.) with a DNA vaccine mixture consisting of SIV Gag DNA, HIV Env DNA and IL-12 DNA formulated in 1 ml PBS. The HIV-1 gp120 protein was formulated in 100μl PBS alone or adjuvanted with 20μg of EM-005 and delivered by conventional IM injection at the same sites as DNA.

2.4. Immunological assays

Cellular responses in splenocytes and the phenotype and cytokine production of dendritic cells (DCs) were monitored using flow cytometry, and humoral immune responses were monitored for bAb and neutralizing (NAb) antibodies (see Supplementary Methods).

3. Results

3.1. DNA&protein co-immunization elicits high levels of humoral immune responses in mice

Different vaccination strategies with DNA and protein, either alone or in combination using a standard prime/boost or a co-immunization regimen, were compared in mice (N= 5 per group) after two immunizations (Fig. 1A). The DNA component of the vaccine was a mixture of plasmids encoding SIV Gag and SIV or HIV-1 Env, whereas the protein component consisted of SIV or HIV gp120 Env adjuvanted in EM-005 (GLA-SE). Blood was collected and assessed for the presence of Env and Gag bAb responses. DNA only, gp120/EM-005 only, or the DNA prime/protein boost immunization regimens yielded similar Env bAb titers (Fig. 1B, upper panel). In contrast, the DNA&protein/EM-005 co-immunization regimen elicited significantly higher (5–10 fold) responses. A vaccine that contained SIV gp160 Env DNA and gp120 protein/EM-005 instead of the HIV components also showed that the co-immunization regimen induced significantly higher bAb titers than any of the other vaccine regimens (Fig. 1C, upper panel). Both studies further showed that the DNA&protein co-immunization elicited early higher humoral responses, with differences detectable even two weeks after the 1st immunization (data not shown). Thus, DNA&protein co-immunization regimen is more effective in rapidly inducing high levels of humoral immune responses.

Fig. 1.

DNA&protein co-immunization induces highest Env bAb in vaccinated mice. (A) Mice (N= 5) were immunized twice (week 0 and 4) with DNA only; protein only; co-immunized with DNA&protein; or received a DNA prime/protein boost (DNA/protein). Blood was collected 2 weeks after the 1st and the 2nd vaccinations (V) and the mice were sacrificed at week 6. (B) Humoral immune responses of mice that received SIV gag and HIV gp160 env DNA and/or HIV gp120 protein. Env-specific (upper panel) and SIV Gag-specific (bottom panel) bAb were measured at 2 weeks after 2nd vaccination. Endpoint antibody titers of individual mice (mean±SEM) from a representative out of 3 independent experiments are shown. Mean±SEM and P values are indicated (Mann–Whitney). (C) Humoral immune responses of mice that received SIV gag and gp160 env DNA and/or SIV gp120 protein. SIV Env-specific (upper panel) and Gag-specific (bottom panel) bAb were measured as described for panel B. Endpoint antibody titers from individual mice from a representative out of 2 independent experiments are shown. Mean±SEM and P values are indicated (t test).

We also measured humoral responses to Gag, which was a component of all the DNA vaccine mixtures, but was not given as protein (Fig. 1B and C, bottom panels). Similar levels of Gag bAb were detected 2 weeks after the 1st vaccination in all the groups that received the DNA vaccine (data not shown). At 2 weeks after the 2nd vaccination, the DNA-only and DNA&protein co-immunization groups showed similar anti-Gag bAb responses, indicating that the presence of the EM-005 adjuvanted Env protein did not impair the development of humoral immune responses to Gag. As expected, the DNA prime/protein boost group showed lower Gag bAb levels since these animals did not receive a 2nd DNA vaccination. Together, the DNA&protein co-immunization (HIV and SIV vaccine platforms) was the most efficient regimen, inducing the highest Env bAb titers without impairing the Gag bAb responses.

3.2. DNA and protein co-immunization elicits higher level of humoral immune responses in macaques

Encouraged by the mouse data, we tested the co-immunization regimen in rhesus macaques and compared the humoral responses after 2 vaccinations using the same mixture of DNAs expressing SIV Gag and HIV Env, either alone or combined with HIV gp120 Env protein (Fig. 2A). Animals (N= 3/group) were immunized with DNA only or DNA&protein using 100μg gp120 protein in the absence of adjuvant. Co-immunization with HIV gp120 protein induced significantly higher bAb responses compared to the DNA only (Fig. 2B, upper panel), while similar levels of Gag-specific bAb titers were found in both groups (bottom panel), indicating that the presence of Env protein in the vaccine did not affect the induction of Gag-specific responses. These findings support the data from the mouse studies and show that the co-immunization protocol elicits robust humoral immune responses to Gag and Env. No impairment in the development of either of the responses was noted.

