Intramuscular administration of influenza hemagglutinin in one hind leg elicits similar protective immune responses as administration in two hind legs Authors: Naoko Uno, Ted M Ross Published: 2026 ABSTRACT AbstractThe objective of this study was to determine if distributing vaccine antigen across two intramuscular vaccination sites, rather than a single site could elicit more efficient protective immune responses. Most commercial injectable vaccines are administered unilaterally in a single location. The hypothesis of this study was vaccination in two distinct intramuscular sites would produce a more robust immune response compared to a single injection site. Mice vaccinated with a bilateral vaccination strategy may elicit broader antigen dissemination leading to increased germinal center activation and higher protective immune responses. Mice were vaccinated in the left and/or right hind leg quadricep muscle with recombinant H1 and/or H3 computationally optimized broadly reactive antigen (COBRA) hemagglutinin influenza protein and then challenged with either an H1N1 or H3N2 influenza virus. Hemagglutination inhibition (HAI) assays were performed on sera collected after final vaccination. There were no significant statistical differences in HAI titers between mice receiving a single HA protein (monovalent) in one versus two legs. The lowest antibody response was observed in mice receiving the bivalent mixture in a single injection site. Similarly, B cells collected from the spleen or draining lymph nodes had anti-HA IgG levels that were not significantly improved following bilateral vaccination compared to single-site vaccination. Furthermore, there were no statistically significant differences in weight loss or lung viral titers following influenza virus challenge in vaccinated mice, regardless of the vaccine regimen. These results indicate that administering intramuscular vaccinations in two locations does not augment immune responses compared to a single site administration.Keywords: Influenza, Hemagglutinin, Intramuscular vaccination, Humoral immunity Introduction Most licensed vaccines are administered via intramuscular (IM) route in a single site in the body. The preference for this method of vaccination is due to convenience, lower reactogenicity, and established dosing regimen (Cook, 2008; Gutierrez et al., 2023) as opposed to a mechanistic understanding of how injection site distribution may influence immune response. After IM vaccination, antigen drainage is restricted to the regional draining lymph nodes (dLNs), where adaptive immune responses are activated (Dupuis et al., 1998). The site of immunization elicits LN engagement, germinal center activity, and humoral immune generation (Cirelli et al., 2017; Turner et al., 2020). Increasing the number of vaccine administration sites may augment protective immune responses. Most seasonal influenza vaccines are typically administered intramuscularly, with effectiveness varying each year, ranging from 10 to 60% protective efficacy (Lewis et al., 2022). Vaccines have to be reformulated annually, largely due to the frequent mutations of the surface glycoprotein hemagglutinin (HA) resulting in antigenic drift (Carrat et al., 2007). The vaccine strains are mismatched due to this variability and result in weaker levels of immune protection against infection. (Altman et al., 2018). Computationally optimized broadly reactive antigen (COBRA) methodology has been developed to address this issue. Next generation recombinant HA proteins have been designed for different influenza subtypes by producing layered consensus sequences from thousands of human clinical isolates to capture conserved epitopes (Giles et al., 2011). In preclinical studies, recombinant COBRA-based H1 and H3 HA proteins elicit antibody response that recognize many viral strains over multiple seasons (Allen and Ross, 2022, 2025; Huang et al., 2021). After intramuscular vaccination, antigen-presenting cells (APCs) capture and process the antigen at the site of injection and migrate to nearby dLN and present antigen to naïve CD4+ T cells. This interaction initiates the differentiation of these cells into T follicular helper (Tfh) cells, which are an important component in B cell activation and differentiation (Sage et al., 2015). Activated