3.2. Humoral Immune Responses to SUDV Proteins and Neutralization Profiles
We assessed the profile of IgG immune-reactivity using a custom made ELISA previously described [18
]. Survivor serum samples were analyzed against SUDV recombinant proteins NP, VP30, VP40, GP1-649
, and irradiated purified whole virus antigen. The results presented in Figure 1
A–E) and Table 1
demonstrate that in the 15 SUDV survivors tested, 11 displayed positive antibody-reactivity to NP, nine to GP1-649,
and VP40, and four to VP30. Six survivors exhibited a positive antibody response against the irradiated SUDV whole purified virus antigen. Serum from uninfected controls displayed no reactivity to any of the viral antigens tested. A plaque reduction neutralization test (PRNT) was performed to determine the neutralization capacity of survivor sera at various dilutions (Figure 1
F and Table 1
). We defined neutralization as the ability of a given serum sample, regardless of dilution, to neutralize 80% of SUDV plaque formation (PRNT80
) relative to controls. The results demonstrate that of the 15 SUDV survivors tested, sera from 6 survivors displayed a PRNT80
capacity. The remaining nine survivors and five uninfected controls were non-neutralizing by the applied definition.
3.3. Flow Cytometry
Flow cytometry analysis of IL-4, TNFα and IFNγ cytokine levels was performed following cell stimulation with SUDV inactivated whole antigen. Based on the humoral reactivity of serum samples observed in both the ELISA and PRNT80
assays (Table 1
), we chose to group the survivors into the following categories for analytic purposes: those who have no immune-reactivity against SUDV GP1-649
or inactivated SUDV and were PRNT80
negative (Ab−/Neut−); those who react against SUDV GP1-649
and/or inactivated SUDV but were PRNT80
negative (Ab+/Neut-); and those who react against SUDV antigens and also have a PRNT80
capacity against live SUDV in vitro
(Ab+/Neut+) (Table 1
). Uninfected controls were also included and grouped separately.
Flow cytometry data showing IFNγ and TNFα expression by cells after stimulation with SUDV antigen for all survivors and controls is presented in Figure 2
A. All plots shown are gated hierarchically as follows: lymphocytes (FSC-A vs.
SSC-A), singlets (FSC-A vs.
FSC-H), live cells (Aqua live/dead negative), CD3+ (CD3 vs.
SSC-A), and CD4+ CD8− (Figure 2
B). Resting values for each survivor or control are included in each quadrant in parentheses (Figure 2
A). Cells from uninfected control samples did not express any IFNγ or TNFα in response to SUDV stimulation. One survivor in the SUDV Ab−/Neut− group had a diverse CD4 T cell response, comprised of IFNγ and TNFα-single positive as well as double positive cytokine-producing cells (Figure 2
A). All other Ab−/Neut− and Ab+/Neut− survivors demonstrated an absence of cytokine production in response to SUDV stimulation. In stark contrast, survivors with PRNT80
serum capacity against SUDV in vitro
all had multivariate cytokine responses (Figure 2
A). The extent of response varied from survivor to survivor but overall, IFNγ and TNFα double-positive (DP) responses predominated. IFNγ single positive (SP), TNFα SP, and DP frequencies in the Ab+/Neut+ were significantly higher than all other groups for each subset of cytokine-producing cells (p
< 0.05) (Figure 2
C). Flow cytometry data of IL-4 cytokine levels following cell stimulation with SUDV inactivated whole antigen showed no detectable signals in both survivors and controls.
A correlation analysis between cytokine expression and neutralization activity (at 1:80 dilution) demonstrated a significant correlation between the cytokine and neutralization responses (Table 2
). The correlation between neutralization and IFNγ, TNFα DP and TNFα SP cytokine responses was slightly higher than for IFNγ SP responses (Table 2
). Only two of the 15 survivors had apparent CD8 T cell responses to SUDV (Supplemental Material Figure S1
) consisting of IFNγ and TNFα expression. Due to the rarity of these responses, we were unable to do further analysis.
We also performed a memory subset analysis for survivors that had SUDV-specific CD4 T cell cytokine responses, regardless of whether or not they had SUDV-specific antibodies or serum neutralization capacity. We used CD62L and CD45RO in this analysis to identify central memory T cells (TCM
, CD45RO+ CD62L+),
effector memory T cells (TEM
, CD45RO+ CD62L−), terminally-differentiated effector memory T cells (TEMRA
(CD45RO− CD62L−), and naïve T cells (CD45RO−, CD62L+). Each cytokine-producing subset was overlaid upon the parent CD4 T cell population for reference. Virtually all IFNγ SP and IFNγ, TNFα DP cells were in the TEM
subset (Figure 3
A) with two survivors (upper right, lower left) that also had TEMRA
cells. Only TNFα SP cells were found in the TCM
subset (Figure 3
A). As CD62L can be cleaved from the cell surface upon T cell activation, we performed an ELISA for soluble CD62L (sCD62L) in the supernatants of these cultures. We were unable to establish a pattern of increased sCD62L in cultures that corresponded to cytokine production, which would have indicated that CD62L had been cleaved due to T cell activation (Figure 3
B). sCD62L levels were variable regardless of stimulation or resting culture for all controls and survivors (Figure 3
3.4. Cytokine and Chemokine Levels in Whole Blood Stimulation by Multiplex ELISA
In addition to flow cytometric analysis, multiplex ELISA was performed on supernatants to evaluate secreted cytokines and chemokines following whole blood stimulation using SUDV. During this assay, PHA-L (Phytohaemagglutinin Leukocytes) stimulation was used as a positive control for immune cell responses, and unstimulated samples served as baseline controls. The unstimulated control results demonstrated that SUDV survivors exhibited equivalent baseline levels of cytokines production compared to healthy, uninfected control patients, and PHA-L stimulation resulted in equally robust cytokine expression in all patient groups.
Whole blood stimulation results with irradiated SUDV elicited significantly elevated secretion of IL-2, IFNγ, IP-10 and MCP-2 in Ab+/Neut+ survivors, relative to unstimulated, resting cultures (Figure 4
A,C,F,G). IFNγ, IP-10 and MCP-2 secretion levels in Ab+/Neut+ survivors were also significantly increased relative to the other survivor groups tested. SUDV stimulation demonstrated no significant increase in secretion levels of IL-5, TNFα, GROα and Eotaxin. Multiplex ELISA results of IL-1β, IL-4, IL-6, IL-10, IL-17, IL-23, I-309, and MCP-1 showed no differences of cytokine levels or detectable cytokine signals in survivors and control groups. Further analysis between percent neutralization (at 1:80 dilution) and cytokine and chemokine secretion levels following stimulation revealed significant correlation to IL-2 (p
= 0.0378), IFNγ (p
= 0.0011), IP-10 (p
= 0.0006), and MCP-2 (p
= 0.0008) expression (Table 2
). The complete multiplex ELISA results for the individual survivors and uninfected controls are presented in Table 3