Studying the immunopathogenesis of Ebola virus disease using flow cytometry

Full Text

The disease caused by the Ebola virus disease (EVD) has been a constant threat to public health since it was first identified in 1976¹. Much attention has been paid to studying its pathogenesis and developing and testing vaccines. Flow cytometry (FC), an important method for studying the mechanisms of immune response development, has been used in this process. FC is a technology for rapidly measuring the characteristics of cells in suspension using monoclonal antibodies (mAbs) or other probes conjugated with a fluorescent label, based on light scattering and fluorescence parameters [1, 2]. The method allows hundreds of thousands of cells in suspension (events) to be recorded in a very short time, which ensures high accuracy of the results obtained. At the same time, the use of flow cytometry when working with Group I pathogens, which include EVD, is associated with a number of difficulties related to both sample preparation and compliance with safety requirements when working with a flow cytometer. In this regard, it is interesting to consider the features of the immunopathogenesis of the disease caused by EVD, the study of which has become possible thanks to the use of the FC method.

FC allows phenotyping of immune system cells, which is actively used in the study of the disease caused by EVD. Thus, when modeling infection in Javanese macaques (Macaca fascicularis) [3], assessment of the population composition of peripheral blood mononuclear cells (PBMCs) revealed the dynamics of T-lymphocyte depletion, as well as NK cells. The number of circulating B-lymphocytes remained constant. Furthermore, staining for surface markers allows us to identify not only the cell's belonging to a subpopulation, but also its state — assessment of the expression of activation markers (CD69, CD25, CD44) showed that the activation of T and B lymphocytes after exposure to EVD was insignificant [3]. Assessment of surface markers using FC allows the activation of specific subpopulations of peripheral blood mononuclear cells to be evaluated without the need for their isolation and subsequent in vitro cultivation. This significantly reduces the duration of the experiment and the number of manipulations, thereby increasing the safety of the studies conducted. Similar data were obtained in another study [4]. In this study, the authors also used FC to assess the level of apoptosis by performing intracellular staining of cells for caspase expression. The detection of increased expression of CD95 (Factor-related apoptosis receptor, FasR) on the cell surface suggested the involvement of FasR/FasL signaling in the induction of apoptosis. Simultaneous analysis of MHC-II, CD14, CD80 and CD86 expression on a single cell allowed us to identify MHC-II+ cells as dendritic cells and assess their degree of maturity and activation [5]. The results showed that EVD had a negative effect on dendritic cells and their activation. So, staining cells for several markers at once, including intracellular ones, lets you accurately tell apart different cell populations and check out their functional state.

Another example is combined staining for membrane markers: CD3 — T-cell marker; CD4 — T-helper marker; CD8 — cytotoxic T-lymphocyte marker; combination of CD28 and CD95 — memory cells, as well as intracellular staining for cytokines — interleukin (IL)-2, interferon (IFN)-γ, and tumor necrosis factor (TNF)-α allows the activation of T-cell subpopulations under the action of antigen and, accordingly, the content of antigen-specific CD4+ and CD8+ T-cells to be assessed. Assessment of this parameter is important because CD8+ T lymphocytes play an important role in protecting against EBV infection induced by the administration of the rAd5GP vaccine [6]. The main criterion for evaluating the antigen-specific T-cell immune response is the proportion of T-lymphocytes producing IFN-γ in response to antigenic stimulation [7], but measuring only one parameter does not allow the entire functional potential of T-cells to be reflected. Therefore, the assessment of intracellular production of three cytokines (IFN-γ, TNF-α, and IL-2) is considered more informative. IFN-γ and TNF-α are considered the main effector cytokines, while IL-2 does not have a direct antiviral or antibacterial effect, but stimulates the proliferation and clonal expansion of T cells, which leads to an enhanced immune response. Multifunctional IFN-γ+TNF-α+IL-2+ T lymphocytes are characterized by a long life cycle and the ability to produce large amounts of cytokines [7–10].

An alternative to FC for detecting these indicators may be the isolation of T-lymphocyte subpopulations, their specific activation in vitro, followed by assessment of cytokine levels in supernatants, which will significantly complicate the experiment and increase its duration. At the same time, the proportion of activated (antigen-specific) cells can also be assessed using fluorescence microscopy, but the sensitivity of this method is significantly inferior to FC. For the detection of cells that simultaneously express several markers, FC is virtually the only method available. Analysis of immune system cell subpopulations also allows for the assessment of the dynamics of the infectious process [11, 12], including the combination of immune system activation and suppression processes [13] in diseases caused by EVD.

After 2000, studies began to appear that used FC when working with samples obtained from humans. The need to use high-tech devices to perform FC limits the use of the method in the field and requires the organization of a mobile laboratory. At the same time, the number of parameters determined simultaneously directly depended on the level of the devices used. The individual dynamics of the infection were represented by isolated clinical cases, which allowed the authors to use the data only to confirm or refute the assumptions or theories about the pathogenesis of the disease.