B cells can either differentiate into short-lived plasmablasts that secrete early antibodies or enter germinal centers (GCs) to undergo somatic hypermutation, affinity maturation, and class-switching (Schmidlin et al., 2009; Mesin et al., 2016; Chen et al., 2023). GC B cells can become long-lived plasma cells that reside in the bone marrow and sustain antibody production or memory B cells that can respond more rapidly upon re-exposure to the antigen (Slifka et al., 1998; Ochsenbein et al., 2000). Initial immune responses are localized within the dLN at the site of vaccine administration (Gonzalez et al., 2011) then activated lymphocytes will exit the dLNs and disseminate systemically to secondary lymphoid organs such as the spleen (Jiang et al., 2017). Increasing vaccine administration to two distinct intramuscular sites may activate multiple dLNs simultaneously, increasing the number of APCs, Tfh cells, and GC B cells, leading to a more robust humoral immune response. In this study, bilateral intramuscular vaccination of COBRA H1 and/or H3 HA proteins in two locations were compared to unilateral vaccination at a single site in mice. Antibody responses, B cell activation, and protection against viral infection were analyzed to determine if increasing the distribution of antigen would increase immune responses. Materials and methods Viruses and antigens Viruses used for assays or infections were obtained from Virapur (San Diego, CA, USA), Center for Disease Control and Prevention (Atlanta, GA, USA), or BEI Resources (Manassas, VA). All seed stocks were propagated in specific pathogen free embryonated chicken eggs, and single-use aliquots were stored in −80 °C. H1N1 strains: A/California/07/2009 (CA/09), A/Michigan/45/2015 (Mich/15), A/Brisbane/02/2018 (Bris/18), A/Guangdong-Maonan/SWL1536/2019 (GD/19), A/Victoria/2570/2019 (Vic/22). H3N2 strains: A/Switzerland/9715293/2013 (Swiz/13), A/Kansas/14/2017 (KS/17), A/Hong Kong/2671/2019 (HK/19), A/Massachusetts/18/2022 (Mass/22), A/District of Columbia/27/2023 (DC/23). Antigens used for vaccinations and ELISPOT were produced as described previously (Ecker et al., 2020). Y2 and NG5 were chosen for H1 and H3 vaccinations, respectively. These two antigens have been previously produced in our lab using computationally optimized broadly reactive antigen (COBRA) methodology and have been shown to elicit broad protection against multiple strains compared to wild type HA (Allen et al., 2025; Huang et al., 2021). Y2 HA sequence was generated by COBRA methodology using circulating H1 strains from 2016 to 2018 (Huang et al., 2021). NG5 HA sequence was created by COBRA methodology using circulating H3 strains from 2018 to 2021 (Allen et al., 2025). Phylogenetic trees were generated using the Geneious tree builder (Geneious Bioinformatices Software), and multiple rounds of consensus sequencing were implemented to produce a single H1 (Y2) or H3 (NG5) sequence. These sequences were cloned into plasmid vectors and transfected into HEK293T cells to produce trimerized, soluble proteins as previously described (Ecker et al., 2020). Mouse vaccinations and infections All animal procedures were approved by the University of Georgia's Institutional Animal Care and Use Committee (IACUC) (IACUC approval #A2024 06-001-Y1-A1). DBA/2J female mice (6–8 weeks old) were purchased from Jackson Laboratories (Bar Harbor, ME, USA). Mice were housed in microisolation cages and given access to food and water freely. Mice were randomly divided into eight vaccine groups (n = 14 per group) and given intramuscular vaccinations of next generation influenza recombinant proteins: Y2 and NG5 HA protein; One Leg group was vaccinated in the left leg with a mixture of 3 μg of Y2 and 3 μg of NG5; Y2 One Leg group was vaccinated in the left leg, with 3 μg of Y2; NG5 One Leg group was vaccinated in the right leg, with 3 μg of NG5; Y2, NG5 One Leg Each group was vaccinated with Y2 in the left leg and NG5 in the right leg, with 3 μg of each antigen; Y2 and NG5 Two Legs group was vaccinated with a mixture of Y2 and NG5 in the left and right legs with 1.5 μg of each antigen in each leg; Y2 Two Legs group was vaccinated in the left and right legs, with 1.5 μg of Y2 per leg; NG5 Two Legs group was vaccinated in the left and right legs, with 1.5 μg