Fig. 2.

DNA&protein co-immunization of macaques. (A) Macaques (N= 3) were immunized twice (week 0 and 4) and 2 weeks after the 2nd vaccination, plasma was collected for endpoint bAb titer determination. The animals were vaccinated with DNA only or with the DNA&protein co-immunization regimen using gp120 alone or adjuvanted in EM-005. (B) Endpoint titers of Env-specific (upper panel) and Gag-specific (lower panel) bAb are shown. The p values from the one-way ANOVA analysis with Bonferroni’s multiple comparisons test are given.

We further examined whether the inclusion of the EM-005 adjuvant was able to further increase the robust humoral responses obtained using the DNA&protein co-immunization protocol (Fig. 2B, upper panel). We tested the contribution of the EM-005 adjuvant using 2 doses of BaL gp120 protein (100μg and 20μg). Immunization with the 100μg-dose of adjuvanted protein showed a significant increase (∼5x) in bAb titers, whereas intermediate levels of bAb titers were noted using the lower Env dose. The presence of Env protein in the vaccine in the absence or presence of the EM-005 adjuvant did not affect the development of Gag specific bAb responses (Fig. 2B, bottom panel). Analysis of the neutralization capacity of the antibodies showed that DNA&protein regimen, especially the group that received the higher dose of adjuvanted gp120 protein, developed NAb to several clade B Env proteins and showed broader cross-neutralizing activity to clade A and clade C Env (Table 1). No NAb to tier 2 Envs were induced after two vaccinations.

Table 1.

DNA&protein co-immunization with EM-005 adjuvanted gp120 induces broader NAb.

Group	Vaccine	Animal	Clade B				Clade C	Clade A	Negative control
			SF162.LS	BaL.26	SS1196.1	Bx08.16	MW965.26	DJ263.8	MuLV
1	DNA	A6X018	<20	<20	<20	23	22	<20	28
		A5X026	<20	<20	<20	<20	24	<20	<20
		A5X007	<20	<20	<20	<20	29	<20	<20
2	DNA&BaLgp120 (100μg)	A6X013	31	<20	22	20	61	<20	<20
		A4X018	<20	<20	<20	<20	36	<20	<20
		A4X010	64	<20	35	48	160	26	30
3	DNA&BaLgp120 (100μg) in EM-005	A4E056	398	98	83	147	498	49	35
		A4E055	65	26	29	43	88	30	<20
		A4X017	95	34	30	48	146	20	25
4	DNA&BaLgp120 (20μg) in EM-005	A6X006	36	34	41	56	146	40	24
		A5X012	248	46	49	65	257	24	<20
		A4X049	26	21	<20	20	52	<20	<20

In conclusion, the macaque and mouse studies show consistent results and demonstrate significantly higher Env humoral responses upon DNA&protein co-immunization. The macaque study further showed a positive effect of EM-005 in combination with the efficient DNA delivery by EP, indicating the importance of inclusion of an appropriate adjuvant to maximize the humoral immune responses in non-human primates.

3.3. DNA and protein co-immunization generates polyfunctional cellular immune responses in mice

We next investigated the development of cellular immune responses in mice that received the different regimens described in Fig. 1A. The frequency of antigen-specific cellular responses was determined in splenocytes stimulated with peptide pools covering HIV-1 Env (Fig. 3) and SIV gag, respectively (Fig. 4), and cytokine production was monitored by ICS staining followed by flow cytometry.

Fig. 3.

DNA and protein co-immunization elicited Env-specific polyfunctional cellular responses in mice. Splenocytes were isolated from immunized mice (described in Fig. 1A) at V2wk2. Sham DNA immunized mice served as negative control. The HIV Env-specific IFN-γ positive T cells were analyzed by flow cytometry. (A). The antigen-specific T cells (upper panel) and the distribution of IFN-γ positive cells in CD4⁺ and CD8⁺ T cells (bottom panel) are shown for a representative animal from each vaccination group. (B) The mean±SEM of Env-specific IFN-μ⁺ T cells (left panel) and the distribution in CD4⁺ and CD8⁺ T cells (right panel) for each group of mice is shown, from one of three representative experiments. (C) Mean frequency (±SEM) of Env-specific cells with 1 function (IFN-γ⁺ or TNF-α⁺), 2 functions (IFN-γ⁺TNF-α⁺, IFN-γ⁺CD107a⁺, TNF-α⁺CD107a⁺) or 3 functions (IFN-γ⁺TNF-α⁺CD107a⁺) among the CD4⁺ and CD8⁺ T cells is shown. (D) Pie charts show the distribution of Env-specific cells with 1 (blue), 2 (red) or 3 (green) functions among the CD4⁺ and CD8⁺ T cells. P values (Mann–Whitney test) are given. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

Fig. 4.