A. Sanchez et al. published an article on the results of studying peripheral blood samples obtained from patients during an outbreak of infection caused by EVD (Sudan Ebola Virus, SUDV) in Uganda in 2000 [14]. The authors used FC to assess the dynamics of the ratio of T and B lymphocytes, NK cells, and the expression of Fas and FasL depending on the patient's condition, which gave reason to assume the development of immunosuppression during EVD infection. In 2007, M. Gupta et al. published a study on the results of FC analysis of PBMC infected with EVD in vitro [15]. The authors measured the expression of apoptosis markers Fas, FasL, Caspase-3, Bcl2, and annexin-V by CD4⁺ T cells, cytotoxic T lymphocytes, and monocytes. The authors showed that 70% of macrophages undergo apoptosis on the fourth day after infection and 30–40% of T-helpers on the eighth day. Among cytotoxic T lymphocytes on the 8th day after infection, the level of cells expressing Bcl2, active caspase-3, and stained with annexin V exceeded the other two populations by 2–3 times. Signals of apoptosis of the Fas⁺/FasL system were also detected on T cells and macrophages. Using the RNase Protection Assay [16], an increase in the expression of TRAIL mRNA, a cytokine of the TNF family (a ligand that induces apoptosis), was detected in CD4⁺ and CD8⁺ T cells on the 7th day after infection. Based on the data obtained, the authors concluded that EVD evades the immune response, causing massive lymphocyte death.

N. Wauquier et al. published the results of a FC study of nine PBMC samples cryopreserved during EVD outbreaks in Zaire (ZEBOV) in Gabon and Congo in 1996–2005 [17]. The CD4⁺ and CD8⁺ T-lymphocyte counts in 2 healthy individuals, survivors during the acute phase, and 3 survivors after recovery were comparable (43.6 and 22.4%, 46.2 and 24.1%, and 36.6 and 17.4%, respectively). In contrast, the proportion of T-lymphocyte populations in the three deceased patients was significantly lower. These data were consistent with the facts about massive lymphocyte death observed in studies on experimental infection of animals [3, 4] and PBMC in vitro [15]. CD3⁺CD4⁺CD95⁺ and CD3⁺CD8⁺CD95⁺ cells accounted for 54.1% and 75.8% of PBMC in ZEBOV fatalities, respectively, compared to 5.6% and 6.8% in 2 healthy individuals. CD3⁺CD4⁺CD95⁺ cells in survivors during the acute phase of ZEBOV infection accounted for 11%, while the relative content of CD3⁺CD4⁺CD95⁺ and CD3⁺CD8⁺CD95⁺ in three samples after recovery was 20.8% and 18.5%, respectively. Thus, those who died from the disease caused by EVD showed immunosuppression, manifested by the loss of peripheral CD4⁺ and CD8⁺ lymphocytes through Fas/FasL-mediated apoptosis.

In 2015, A. McElroy et al. published a report on the results of a study of the cellular and humoral immune responses of 4 people who survived EVD infection and were treated at Emory University from August to October 2014 [18]. CD4⁺ and CD8⁺ T cells were analyzed for expression of HLA-DR and CD38 activation markers. Antibody-secreting cells (plasma blasts) were identified by CD27 and CD38 expression on CD19⁺ cells (B lymphocytes). Phenotyping of activated CD8⁺ T cells was performed after the onset of symptoms to detect the expression of markers activated via the T cell receptor: Bcl-2l, CD127, granzyme B, PD-1, CX3CR1, CD45RA. Intracellular FC was used to measure the production of cytokines IFN-γ, TNF-α, and IL-2 by CD8⁺ and CD4⁺ T cells and to determine their antigen specificity [18]. Experimental treatment was initiated in all patients after the onset of symptoms. Thus, the initiation and activation of the adaptive immune response occurred under natural conditions without intervention. Phenotyping of the plasmoblast population showed that their peaks coincided with the generation of EVD-specific IgG, which indicated a sustained humoral response in all patients. It was also shown that most plasmablasts were IgG-positive, as demonstrated in other acute viral infections [19]. It is worth noting that EVD-specific plasmablasts were present in the blood even 2 months after the onset of symptoms.

A high percentage of activated CD8⁺ and CD4⁺ T cells was observed for 60 days after the onset of symptoms. This result contrasted with the responses of activated CD8⁺ T lymphocytes in other acute viral infections, where the percentage of activated T cells returned to baseline much more rapidly after recovery [20]. In the case of EVD-induced disease, the elevated T-cell content implied the presence of a viral antigen that continuously restimulated the immune response. It was suggested that the viral antigen most likely persists not in the blood but in other tissues, because in all patients, 23 days after the onset of symptoms, viral RNA could not be detected in the blood. The fact that prolonged activation was observed in patients with severe and moderate disease was consistent with the presence of high levels of viral antigen in their blood over a long period of time. Unlike patients with moderate and severe disease, patients with mild disease experienced a rapid decrease in viremia and activated T-cell populations.

Patients with mild disease did not have two peaks of activated CD4⁺ and CD8⁺ T cells. This phenomenon was observed in patients with severe and moderate disease, which, according to the authors, represented the return of tissue T cells to the peripheral blood from the affected organs [18]. The antigen specificity of CD4⁺ and CD8⁺ T cells was determined by the expression of IFN-γ and TNF-α in response to stimulation with pools of peptides from EVD proteins. The strongest responses were mediated by CD8⁺ T cells and directed against the EVD nucleoprotein. All patients had low levels of CD8⁺ T-cell responses to VP40. Only 2 of the 3 patients examined had a detectable CD8⁺ T-lymphocyte response against the viral glycoprotein (GP). The authors concluded that the virus nucleoprotein, in addition to GP, can generate a sustained T-cell response during vaccination [18].