of NG5 per leg. Each vaccine group was adjuvanted with AddaVax (InvivoGen, San Diego, CA, USA) at 1:1 vol ratio, with 25 μL of antigen and 25 μL of AddaVax- 50 μL total volume per leg. One week after final vaccination, spleens and lymph nodes (n = 3) were harvested from each group. Popliteal and inguinal lymph nodes were collected from the left and right sides of each mouse. Splenocytes and lymphocytes were processed and frozen in 10% DMSO + B Cell Media (RPMI w/L-glutamine +10% FBS +50 μM β-mercaptoethanol). Mice were infected with lethal doses (10LD50) of H1N1 A/Brisbane/02/2018 (3.6 × 104 PFU) or H3N2 mouse-adapted A/Switzerland/9715293/2013 (5 × 106 PFU) four weeks after final vaccination (n = 11 per group). Total volume of 50 μL were administered intranasally. Weights and clinical signs were monitored daily. Clinical signs were scored by following criteria: Weight loss between 15% and 25% = 1, lethargy = 1, dyspnea = 2, weight loss greater than 25% = 3, severe respiratory distress = 3, unresponsiveness to external stimuli = 3. Mice with cumulative clinical scores (CS) of 3 or above indicated reaching humane endpoint and were euthanized. Lungs were harvested on day 3 (n = 3) and day 6 (n = 3) for quantifying viral load in lungs (Fig. 1). Diagram of vaccination timeline and groups. (A) Naïve DBA2/J mice were vaccinated on day 0, 28, and 56, spleens and lymph nodes taken on day 63, mice were challenged on day 84, and lungs were taken on days 87 and 90. (B) There were eight different vaccine groups (n=14 per group). Hemagglutination inhibition assay Mouse sera were treated with receptor-destroying enzyme (Denka Seiken, Tokyo, Japan) at 1:3 ratio according to manufacturer's instructions. Hemagglutination inhibition (HAI) activity of sera samples were determined as previously described (Allen et al., 2022). Briefly, sera were diluted two-fold starting at 1:10 dilution and incubated with virus at 1:8 HA titer for 30 min in v-bottom 96-well plates. 0.8% turkey red blood cells were added for H1N1 influenza viruses and incubated for 30 min or 0.8% guinea pig red blood cells +20 nM oseltamivir caboxylate were added for H3N2 influenza viruses and incubated for 1 h. After incubation, plates were tilted and the serum titer at which the blood did not agglutinate was determined. The reciprocal of the highest serum dilution was identified as HAI titer. ELISPOT Mouse spleens and lymph nodes collected one week after final vaccination was analyzed for IgG antibody secreting cells (ASCs). For single cell suspensions, spleens were disrupted using gentleMACS Dissociator (Miltenyi Biotec, Waltham, MA, USA), filtered through 70 μm MACS SmartStrainers (Miltenyi Biotec), centrifuged at 1000 rpm for 5 min at RT, and resuspended in B cell media (RPMI+10% FBS+50 μM βME+100 U/mL Pen/Strep). Lymph nodes were disrupted manually using sterile glass slides, filtered through 70 μm MACS SmartStrainers, centrifuged at 1000rmp for 5min at RT, and resuspended in B cell media. Mouse IgG ELISPOT (CTL, Shaker Heights, OH, USA) assay was performed following manufacturer's instructions. Briefly, ELISPOT plates were first coated with Y2, NG5, or positive control (IgG capture antibody) and incubated overnight at 4 °C. Plates were washed with PBS and splenocytes (n = 3) or lymphocytes (n = 3, pooled) were added at 3 × 105 cells per well and serially diluted. The cells were incubated for 6 h at 37 °C with 5% CO2. Cells were washed off and plates were incubated with IgG detection antibody and incubated overnight at 4 °C. Plates were washed with PBS+0.5% Tween-20 and incubated with tertiary antibody at RT in the dark for 1 h. Plates were washed with DI water and substrate was added for 7 min at RT in the dark. After color development, plates were washed with tap water and left to dry overnight. Spots were scanned and counted using the ImmunoSpot S6 Ultimate Analyzer (CTL). Viral plaque assays Mouse lungs were harvested on days 3 and 6 post infection. Supernatant from homogenized lungs were incubated with MDCK cells to determine viral lung titer as previously described (Uno et al., 2024). Statistical analysis Statistical differences in HAI responses were done using Two-Way ANOVA with multiple comparisons. Statistical differences in weight loss curves, ELISPOT splenocyte responses, and lung