Alphavirus Replicon Particle Vaccine Breaks B Cell Tolerance and Rapidly Induces IgG to Murine Hematolymphoid Tumor Associated Antigens

Abstract

De novo immune responses to myeloid and other blood-borne tumors are notably limited and ineffective, making our ability to promote immune responses with vaccines a major challenge. While focus has been largely on cytotoxic cell-mediated tumor eradication, B-cells and the antibodies they produce also have roles in anti-tumor responses. Indeed, therapeutic antibody-mediated tumor cell killing is routinely employed in patients with hematolymphoid cancers, but whether endogenous antibody responses can be incited to blood-born tumors remains poorly studied. A major limitation of immunoglobulin therapies is that cell surface expression of tumor-associated antigen (TAA) targets is dynamic and varied, making promotion of polyclonal, endogenous B cell responses appealing. Since many TAAs are self-antigens, developing tumor vaccines that enable production of antibodies to non-polymorphic antigen targets remains a challenge. As B cell responses to RNA vaccines are known to occur, we employed the Viral Replicon Particles (VRP) which was constructed to encode mouse FLT3. The VRP-FLT3 vaccine provoked a rapid IgG B-cell response to this self-antigen in leukemia and lymphoma mouse models. In addition, IgGs to other TAAs were also produced. Our data suggest that vaccination with RNA viral particle vectors incites a loss of B-cell tolerance that enables production of anti-tumor antibodies. This proof of principle work provides impetus to employ such strategies that lead to a break in B-cell tolerance and enable production of broadly reactive anti-TAA antibodies as potential future therapeutic agents for patients with hematolymphoid cancers.

Keywords: B cell tolerance; VRP; alphavirus replicon particle; antitumor antibody; autoantigen; cancer vaccine; hematolymphoid tumor.

Figure 1

The mouse lymphoma model and leukemia model, and the VRP vaccination strategy. FLT3-expressing A20 tumor cells were inoculated subcutaneously on the back of BALB/c mice on day 0. Tumor-challenged mice were vaccinated on days 4 and 18 with 1 x 10⁶ IU of VRP-Ctrl or VRP-FLT3 on the footpad. Control mice received PBS (n= 5 per group). FLT3-expressing C1498 tumor cells were inoculated intravenously through tail vein of B6.SJL mice on day 0. Mice were vaccinated on day 4 and 18, and day 32 in one repeat of this model with 1-2.5 x 10⁶ IU of VRP-Ctrl or VRP-FLT3 on the footpad. The control mice received PBS (n= 5 or 10 per group). Blood samples were harvested on day 14, 28, and 35-42.

Figure 2

VRP-FLT3 vaccination elicits a FLT3-specific antibody response in a mouse lymphoma model. Mice challenged with A20-FLT3 tumor cells were vaccinated as described in Figure 1. (A, B) Blood samples harvested on days 14 and 28 were analyzed by flow cytometry for the frequencies of (A) lymphocytes and (B) CD19⁺ B cells. (C) Concentrations of total IgG in plasma on day 28 were determined by ELISA. Each symbol represents an individual mouse (n=5 per group). Bars represent the mean ± SD. Statistical significance was determined by one-way ANOVA, *P<0.05, **P<0.01. (D) Western blot of recombinant mouse FLT3 protein. Plasma collected on days 35-42 was 1:250 diluted and used for probing. Arrow indicates the FLT3 protein. Two representative samples per group are shown. Positive Ctrl represents FLT3 detection with commercial anti-FLT3 antibody. Panel at left represents MWM, with numbers indicating kDa for the bands shown. (E) Dot blot of recombinant mouse FLT3 protein. Protein was spotted in duplicate and probed with 1:500 diluted plasma collected on indicated time points. Two to three representative samples per group are shown. In both (D, E), fluorochrome-conjugated anti-mouse IgG secondary antibody was used for signal development.

Figure 3

VRP-FLT3 vaccination alters the composition of circulating B cells in a mouse lymphoma model. Mice challenged with A20-FLT3 tumor cells were vaccinated as described in Figure 1. Blood samples harvested on days 14 were analyzed by flow cytometry for the frequency of (A) GL7⁺ B cells, (B) CD93⁺ mature B cells and CD93^– non-mature B cells, or (C) CD93⁺ IgM⁺ CD23^– T1 B cells (n=5 per group). A representative sample of each group is shown in contour plot on the right. Bars represent the mean ± SD. Statistical significance was determined by one-way ANOVA, **P<0.01, ***P<0.001, ****P<0.0001.