In a study by S. Agrati et al., they evaluated the kinetics and functionality of T-cell subpopulations, as well as the expression of markers of activation, autophagy, apoptosis, and exhaustion during the acute phase and until recovery in patients treated for EVD infection at the National Institute of Infectious Diseases in Rome [21]. Male patients (patient No. 1, 51 years old, and patient No. 2, 37 years old) with EVD-induced disease were hospitalized on the 5th and 3rd days after the onset of symptoms, respectively [22, 23]. The clinical course of the disease in patient No. 1 was longer and more severe. Both patients experienced activation of Epstein-Barr virus (EBV) infection. Patient No. 1 was treated with high doses of favipiravir, administration of convalescent plasma, melanocortin, and the drug ZMab (Mapp Biopharmaceutical). Patient No. 2 was treated with favipiravir and Mill77, a cocktail of mAbs. Peripheral blood samples were collected at various times from admission to the hospital until recovery.

Patients showed a marked decrease in the relative content of CD4⁺ T lymphocytes to below 20% and a significant increase in CD8⁺ T lymphocytes during the viremia period, which was followed by a recovery phase accompanied by a persistent inversion of the CD4⁺/CD8⁺ T cell ratio. When determining the absolute number of CD4⁺ and CD8⁺ T lymphocytes, a decrease was initially observed, followed by an increase, which can be explained in patient No. 1 by treatment with melanocortin, and in patient No. 2 by a homeostatic response to virus-induced lymphopenia or a lymphoproliferative response to EBV reactivation [24]. According to the authors, a significant reduction in the CD4⁺ T-cell population is an important factor in the overall loss of lymphocytes and may limit the initiation and maintenance of effective humoral and cytotoxic T-cell immunity, leading to reversible immunosuppression.

In both patients, the content of activated CD38-expressing CD8⁺ T cells was low in the first days after infection and then increased, reaching a maximum on day 13 for patient No. 1 (79.6%) and on day 8 for patient No. 2 (79.5%), followed by a decline. A similar pattern was observed for CD4⁺ T cells. Analysis of HLA-DR and CD38 co-expression on CD8⁺ T cells revealed delayed activation and lower activation of CD4⁺ T cells (based on HLA-DR expression) in patient No. 1. Patient No. 1 had a longer CD8⁺ T-cell activation profile compared to patient No. 2, possibly due to longer HIV viremia. Most activated T cells did not express either CD45RA or CCR7, indicating an effector memory (EM, CCR7^–CD45RA^– phenotype) [21, 25]. The cytotoxic profile, represented by increased expression of granzyme B on CD8+ T cells compared to healthy donors, persisted after virus elimination. The content of proliferating T cells (carrying the Ki67 receptor) was low immediately after infection, then increased to 11–13 days and decreased immediately before virus elimination. Both patients showed a decrease in the content of B and NK cells [24]. Analysis of IFN-γ production after stimulation with phytohemagglutinin during the viremic phase of infection revealed weak responses, indicating functional T-cell anergy, which was transient. It is noteworthy that T-cell anergy was accompanied by reactivation of the infection caused by EBV. In both patients, during the viremia phase, an increase in AMBRA-1 (activating molecule in Beclin1-regulated autophagy)-positive cells [24] and an increase in CD95 expression on CD4⁺ and CD8⁺ T cells, followed by a decrease, as well as a short-term increase in PD-1 protein content on the surface of CD4⁺ T lymphocytes, which indicated the possible involvement of autophagic/apoptotic pathways in the massive loss of lymphocytes during EVD infection and depletion of T helper cells [21].

As in most reported cases of the disease outside Africa [26], the treatment received by patients was not uniform, nor were their physical conditions and infection conditions similar. Accordingly, the kinetics of immunological parameters varied greatly. In patient No. 2, the decrease in viral load was associated with low levels of proliferating T cells and a decrease in the level of granzyme B-expressing CD8⁺ T cells. However, the decrease in viral load did not prevent the loss of CD4⁺ T cells and an increase in the expression of activation markers. In patient No. 1, the rate of T-cell proliferation corresponded to the kinetics of viral load, while cytotoxic T-lymphocytes persisted as long as the virus was present. Thus, the data of C. Agrati et al. [21]. confirmed the combination of both models of immune response to EVD [18, 27], since strong immunosuppression (leading to EBV reactivation) was observed simultaneously with the activation of CD8⁺ T cells, which gradually removed EVD RNA from the blood.

In 2017, the results of a comprehensive study of a patient who survived Ebola infection, contracted in Sierra Leone and treated in Germany, who received only supportive therapy without experimental drugs, were published [28]. During the course of the disease, he had numerous complications [29]. The last positive results for the detection of viral RNA in plasma by reverse transcription polymerase chain reaction were obtained on the 17th day after the onset of the disease. Discharge was delayed due to the detection of viral RNA in urine on the 30th day and in sweat on the 40th day.