viral titers were done using Mann-Whitney test. Analyses performed by GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). A p value of less than 0.05 was defined as statistically significant (*, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001). ELISPOT lymphocytes were pooled, so statistical analysis was not performed. Results Assessment of serological responses following vaccination The goal of this study was to determine if vaccinating a bivalent rHA vaccine in the same leg was as effective at eliciting protective antibodies against H1N1 and H3N2 influenza viruses as mice receiving the same vaccines in two different legs. Naïve DBA/2J mice were given intramuscular vaccinations of Y2 and/or NG5 with AddaVax adjuvant in one or two legs in prime-boost-boost fashion (Fig. 1). For Y2 and NG5 One Leg group, mice were given a mixture of 3 μg Y2 and 3 μg of NG5 in the left leg only; for Y2 One Leg group, mice were given 3 μg of Y2 in the left leg only; for NG5 One Leg group, mice were given 3 μg of NG5 in the right leg only; for Y2, NG5 One Leg Each group, mice were given 3 μg of Y2 in the left leg and 3 μg of NG5 in the right leg; for Y2 and NG5 Two Legs, mice were given a mixture of 1.5 μg of Y2 and 1.5 μg of NG5 in the left leg and 1.5 μg of Y2 and 1.5 μg of NG5 in the right leg; for Y2 Two Leg, mice were given 1.5 μg of Y2 in the left leg and 1.5 μg of Y2 in the right leg; for NG5 Two Leg, mice were given 1.5 μg of NG5 in the left leg and 1.5 μg of NG5 in the right leg (Fig. 1B). Antibody responses following the final vaccination were analyzed for hemagglutination inhibition (HAI) activity (Fig. 2). Mice vaccinated with Y2 HA protein had, on average, seroprotective HAI titers (≥1:40) against all 5 H1N1 influenza virus strains, regardless of if the Y2 HA was administered in one or two legs or administered in combination with the NG5 HA protein (Fig. 2A). Y2 HA administered in the left leg only or both right and left legs had no statistically significant differences in HAI titers against any of the H1N1influenza virus strains. However, mice vaccinated with a mixture of Y2 and NG5 HA proteins in one leg had serum HAI titers against the panel of H1N1 strains that were 2–4-fold lower than mice vaccinated with any of the other vaccinated groups of mice (Fig. 2A). This group had statistically lower HAI titers compared to Y2 only groups for CA/09, all other vaccine groups for MI/15, and Y2 only and Y2 or NG5 in one leg groups for BR/18. Hemagglutination Inhibition (HAI) response 14 days after final vaccination. Naïve DBA/2J mice were vaccinated as shown in Figure 1. Virus strains are shown on the x-axis in chronological order. HAI titer is shown on the y-axis, with bottom dashed line indication 1:40 HAI titer and top dashed line indicating 1:80 HAI titer. (A) H1 HAI responses were analyzed for groups that contained Y2 in their vaccine formulation. (B) H3 HAI responses were analyzed for groups that contained NG5 in their vaccine formulation. Statistical differences were calculated between vaccine groups using two-way ANOVA with multiple comparisons were performed by GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). A p value of less than 0.05 was defined as statistically significant (*, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001). For mice vaccinated with NG5 HA, there were no significant differences in HAI activity in mice administered NG5 protein in one leg compared to mice vaccinated in two legs (Fig. 2B). Mice vaccinated with a mixture of Y2 and NG5 proteins in one leg had the lowest HAI titers against the panel of H3N2 influenza viruses compared to mice vaccinated with other vaccine regimens (Fig. 2B). Mice vaccinated with a mixture of Y2 and NG5 proteins had statistically lower HAI titers against KS/17 and HK/19, as well as mice vaccinated with NG5 protein in one leg against HK/19 compared to mice vaccinated with the NG5 protein in two legs. Vaccine elicitation of anti-HA antibody secreting cells Spleens and draining lymph nodes collected from vaccinated mice were assessed for anti-HA antibody secreting cells (ASC). Mice vaccinated with Y2 protein in one leg had similar anti-Y2 HA ASCs as mice vaccinated with Y2 protein in two legs (Fig. 3A). Mice vaccinated with Y2 protein mixed with NG5 protein administered to