Figure 4

VRP-FLT3 vaccination elicits a FLT3-specific antibody response in a mouse leukemia model. Mice challenged with C1498-FLT3 tumor cells were vaccinated as described in Figure 1. (A, B) Blood samples harvested on days 14 and 28 were analyzed by flow cytometry for the numbers of (A) lymphocytes and (B) CD19⁺ B cells. (C) Concentrations of total IgG and (D, E) levels of FLT3-specific IgG in day 28 plasma samples were determined by ELISA. The titration curves (D) and O.D. values at 1:500 dilution (E), indicated as the vertical dashed line in (D), of anti-FLT3 IgG ELISA are shown. Each symbol represents an individual mouse (n=5 per group). Bars represent the mean ± SD. Statistical significance were determined by one-way ANOVA, *P<0.05, **P<0.01, ****P<0.0001. Data represents one of two independent experiments.

Figure 5

IgG2b and IgG2c are the predominant subclasses of FLT3-specific IgG induced by VRP-FLT3 vaccine. Antigen-specific IgG1 (A), IgG2b (B), IgG2c (C), and IgG3 (D) responses were detected by ELISA using the recombinant FLT3 as the capture antigen. (A–D) Titration curves of samples was shown. One sample in PBS and VRP-Ctrl groups was tested, and ten samples in VRP-FLT3 groups were tested. (E) Levels of anti-FLT3 IgG subclasses were compared by calculating the equivalent IgG subclass concentration using O.D. values at 1:300 dilution of anti-FLT3 IgG1 and IgG3 ELIS as indicated as the vertical dash line in (A) and (D), and O.D. values at 1:2700 dilution of anti-FLT3 IgG2b and IgG2c indicated as the vertical line in (B, C) ELISAs. Each symbol represents an individual mouse (n=10). Bars represent the mean ± SD. Statistical significance were determined by one-way ANOVA, *P<0.05, **P<0.01, ***P<0.001. Data represents one of two independent experiments.

Figure 6

VRP-FLT3 vaccination induces broad reactivity of antibody to mouse leukemia cell membrane antigens. Mice challenged with C1498-FLT3 tumor cells were vaccinated as described in Figure 1. Plasma samples collected on days 35-42 were diluted and were incubated with (A) C1498-mock cells or (D) C1498-FLT3 at a level of 1 ug total IgG per reaction. The tumor cell-bound IgG was detected by a fluorophore-conjugated secondary anti-mouse IgG antibody using flow cytometry. Heatmaps of MFI of each sample are shown (n=8 or 10 per group). Statistical significance was determined by Kruskal-Wallis test (nonparametric) ANOVA, *P<0.05. A representative sample of each group is shown in Heatmaps on the right (2^nd antibody alone in shade grey and samples in blue line). (B, C) Plasma IgG bound to C1498-mock membrane extracts were determined by ELISA. Days 35-42 plasma samples were tested. (B) Titration curves (symbols represent the mean values in each group). (C) O.D. values obtained with 1:25 diluted plasma were normalized by the IgG concentration of each sample (O.D. / total IgG concentration). Each symbol represents an individual mouse (n=7 per group). Bars represent the mean ± SD. Statistical significance was determined by one-way ANOVA, **P<0.01.

Figure 7

VRP vaccination attenuates the growth of FLT3-expressing A20 tumor cells and FLT3-expressing C1498 tumor cells. (A) Mice challenged with A20-FLT3 tumor cells were vaccinated as described in Figure 1. Tumor area of individual mice are depicted (n=5 per group). Lines represent the mean value of each groups. Statistical significance was determined by one-way ANOVA, *P<0.05, ^**P<0.01. (B) Inoculation of FLT3-expressing C1498 tumor cells and the vaccination schedule are as described in Figure 1. C1498-FLT3 tumor cells in peripheral blood were determined by Flow cytometry. Each symbol represents an individual mouse (n=5 per group). Bars represent the mean ± SD. Statistical significance was determined by one-way ANOVA, *P<0.05.