The proportion of effector CD4⁺ and CD8⁺ memory T cells (CCR7^–CD45RA^–) was increased on days 37 and 46 compared with control values, as was the proportion of CD8⁺ and CD8⁺EM cells co-expressing HLA-DR and CD38. An increase in activation cells was also observed among CD4⁺ T cells and their corresponding CD4⁺ EM subset, but to a lesser extent. CD8⁺ T cells showed a high proportion of cytotoxic granzyme B-positive cells. The population of regulatory T cells was quantitatively comparable to that of healthy donors. To determine the EVD-specific cellular response, the induction of four cytokines in response to GP-specific stimulation with overlapping peptides was investigated. The content of EVD-GP-specific T cells was insignificant; they were predominantly found in the CD8⁺ T cell population with the highest proportion of cells producing TNF-α, but not IFN-γ, IL-2, or macrophage inflammatory protein (MIP-1β), as well as among CD107a⁺ cells [28]. Subpopulations of B cells: regulatory (CD19⁺CD24^hiCD38^hi) and transient (CD24^hiCD27⁺) were lower than control values, indicating an autoimmune process in the body. An increase in plasmablasts (CD19⁺CD24^-CD38^hi) and the detection of EVD-specific neutralizing antibodies with reverse geometric mean titers of 91 and 76 on days 37 and 46, respectively, indicated activation of the humoral immune response. Thus, both FC and ELISPOT revealed EVD-specific antiviral T-cell cytotoxicity and activity, albeit at low levels, one month after the end of viremia and clinical recovery of the patient, as well as the formation of cellular and humoral immunity and immunological memory. Prolonged presence of the virus in the body can contribute to the formation of post-ebola syndrome [30], so the authors concluded that the results obtained are valuable for studying this syndrome, as well as for developing vaccines against EVD.

S.M. LaVergne et al. investigated EVD-specific T-cell responses and humoral immunity in 37 survivors of EVD disease in Sierra Leone [30]. To stimulate PBMC and study the specific response, they used a recombinant vesicular stomatitis virus encoding EVD proteins, strain Makona G3845 [31]. The isolated lymphocytes were placed in a plate in RPMI culture medium and infected with recombinant vesicular stomatitis virus at an infection multiplicity of 15 relative units of luminescence. For negative control, cells were infected with recombinant vesicular stomatitis virus encoding enhanced green fluorescent protein (EGFP). For positive control, cells were incubated with antibodies to CD3 and CD28. The cells were then stained with antibodies to CD3, CD4 and CD8, fixed, permeabilized, stained with antibodies to IFN-γ, TNF-α, and IL-2, and analyzed by flow cytometry.

In the group with post-ebola syndrome, 84% of participants had positive CD8⁺ T-cell responses to 7 EVD antigens, compared with 47% positive responses in the group without post-ebola syndrome. Similarly, the first group had a higher level of positive CD4⁺ T cells (63% vs. 27%). A significant correlation was found between the sum of specific CD8⁺ T-cell responses to any EVD antigen and IgG titers of specific antibodies to EVD among individuals with a T-cell response to CD8⁺ antigen.

Analyzing their results and data from other studies, the authors suggested that post-EVD syndrome is associated with higher levels of viremia during infection [32] and the presence of the virus in immune-privileged organs during recovery [33–35]. Cell damage during acute infection may be more significant in individuals with higher virus titers. The presence of the virus during recovery may cause cell damage either through the direct cytopathic action of the virus or as a result of the immune response to infected cells.

Continued close attention to EVD over the past decade, the development of modern technologies for storage, transportation, and sample preparation in virological studies, along with increased transport accessibility to endemic regions of Africa, have made it possible to obtain and summarize data on mass outbreaks of the disease.

In 2020, R. Thom et al. published data on the specific immune response to EVD in 206 people from two prefectures in Guinea (117 people survived EVD infection; 66 people had contact with infected individuals but did not become ill; 23 people were negative controls) [8]. Intracellular determination of IFN-γ, TNF, and IL-2 content using FC was performed after incubation of PBMC with antibodies to CD28, CD49d, and CD107a and stimulation with a pool of peptides specific to EVD GP. Antibodies to CD3, CD4, CD8, CD19, CD14, CCR7, CD95, and CD45RO were used for phenotyping. IFN-γ was determined using ELISpot.

The antibody titer among those who had recovered from the disease 3–14 months after infection was 10 times higher in 95% of cases than the titer formed 1 month after administration of a single dose of the EVD vaccine. The dominant phenotype of multifunctional CD8⁺ T cells measured among 53 individuals who survived EVD infection was IFN-γ⁺FNF⁺IL-2⁺, with an average response to EVD GP of 0.046% of the total number of CD8⁺ T cells. Furthermore, both neutralizing antibodies and T-cell responses were detected in 6 (9%) of 66 patients who had been in contact with infected individuals. It was noted that 4 (3%) of 117 people infected with EVD did not have circulating antibodies specific to the virus 3 months after infection.

When phenotyping IFN-γ⁺ T cells, most CD8⁺ lymphocytes showed intermediate CCR7 expression and low CD45RO expression, which corresponded to the phenotype of naive lymphocytes. In contrast, CD4⁺ T cells showed mostly positive CD45RO expression and low or intermediate CCR7 expression, corresponding to the phenotype of central memory or EM cells [25]. No significant correlation or difference was found between age, sex, or viral load and any of the measured immunological parameters. Phenotyping of CD8⁺ T cells revealed a possible subset of stem memory T cells, which are a self-renewing population of lymphocytes with a CD45RO^–CCR7⁺CD27⁺CD95⁺ phenotype [36]. The phenotype of IFN-γ-producing CD4⁺ T cells was CD45RO⁺, therefore, these cells were memory T cells rather than terminally differentiated effectors (TEMRA with the phenotype CD45RA⁺CD45R0^–CD62L^–CCR7^–CD27^–CD28^–) that had recently been exposed to antigen. The authors concluded that high titers of neutralizing antibodies and an enhanced T-cell response may indicate long-term immunity in survivors. The detection of antibodies and T-cell responses upon contact with individuals infected with EVD provides further evidence for the existence of subclinical EVD infection.