one leg had similar number of ASCs observed as mice vaccinated with Y2 in the left leg and NG5 in the right leg (Fig. 3A). Both groups of mice that received bivalent vaccination regimen had lower number of ASCs compared to mice administered a mixture of Y2 and NG5 in both legs. Mice vaccinated with NG5 protein only had low to undetectable levels of Y2 HA-specific ASCs. The highest responses were detected in mice vaccinated with Y2 HA in two legs and mice administered Y2 and NG5 HA proteins in two legs. These two groups of mice had statistically higher serum HAI titers than mock vaccinated mice or mice vaccinated with NG5 only. Mice vaccinated with Y2 and NG5 proteins in one leg had the fewest number of ASCs detected against the Y2 HA. IgG B cell response in splenocytes after vaccination. Naïve DBA2/J mice were vaccinated as shown in Figure 1. IgG antibody secreting cells (ASCs) for (A) Y2 or (B) NG5 were enumerated from splenocytes harvested on day 63 (n=3). Statistical analyses between vaccine groups were done using Mann-Whitney test performed by GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). No significance between vaccine groups were found. These same mice were assessed for ASCs against the NG5 protein following vaccination (Fig. 3B). Mice that were vaccinated with a mixture of Y2 and NG5 HA in both legs had the highest number of ASCs than mice administered Y2 HA in left leg and NG5 HA in right leg (Fig. 3B). In addition, these mice also had higher number of anti-NG5 HA ASCs compared to mock vaccinated mice or mice vaccinated with Y2 HA only. Mice given Y2 HA in the left leg and NG5 HA in the right leg had the lowest ASC response compared to other groups with NG5 in their formulation. Mice that received a mixture of Y2 and NG5 HA proteins in the same leg had twice as many ASCs, on average, as mice that received these vaccines in separate legs, but had similar responses as mice that received the mixed vaccine in two legs (Fig. 3B). Mice that received NG5 protein only in one leg or in two separate legs had similar responses. To assess ASC responses in the draining ipsilateral proximal lymph nodes, left and right lymph nodes were examined separately in each group of vaccinated mice (Fig. 4). Mice vaccinated with Y2 HA in the left leg had predominately Y2-specific ASCs detected in the left lymph node, regardless if vaccinated alone or mixed with the NG5 HA protein (Fig. 4A). Mice vaccinated with the Y2 HA protein in both right and left legs had similar number of Y2-specific ASCs detected in the left lymph node (2800 spots) with lower, but substantial number of ASCs (1700) were detected in the right lymph node. Interestingly, mice vaccinated in both legs with a mixture of Y2 and NG5 HA proteins had higher number of Y2 HA-specific ASCs in the right lymph node with 3 times the number of ASCs detected in the right lymph node compared to the left lymph node. IgG B cell response in lymph nodes after vaccination. Naïve DBA2/J mice were vaccinated as shown in Figure 1. IgG antibody secreting cells (ASCs) for (A) Y2 or (B) NG5 were enumerated from lymphocytes harvested on day 63 (n=3, pooled). The x-axis indicates left (L) or right (R) lymph nodes from each group. These same mice were assessed for NG5 HA-specific ASCs in the lymph nodes following vaccination (Fig. 4B). Mice vaccinated in the right leg with NG5 HA protein alone had higher NG5 HA-specific ASCs in the right lymph node compared to the left lymph node. Similar results were observed in mice vaccinated with NG5 HA protein in the right leg and Y2 HA protein in the left leg or mice vaccinated with a mixture of Y2 and NG5 proteins in both legs. Mice vaccinated with NG5 HA protein in both legs had similarly high numbers of ASCs detected in both the right and left lymph node. Influenza virus challenge Mice were divided into two groups and challenged with lethal doses of either BR/18 (H1N1) or SW/13 (H3N2) influenza virus (Fig. 5). Mock vaccinated mice challenged with either BR/18 or SW/13 rapidly lost weight and all mice reached clinical endpoints between days 4–6 post-infection. All Y2 HA vaccinated mice survived BR/18 challenge with little weight loss (>5% of original body weight) and no clinical symptoms (Fig. 5A). In contrast, after SW/13 viral challenge, all mice vaccinated with NG5 