Gag-specific cellular responses induced by DNA alone and DNA&protein co-immunization in mice. Splenocytes (described in Fig. 3) were stimulated with SIV gag peptide pool. (A) The mean±SEM of gag-specific IFN-γ⁺ T cells (left panel) and the distribution in CD4⁺ and CD8⁺ T cells (right panel) for each group of mice is shown, from one of three representative experiments. (B) Mean frequency (±SEM) of Gag-specific cells with 1 function (IFN-γ⁺, TNF-α⁺), 2 functions (IFN-γ⁺TNF-α⁺, IFN-γ⁺CD107a+, TNF-α⁺CD107a⁺) or 3 functions (IFN-γ⁺TNF-α⁺CD107a⁺) in CD4⁺ T cells is shown. (C) Pie charts show the distribution of Gag-specific cells with 1 (blue), 2 (red) or 3 (green) functions among the CD4⁺ T cells. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

Mice vaccinated with DNA-only or co-immunized with DNA&protein developed similar levels of Env-specific IFN-γ⁺ T cell responses (Fig. 3A, B), indicating that the presence of the EM-005 adjuvanted protein in the co-immunization regimen did not impair the development of cellular immune responses. As expected, immunization with the DNA prime/protein boost regimen showed lower cellular immune responses, since a single DNA immunization was given. Immunization with protein only induced the lowest levels of cellular immunity. The antigen-specific responses were mainly mediated by CD4⁺ T cells. The DNA only and DNA&protein co-immunization groups, which received 2 DNA vaccinations, had increased levels of Env-specific CD8⁺ T cell responses (Fig. 3B). As expected, splenocytes from mice immunized with sham DNA did not respond to peptide stimulation (Fig. 3A and B, left panel).

We examined the poly-functional profile of the Env-specific T cells (Fig. 3C) by measuring their cytokine production (IFN-γ and TNF-α) and degranulation potential (CD107a) by flow cytometry. Mice that received two vaccinations (DNA only and DNA&protein co-immunization) showed similar pattern of responses including single, 2-function and 3-function within both the CD4⁺ and CD8⁺ T cell populations. The co-immunization group had a higher frequency of CD4⁺ T cells with 3 functions than the other groups and this difference was statistically significant compared to the DNA prime/protein boost group (with a single DNA dose) which elicited lower T cell responses with reduced multi-functionality. The distribution of Env-specific multifunctional T cells for all the groups that received DNA as a vaccine component is shown in the pie charts (Fig. 3D). The distribution of IFN-γ, TNF-α and multi-functionality among CD8⁺ T cells was similar both in the DNA only and the co-immunization group. Similar results were obtained in studies using SIV Env DNA combined with SIV gp120 protein (data not shown). These results show that co-immunization with DNA and protein did not impair the induction of multifunctional cellular immune responses.

3.4. Gag-specific cellular responses are not impaired in DNA and protein co-immunization protocol in mice

We also determined whether the co-immunization protocol affected the development of cellular immune responses to Gag, a second antigen that was part of the DNA vaccine mixture but that was not provided as protein. The level of Gag-specific IFN-γ⁺ T cells was similar among the groups that received 2 DNA immunizations (Fig. 4A), and showed that co-administration of the adjuvanted Env protein did not affect the development of Gag-specific cellular responses, excluding the possibility of a general T cell activation induced by the adjuvant included in the vaccine. This conclusion is further supported by the fact that the protein-only group did not show any response upon stimulation with Gag peptides (Fig. 4A). As expected, the levels of Gag-specific responses were lower in the DNA prime/protein boost group, which only received a single DNA vaccine at prime. The Gag-specific IFN-γ⁺ T cells showed mainly a CD4⁺ phenotype in all three groups. Analysis of the multi-functionality of the Gag-responses demonstrated similar frequency among the groups that received the DNA twice, including cells with 2 or 3 functions. The mice from the DNA prime/protein boost had lower degranulation potential (Fig. 4B). The pie charts show that the distribution of 1, 2 and 3 functions between the antigen-specific T cells is similar among the groups (Fig. 4C). These results demonstrate that co-immunization with DNA (Gag and Env) and protein (Env) did not affect cellular responses induced by an antigen (Gag) included only as DNA.