Over the past 20 years, no work on the creation or testing of vaccines against diseases caused by EVD has been done without the use of FC [7]. Unfortunately, immunization regimens aimed at creating humoral immunity are not always effective. In this regard, special attention is being paid to the development of vaccines that induce the formation of a T-cell response. In this regard, FC measurement of T-cell immunogenicity is of great importance for current research.

The results of cell response studies conducted by R. Thom et al. on individuals who had recovered from EVD infection several years earlier [8] confirmed the conclusion made by D.A. Stanley et al. in their study on testing vaccines against EVD infection in lower primates [7], namely that the most intense and long-lasting immunity against EVD is induced by vaccines that stimulate the production of multifunctional CD8⁺ T cells, i.e., those that simultaneously express the cytokines IFN-γ, TNF-α, and IL-2. The T-cell responses measured by R. Thom et al. [8] were comparable to the results observed 6 months after vaccination of volunteers with a single dose of a vaccine based on chimpanzee adenovirus type 3 (ChAd3), encoding the ZEBOV surface GP with the addition of a booster dose of a modified vaccine virus, Ankara strain, encoding the same GP [37, 38].

Intracellular cytokine staining and measurement by FC were used in Phase I and II clinical trials of the Zabdeno/Mvabea vaccine, approved by the World Health Organization in April 2021 for emergency use in adults and children aged 1 year and older [39], and the rAd5 vaccine. ZEBOV, developed in China and approved by local regulators [40], and a number of candidate vaccines [41–43].

V. Raabe et al. used FC to study intracellular cytokine secretion after immunization with a recombinant vesicular stomatitis virus-based Ebola vaccine (rVSVΔG ZEBOV-GP) in a prospective multicenter study [10]. CD4⁺ and CD8⁺ T-cell responses to Ebola GP were observed in 47% and 59% of participants at baseline and in 100% and 72% of participants 1 month after vaccination, respectively. The T-cell response was predominantly monocytic, with the exception of the CD8⁺ response to GP EVD in participants previously vaccinated with a heterologous vaccine against EVD.

FC has revolutionized the ability to measure T-cell proliferation, a functional feature important for T-cell control during vaccination [44]. Since FC can be used to assess not only the presence but also the intensity of fluorescence, staining cells with the fluorescent dye carboxyfluorescein succinimidyl ester (CFSE) allows tracking cell division in vivo and in vitro and can be easily combined with phenotypic and intracellular measurements. CFSE easily penetrates cell membranes; intracellularly, it is deacetylated and then covalently binds to cytoplasmic proteins, thus remaining inside the cell. As labeled cells divide, the dye is distributed among the daughter cells, marking subsequent generations with reduced fluorescence [44, 45]. In this way, up to 7 cell divisions can be detected, allowing cell populations with increased proliferation to be identified by the degree of dye dilution. Therefore, CFSE can be used as an indicator of antigen specificity to obtain information about the frequency of proliferating lymphocytes induced by vaccination in laboratory models and in PBMC isolated from the blood of volunteers.

Furthermore, lymphoproliferation with CFSE can be combined with other functional assays to obtain a more complete picture of cell reactivity. Thus, using lymphoproliferative analysis of CFSE-stained T cells and FC, the antigen-specific T-cell response to the combined vector vaccine against EVD, GamEvak-Combi, developed and registered in Russia, was evaluated. Proliferating CD4⁺ and CD8⁺ T lymphocytes were identified by forward and side light scattering based on CD3, CD4, CD8 expression and low CFSE fluorescence. Proliferation in unstimulated cells was subtracted from stimulated cell proliferation, and negative differences were assumed to be zero. The vaccine demonstrated a favorable safety profile and induced a robust antigen-specific humoral and cellular immune response in 100% of volunteers in Phase I–II clinical trials [46]. Thus, FC has significantly expanded our understanding of the mechanisms of immune responses that occur when various antigens are introduced in the creation of vaccines against EVD.

In the last decade, studies using FC have appeared that clarify the mechanism of EVD penetration and spread in the body and explain the disorders caused by EVD. Numerous receptors on the cell surface mediate the binding of filoviruses and their penetration into the endosomal compartment of cells, including phosphatidylserine receptors [47]. At the beginning of the century, reports appeared about the involvement of the costimulatory T-cell molecule Tim-1 (T-cell immunoglobulin mucin domain-1, mucin domain-1 of T-cell immunoglobulin) in the immune response to viral entry into the body, particularly in binding to viral envelopes [48], and in the immunomodulation of inflammatory reactions after signaling pathway activation [49, 50]. Tim-1 is a EVD attachment factor, a phosphatidylserine receptor that mediates the entry of filoviruses into cells through interaction with phosphatidylserine on virions [48]. Human and mouse macrophages express Tim-1 and phagocytose apoptotic bodies through interaction with phosphatidylserine [51]. The costimulatory function of Tim-1 occurs after interaction of the T-cell receptor with the antigen-presenting complex due to an increase in the level of the Tim-1 ligand [52].