HA protein in one leg or two legs rapidly lost weight by day 4 post-infection, but slowly began to recover, but did not return to full weight following 14 days of observation. There were no significant differences in weight loss between NG5 HA administered in one leg and NG5 HA administered in two legs. Weight loss after viral challenges. Naïve DBA2/J mice were vaccinated and challenged as shown in Figure 1. (A) Weight loss curve after lethal A/Brisbane/02/2018 (H1N1) challenge for groups (n=5) with Y2 in their formulation and mock. (B) Weight loss curve after lethal mouse-adapted A/Switzerland/2013 (H3N2) challenge for NG5 or mock groups (n=5). Statistical analyses between vaccine groups were done using Mann-Whitney test performed by GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). No significance between vaccine groups were found. Lung viral titers were determined at days 3 and 6 post-infection (Fig. 6). Mock vaccinated mice had high viral lung titers ~1 × 106 pfu/g following BR/18 influenza virus infection (Fig. 6A and B). There were no detected viral lung titers in any vaccinated mice following this infection (Fig. 6A and B). Mock vaccinated mice challenged with SW/13 had average viral lung titers of ~1 × 106 pfu/g, but these titers declined ~1 log by day 6 post-infection (Fig. 6C and D). Mice vaccinated with NG5 HA protein in the right leg only had ~3.5 logs lower viral titers (~7 × 102 pfu/g) on day 3 post-infection compared to mock vaccinated mice following SW/13 virus infection, whereas mice vaccinated in both legs had 1.5 logs lower viral titers (~2 × 105 pfu/g) compared to mock vaccinated mice (Fig. 6C). There were no detectable viral titers from either one leg or two leg group of NG5 vaccinated mice on day 6 post-infection (Fig. 6D). Lung viral plaque titers after viral challenges. Naïve DBA2/J mice were vaccinated and challenged as shown in Figure 1. Lung viral plaque titers were determined at (A) day 3 and (B) day 6 after lethal A/Brisbane/02/2018 (H1N1) challenge for groups with Y2 in their formulation and mock. Lung viral titers were determined at (C) day 3 and (D) day 6 after lethal mouse-adapted A/Switzerland/2013 (H3N2) challenge for NG5 or mock groups. Statistical analyses was done using Mann-Whitney test performed by GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). performed by GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). No significance between vaccine groups were determined. Discussion This study investigated whether distribution of recombinant hemagglutinin (HA) antigens in two intramuscular injection sites enhances humoral immunity and protective efficacy compared to conventional single-site administration. Although bilateral vaccination in two legs could potentially engage multiple draining lymph nodes (dLNs) and expand germinal center activity (Cirelli et al., 2017), two-site administration did not significantly improve immunological or protective outcomes against influenza virus in DBA/2J mice. For vaccination at a single location, HAI titers against H1N1 and H3N2 viral panels were comparable between mice administered vaccine in one-leg or two-legs (Fig. 2). Mice vaccinated with Y2 HA had seroprotective HAI titers (>1:40) against all five H1N1 strains, regardless of injection site distribution, and mice vaccinated with NG5 HA had no differences in HAI titers. This suggests that recombinant HA formulated with AddaVax within a single dLN network is sufficient to generate maximal systemic humoral response. Antigen-specific responses were initially localized to the dLN closest to the injection site (Fig. 4), with unilateral vaccinations activating ASCs in the ipsilateral nodes, while bilateral vaccinations activated both nodes. However, distribution of LN activation did not translate to enhanced systemic ASC response (Fig. 3) or elevated HAI titers. This suggests rapid systemic dissemination of activated lymphocytes following vaccination, which may make engaging additional ipsilateral dLNs unnecessary. The group vaccinated with a mixture of Y2 and NG5 HA in one leg consistently had the lowest immune responses compared to mice vaccinated with other vaccine regimens. This group had two to four-fold lower HAI titers against several H1 and H3 strains compared to mice vaccinated with other vaccine