3.5. EM-005 adjuvant enhances the maturation of dendritic cells in mice

We further dissected the mechanism by which EM-005 contributed to the increase in humoral immunity. We hypothesized that the EM-005 may stimulate the maturation of dendritic cells (DCs), thereby enhancing their functional capabilities for antigen presentation. To explore possible effects of EM-005 on the function of DCs, mice were immunized with HIV-1 Env DNA, gp120 protein or DNA&protein in the presence or absence of EM-005 as adjuvant (Fig. 5A). Two days later, spleens and draining lymph nodes were collected and the phenotype and cytokine production of DCs was analyzed by flow cytometry. Vaccine regimens that did not contain the adjuvant (Fig. 5A) showed a DC phenotype similar to that of untreated mice (data not shown). In contrast, vaccine regimens that included the EM-005 adjuvant (EM-005 only, protein formulated in EM-005 or DNA&protein/EM-005) resulted in DCs with a more mature phenotype, characterized by increased expression of the co-stimulatory molecules CD40, CD80 and CD86. As representative examples, we show (Fig. 5B, upper panel) the expression profiles of the co-stimulatory molecules in DCs obtained from mice that received protein only (gray area), DNA&protein (dashed line), EM-005 only (gray line) and DNA&protein/EM005 (bold black line). The percent of CD40⁺, CD80⁺ and CD86⁺ cells in MHC II⁺ DCs from mice immunized with EM-005 only or DNA&protein/EM-005 was significantly higher than that of mice immunized with protein only or DNA&protein (Fig. 5B, lower panel). In addition, we found that DCs obtained from mice injected with EM-005 produced significantly higher level of IL-6 in lymph nodes (Fig. 5C, left panel) and spleens (Fig. 5C, right panel). Therefore, these data suggest that EM-005 increased the immunogenicity of the adjuvanted protein by enhancing the maturation and cytokine production of DCs.

Fig. 5.

The maturation and cytokine production of DCs by EM-005 adjuvant. (A) Mice (n = 3 or 4) were immunized with sham DNA, HIV-1_BaL gp160 DNA only, HIV-1_BaL gp120 protein only, protein/EM-005, DNA&protein, DNA&protein/EM-005 and EM-005 only. Two days after the injection, inguinal LN and spleens were isolated. (B) The maturation of DCs in inguinal LN was analyzed by flow cytometry. The overlays (upper panel) show the expression of the co-stimulatory molecules CD40, CD80 and CD86 in DCs identified as CD3−, CD11c⁺MHC II⁺. Protein only (gray filled area), DNA&protein (dashed line), DNA&protein/EM-005 (black line) and EM-005 only (gray line) are shown as selective examples. The lower panels show the percent of CD40⁺, CD80⁺ and CD86⁺ cells in MHC II⁺ DCs. C) The IL-6 production of DCs in inguinal LN (left panel) and spleen (right panel). Lymphocytes were incubated overnight in the presence of Monensin and IL-6 production was monitored by intracellular staining and flow cytometry. The CD11c⁺ population in CD3⁻ cells was considered to represent DCs. Mean frequency (±SEM) of IL-6⁺ DCs in inguinal LN and spleen is shown. One-way ANOVA was used to analyze the difference between each of the vaccines that lacked (open symbols) or included (filled symbols) EM-005. *P < 0.05.

4. Discussion

The goal of this study was to identify an HIV vaccine platform effective in eliciting robust and balanced immunity combining both cellular and antibody responses. We compared vaccination protocols that used DNA, protein, or a combination of DNA and protein delivered either in co-immunization regimen or using a classical prime/boost regimen as outlined. The rationale for this strategy is that DNA vaccination elicits strong cellular immunity, whereas the induced antibody responses are lower than those obtained with a protein vaccine. In contrast, protein vaccination provides strong antibody response but less robust cellular immunity. Therefore, we hypothesized that a combination of these vaccine components may be better suited to trigger both arms of the immune system.