P. Youkan et al. used FC to demonstrate the central role of Tim-1 in EVD entry into the body and its effect on CD4⁺ T cells in in vivo and ex vivo experiments [53]. Mice with a knockout Tim-1 gene (Tim-1^−/−) and wild-type control mice were infected with adapted EVD, strain Mayinga, by intraperitoneal administration of a dose of 1000 PFU, which in a previous study was found to be equal to 30 LD₅₀ [54]. Symptoms of the disease were observed in both groups, but Tim-1^−/− mice fully recovered by day 16 after infection, while all control animals died or were moribund by day 8. FC analysis showed that on day 6, Tim-1^−/− mice with activated CD4⁺ T cells had elevated levels of IL-2, IFN-γ, and TNF-α compared to wild-type mice infected with EVD.

P. Youkan et al. also conducted work with human PBMC [53]. CD4⁺ T lymphocytes obtained by negative magnetic separation, Vero-E6 cells, 293T and Jurkat cells were infected with recombinant EVD, Mayinga strain, expressing the GFP gene obtained as a result of transfection [55]. At a multiplicity of infection equal to 1 PFU per cell, the percentage of GFP-positive CD4⁺ T lymphocytes was 13.3%. A multiplicity of infection of 3 IU per cell led to an increase in the GP^hi cell population to 34%, which allowed the authors to hypothesize a dose-dependent effect of infection.

The relative degree of T-cell activation caused by EVD was determined by FC by comparing populations positive for activation and proliferation markers in PBMC devoid of virus target cells (dendritic cells and monocytes) and isolated CD4⁺ T cells. Incubation of these cells for 48 hours in the presence of EVD led to an increase in the proportion of CD25⁺, CD69⁺, and Ki-67⁺ populations compared to cells incubated without the virus. The addition of EVD to the “mixed” PBMC culture led to an increase in the proportion of activated CD4⁺ T cells by less than 2-fold. The addition of EVD to isolated CD4⁺ T cells led to the largest relative increase in populations positive for all three activation markers, 3.5–4.5-fold. These results indicated that CD4⁺ T cells are activated more efficiently in the absence of known target cells, demonstrating that EVD directly stimulates CD4⁺ T cells. Intracellular FC performed on primary CD4⁺ T cells after adding EVD with a multiplicity of infection of 0.3 showed a significant increase in the populations of cells producing IL-2, IFN-γ, and TNF-α on days 1 and 4.

Blocking Tim-1 with small interfering RNAs significantly reduced T cell activation based on CD25 and CD69 expression, indicating a direct role for Tim-1 in the development of cytokine storm during EVD infection [53].

To determine whether EVD binding to helper cells was random or targeted to specific subsets, CD4⁺ T lymphocytes were stained with antibodies to HLA-DR, CD38, CD45RO, and CCR7. The addition of EVD led to the formation of a separate population carrying both HLA-DR and CD38 antigens (HLA-DR⁺CD38⁺), representing an activated subset of CD4⁺ T cells. To determine whether EVD binds predominantly to naive cells or memory cells and activates them, lymphocytes were stained with CD45RO and HLA-DR. Most GP⁺-cells that reacted positively to HLA-DR also reacted to CD45RO, indicating that EVD binds to CD4⁺CD45RO⁺ memory T cells. It was further determined that EVD binds mainly to CD45RO⁺CCR7⁺ central memory cells.

Summarizing the study by P. Younan et al., it can be said that the binding of EVD to Tim-1 on the surface of CD4⁺ T lymphocytes leads to their direct activation, indicating the direct role of Tim-1 in the development of cytokine storm. Loss of Tim-1 expression leads to a decrease in viral load in the late stages of infection and significantly reduces EVD -induced mortality in mice. These results indicate that Tim-1 serves as an important receptor for EVD in vivo. Blocking the Tim-1 + EVD interaction could form the basis of an effective antiviral strategy.

R.I. Santos et al. actively used FC when studying the effect of Tim-1 on intercellular EVD transmission [56]. Three cell lines were used, two of which were derivatives of the Huh7.5 line with exons e2 and e3 disabled using CRISPR–Cas9 [57], and control cells transduced with a lentiviral vector expressing scrambled sgRNA. The cell lines were grouped together either as donors (except for Huh7.5ΔTim1-e3 with knockout exon 3) or as acceptors, resulting in 12 combinations. Donor cells were infected with recombinant EVD expressing GFP [55] or its derivative, in which GP EVD was replaced with an analogue from Ebola virus Bundibugyo, at a multiplicity of infection of 5 PFU per cell and incubated for 48 h. Donor cells were treated with mAb targeting the glycan cap, which can neutralize free viral particles [58]. After 1 hour of exposure, the cells were washed in phosphate-buffered saline and placed on top of acceptor cells pre-stained with CellTraceFarRed dye and incubated for another 48 hours. The efficiency of cell-to-cell virus transmission was assessed by FC analysis by counting double-positive CellTraceFarRed⁺eGFP⁺ cells. To ensure that the virus was transmitted exclusively through intercellular junctions, neutralizing mAbs targeting the glycan cap were added to some wells [58].