regimens (Fig. 2). Similarly, the lowest number of detectable anti-Y2 HA ACS were in these mice (Fig. 3A). Local antigenic competition or immunodominance effects may be a potential cause of lower antibody and B cell responses mice vaccinated in both legs compared to mice vaccinated with other vaccine regimens. Co-delivery of distinct antigens can result in suppression of immune responses to one antigen due to competition for APC uptake or T cell help (Zost et al., 2019). In contrast, separating antigen vaccinations into different legs (Y2 and NG5 HA in separate legs) or distributing a mixed formulation (Y2 and NG5 HA in two legs) mitigated reduced immune responses (Figs. 2 and 3). This suggests that distributing antigens may reduce local immune competition. Mice vaccinated in two legs did not have significantly improved protection against Influenza virus challenge. All mice vaccinated with formulations containing Y2 HA were fully protected against lethal H1N1 influenza virus challenge (Fig. 5A) with undetectable lung viral titers after infection (Fig. 6A and B). Mice vaccinated with NG5 HA alone or mixed with Y2 HA were protected from lethal SW/13 H3N2 influenza virus challenge (Fig. 5B). There were no statistical differences in weight loss between mice vaccinated in one leg and two legs. Mice vaccinated with NG5 HA in one leg had lower lung viral titers on day 3 post-infection compared to mice vaccinated with NG5 HA vaccinated in two legs (Fig. 6C), with both groups of mice clearing virus from the lungs by day 6 post-infection (Fig. 6D). These findings have important implications on how COBRA-based vaccines compare to conventional quadrivalent influenza vaccines (QIV). QIV formulations have four distinct HA antigens (H1, H3, and two IBV HAs) that are delivered to a single injection site (Tisa et al., 2016). This is similar to the Y2+NG5 one leg group. Our data suggest that some immune competition may occur, but this can be mitigated by distributing across two anatomical sites or delivering them as separate immunogens. However, it is unlikely to enhance overall magnitude of systemic antibody response or enhance protective efficacy. Additionally, the logistical and compliance challenges associated with multiple injections may outweigh potential immunological benefits. There were some limitations in this study. DBA/2J mice were used for their susceptibility to influenza virus infection (Alberts et al., 2010; Dengler et al., 2012), but their modified genetics reduces innate inflammatory responses to viral infection. Cellular immune responses were not fully characterized, so future studies analyzing Tfh and CD8+ T cell response will be necessary to identify immune differences in the vaccine regimens. Additionally, dosing studies using suboptimal antigen concentrations may reveal an improved immune response elicited by bilateral vaccination compared to vaccinating in a single leg, which was not observed in this study. Overall, the findings in this study demonstrate that intramuscular vaccination in two quadricep leg muscles does not significantly enhance systemic antibody responses or enhance the protective efficacy against influenza virus challenge compared to vaccination in a single leg in mice. However, spatial separation of distinct antigens may mitigate local antigenic competition observed in single-site bivalent formulations. These findings provide mechanistic insight into the relationship between antigen distribution and protective immunity. Funding statement This project has been funded as part of the Collaborative Influenza Vaccine Innovations Centers (CIVICs) by the National Institute of Allergy and Infectious Diseases, a component of the NIH, Department of Health and Human Services, under contract 75N93019C00052. TMR is also supported in part as Eminent Scholar by the Georgia Research Alliance, GRA-001. Footnotes Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. CRediT authorship contribution statement Naoko Uno: Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. Ted M. Ross: Funding acquisition, Project administration, Supervision, Writing – review & editing. Data availability Data is uploaded to NIH Import Data Sharing Site. References Associated Data Data Availability Statement Data is uploaded to NIH Import Data Sharing Site.