Here, we report that the DNA&protein co-immunization induces significantly higher humoral immune responses than DNA alone, and that co-immunization did not impair cellular immunity resulting in the development of polyfunctional cellular immune responses in mice. Similarly, we found that the co-immunization regimen elicited higher humoral responses in macaques, including broader HIV-1 cross-neutralizing antibodies. Interestingly, incorporation of the synthetic EM-005 adjuvant led to further significant increase of the humoral responses in vaccinated monkeys despite the use of the very efficient DNA EP delivery method. These results support the concept of DNA&protein co-immunization as promising strategy for the development of an effective HIV-1 vaccine, and show that inclusion of an adjuvant contributes to maximally augment the humoral responses. Several studies have shown that the EM-005 adjuvant enhances the humoral responses induced by protein immunogens [27–32]. Here, we show that using one of the most effective DNA delivery methods (in vivo electroporation) together with protein co-immunization still benefits from the presence of this adjuvant. Addressing the mechanism for this enhancement, we found that the expression of co-stimulatory molecules was upregulated ex vivo in DCs from mice that received EM-005. DCs with more mature phenotype might migrate more efficiently into secondary lymphoid organs and enhance antigen presentation. Importantly, DC secretion of IL-6, which has been associated with B cell activation and antibody production [33], was significantly increased by the EM-005 adjuvant. These data support the conclusion that the DC activation observed in mice immunized with DNA&protein/adjuvant is due to the effect of EM-005 and not to other components of the co-immunization regimen. Increasing the maturation/activation status of DCs, like EM-005 does, is the most common feature of all TLR ligands [34], and the most likely mechanism responsible for their adjuvant activity.

In our study, co-immunization not only induced the highest Env bAb titers but also elicited strong cell-mediated responses. These findings are in agreement with those obtained in a recent macaque study where vaccines combining DNA with simultaneous or sequential administration of inactivated viral particles, as a source of protein, showed that the co-immunization protocol induced the highest humoral responses [8]. Improved humoral immunity induced by needle/syringe delivered DNA&protein co-immunization compared to the individual components has been reported for a non-HIV vaccine [35], but it was also noted that co-immunization resulted in active immune suppression leading to impaired cellular responses [35–38]. Subsequent works testing some non-HIV vaccines showed that DNA&protein co-immunization resulted in T cell anergy secondary to induction of inhibitory Treg responses via regulatory DCs [39,40], and, therefore, it was proposed as a method to induce tolerance in autoimmune and allergic diseases [41–44]. In contrast, DNA&protein co-immunization was also reported to greatly augment humoral responses in other systems [45–47] with no impairment of cellular immunity [48]. Combining Dengue DNA vaccine with particles in a co-immunization protocol demonstrated higher antibody titers, which associated with protection from Dengue viremia in immunized mice [49]. In agreement with the latter, we recently reported that the co-delivery of electroporated DNA and inactivated viral particles resulted in robust immune responses without any evidence of tolerance, because the vaccinated macaques resisted infection and controlled chronic virus replication [8]. Using the mouse model with DNA and purified Env protein as vaccine components, we report superior induction of humoral immunity with no negative effect on the cellular immune responses using the co-immunization regimen. We hypothesize that the different DNA delivery methods, (needle/syringe [35,36] versus IM/EP (this report), the amounts of DNA and protein used or the nature of immunogens could contribute to the discrepancies in these findings. In our DNA&protein co-immunization protocol using electroporation, we did not observe induction of Treg cells (data not shown) or impairment of cellular responses compared to DNA only group. While our studies were on-going, Jaworski et al. [50] reported that HIV DNA&protein co-immunization was superior to vaccination with either of the two individual components in eliciting humoral immune responses in rabbits and mice. Our results are in overall agreement and we further expand these findings to the rhesus macaque model, which is considered the most relevant for human vaccination. Jaworski et al. [50] reported that DNA&protein co-immunization also enhanced antigen-specific cellular immune responses compared to either DNA or protein alone in mice. In contrast, our results show that DNA&protein co-immunization induced similar levels of cellular immune responses compared to DNA and that these levels are higher than those found in the protein only group. It is possible that the efficiency of the different DNA delivery methods used in these studies (conventional intramuscular injection for DNA delivery [50] versus in vivo electroporation) could contribute to this.

We have previously shown that DNA&protein co-immunization induces robust immune responses in the absence of exogenous adjuvant [8]. Since the protein component was provided as inactivated viral particles at low concentration, we could not exclude an adjuvant effect provided by other particle components. Here, we show that the DNA&protein co-immunization regimen using purified protein in the absence of adjuvant elicits higher humoral responses and that these responses can be further augmented using the EM-005 adjuvant. DNA&protein co-immunization not only elicits high Env bAb titers but it also induces polyfunctional cellular immune responses. Importantly, this strategy induces robust levels of antibodies with cross-clade neutralizing activity in the macaque model. The DNA&protein co-immunization platform provides a promising HIV-1 vaccine candidate with potentially improved efficacy.