The use of cell lines with disabled exons as acceptors significantly reduced the number of double-positive cells compared to control cells with preserved Tim-1 expression, suggesting that Tim-1 is necessary for effective spread across intercellular junctions for both EVD expressing GFP protein, and for Bundibugoy fever virus. Viral diffusion ensures rapid transmission from host to host, but viral particles are vulnerable to host antibody responses. Cell-to-cell spread is a strategy widely used by various viruses, including HIV type 1, hepatitis C, African swine fever, and cowpox viruses [59–61]. For EVD, cell-to-cell spread was demonstrated by C. Miao et al. [62].

R.I. Santos et al. found that mAbs targeting GP MPER (membranous proximal external region) inhibit EVD spread through intercellular connections. The effectiveness of antibody blocking was enhanced by the host protein BST2 (also known as tetherin, CD317, or HM1.24) [63]. BST2 has been identified as an effective cellular factor that prevents the release of human immunodeficiency virus-1 in the absence of the viral accessory protein U [64]. BST2 inhibits the replication of a wide range of enveloped viruses [63]. On the other hand, some viruses have evolved virus-encoded antagonists to counteract the antiviral effect of BST2. EVD also appears to have evolved GP as an antagonist of BST2. However, it has been reported that high expression of BST2 suppresses ZEBOV replication even in the presence of GP [65].

To determine the surface expression of BST2 in various cell lines, cells were cultured in T75 flasks, trypsinized and resuspended in culture medium [56]. They were then placed in separate polystyrene tubes and incubated with mouse mAb specific for human BST2 conjugated to phycoerythrin or phycoerythrin-conjugated mouse isotype control. The cells were then analyzed by flow cytometry to determine the percentage of PE⁺ cells (expressing BST2) in the total cell population. The FC study was duplicated by microscopic analysis. It was established that MPER-specific mAbs completely prevented the virus from entering cells. Thus, the work of R.I. Santos et al. allowed us to conclude that EVD and Bundibugoy fever virus are capable of spreading to neighboring cells through intercellular connections in a process dependent on the Tim-1 protein. This type of Ebola virus transmission can be blocked by mAbs, especially those targeting GP MPER (BDBV223). BDBV223 mAb-mediated inhibition of intercellular transmission of Ebola viruses depends on the host protein BST2. It has been suggested that some protective mechanisms against EVD are not necessarily reflected in antibody activity as measured by plaque reduction assays [56].

Modern cytometers differ from their predecessors of 10–20 years ago in their increased analysis speed and greater number of lasers and fluorescent detectors, allowing dozens of parameters to be studied. The dynamics of cytometry development imply its integration with other methods.

Multiplex microsphere immunoassay (MMIA), based on a combination of FC and indirect enzyme-linked immunosorbent assay (ELISA), has become widely used and is a valuable tool for the quantitative determination of soluble molecules [66]. Among the existing multiplex platforms, Luminex xMAP (multi-analyte profiling) is the most popular—a technology based on the use of FC principles with fluorophore-labeled polystyrene microspheres [67, 68]. Polystyrene microspheres filled with fluorescent dye and immersed in a suspension have an activated surface that binds them to a specific antibody for capture. Antibodies for detection with a fluorescent reporter are added after the incubation and washing steps are complete. Each bead contains a “sandwich” consisting of the captured target analyte and a similar antibody conjugated to a reporter. The reporter constructs from the bead analyte are analyzed in a flow chamber, where lasers excite the reporters and the emitted light is collected by a series of detectors for quantitative analysis.

In 2010, data were published on the results of a study of a collection of 56 blood samples (42 deceased and 14 recovered) from patients of all documented outbreaks of the recombinant vesicular stomatitis virus (ZEBOV) that occurred in Gabon and the Republic of Congo between 1996 and 2005. [17]. Plasma samples were tested using Luminex technology to measure the levels of 50 cytokines, chemokines, and growth factors. It was determined that fatal outcomes were associated with hypersecretion of proinflammatory cytokines (IL-1b, -1RA, -6, -8, -15, and -16), chemokines, and growth factors (MIP-1a, MIP-1b, MCP-1, M-CSF, MIF, IP-10, GRO-a, and eotaxin). It should be noted that no increase in IFN-α2 was found in the plasma of patients. Furthermore, the indicators of the deceased were characterized by very low levels of circulating cytokines produced by T-lymphocytes (IL-2, -3, -4, -5, -9, -13). This largest human study at the time showed that fatal outcomes in EVD-induced disease are associated with an aberrant innate immune response and global suppression of adaptive immunity. Immune responses were characterized by a “cytokine storm” with hypersecretion of numerous pro-inflammatory cytokines, chemokines, growth factors, and the absence of antiviral IFN-α2. Immunosuppression manifested itself in low levels of circulating cytokines produced by T lymphocytes and massive loss of peripheral CD4⁺ and CD8⁺ lymphocytes.

Based on Luminex xMAP technology, an MMIA kit was developed for the simultaneous detection of antibodies to four types of EVD (EBOV, SUDV, BDBV, REST, Reston Ebola virus) in humans and animals [69]. Commercially available recombinant proteins derived from nucleoprotein (NP), VP40 matrix protein, and GP from different EVD strains were used as antigens. A total of 94 serum samples obtained from survivors of the 2014–2015 EVD outbreak in Guinea were tested [70, 71]. For the negative control, 108 plasma samples from the laboratory of the University Hospital of Montpellier, France, were used. The xMAP analysis results were compared with data obtained using commercially available ELISA kits to detect IgG antibodies to GP- and NP-EBOV and to GP-SUDV as separate tests on a single sample panel. The Luminex assay was found to be as sensitive but more specific, more accurate, and more cost-effective than ELISA for detecting EVD IgG in human plasma. The use of Luminex technology has enabled large-scale retrospective and prospective surveys of human and wildlife populations to identify unreported diseases in areas conducive to EVD circulation, as well as to determine the ecological niches occupied by EVD.

R. Surtees et al. published data on the development of an MMIA kit based on Luminex xMAP technology for the simultaneous detection of antibodies in human and animal blood serum samples produced in response to infection by members of the Filoviridae, Phenuiviridae, Nairoviridae and Hantaviridae families [72]. The nucleoproteins of the following viruses were selected as target antigens: Marburg virus, Congo-Crimean hemorrhagic fever virus, Dobrava-Belgrade virus, and Rift Valley fever virus were selected as target antigens, as it had previously been shown that NP are highly immunogenic and their sequences are conserved in different species of viruses of the same genus [73–76]. The kit was used to analyze 129 human serum samples collected in Guinea in 2011–2012. Furthermore, to expand the scope of application and ensure comprehensive serological surveillance, a system for detecting pan-species antibodies was created by replacing the secondary antibody with a mixture of biotinylated protein A and protein G. The kit's ability to detect antibodies from other animal species was confirmed using serum from insectivorous bats caught in Côte d'Ivoire and immunized with virus-like particles of types EVD and Marburg virus. To optimize MMIA and negative controls, 88 human serum samples from a German blood bank were used. Serum samples from lower primates infected with Rift Valley fever virus, Marburg virus, or EVD virus were used for validation (this serum can be used as a substitute for human serum in immunochemical assays [77]). The ability of MMIA to detect IgG antibodies to Dobrava-Belgrade virus NP and Congo-Crimean hemorrhagic fever virus NP in human blood serum was tested using blood serum samples from patients who were characterized for the presence of the pathogen by FC and ELISA. This FC-based assay can be used as a monitoring tool to detect infection with multiple highly pathogenic viruses in humans and wildlife, which is important for risk assessment and prevention of zoonotic disease outbreaks.

As a rule, the study of cellular immunity in patients with EVD infection focuses on a specific cell population. This is partly due to the limitations of flow cytometry depending on the instrument used. This can be overcome by mass cytometry, which combines flow cytometry with mass spectrometry, allowing up to 50 parameters to be determined simultaneously. Cells labeled with antibodies conjugated to lanthanide metals are detected by time-of-flight mass spectrometry, which is why the technology is often referred to as CyTOF (cytometry by time of flight) [78, 79]. A. McElroy et al. published a study using CyTOF to examine cryopreserved MSCs from four patients with EVD infection who were treated at Emory University Hospital in 2014 [18, 80, 81]. Notable results include the loss of non-classical monocytes (CD14^–CD16⁺) and myeloid dendritic cells (Lin^–CD14^–CD16^–HLA^-DR⁺CD1c⁺) during the acute viremic phase, followed by their recovery; preservation and activation (increased CD38 expression) of plasmacytoid dendritic cells (Lin^–CD14^–CD16^–HLA^-DR⁺CD123⁺); signs of emergency hematopoiesis (increased growth factors) and noticeable proliferative activity (increased Ki-67⁺ expression) in many immune cell populations [80].

In recent years, another method derived from flow cytometry has been gaining momentum: flow virometry. This multi-parameter, high-throughput, and sensitive method allows the detection, quantification, and characterization of viral particles based on the biophysical properties of the virus and the expression of proteins on its surface. In particular, by calibrating a flow cytometer using reference materials, it is possible to measure the concentration of intact viral particles in a sample, the amount of target antigen on the surface of the virus, and the relative diameter of the virus. Flow virometry can be used to directly measure submicron particles smaller than 300 nm, including extracellular vesicles, exosomes, and larger viruses. The principle of the method is as follows. Viruses or extracellular vesicles are captured by magnetic nanoparticles with a wavelength of 15 nm, bound to antibodies that recognize one of the surface antigens. The captured virions or vesicles are incubated with fluorescent antibodies against other surface antigens. The resulting complexes are separated from unbound antibodies on magnetic columns and analyzed using flow cytometers optimized for the detection of nanoscale objects. Flow virometry uses violet lasers because they have the shortest wavelength and, therefore, scatter more frequently with sample streams containing particles smaller than or equal to the wavelength. The greater the scattering during sample processing, the easier it is to distinguish it from random background noise and obtain a clear signal [82].

Flow virometry was used to characterize the live vaccine against ZEBOV — ERVEBO [83]. ERVEBO is a live vesicular stomatitis virus genetically modified to express GP ZEBOV to elicit a protective immune response in vaccinees. The paper presents a high-throughput test that makes it possible to directly determine damage during the production of vaccine viral particles with loss of infectivity and to identify the process parameters that contribute to this.

Working with material that contains or is suspected of containing infectious agents of biological hazard groups I or II requires ensuring personal and public biological safety and environmental protection². The lack of sufficient number of biosafety level 4 facilities and trained personnel, as well as the complexities associated with working in biosafety level 4 laboratories, seriously hamper fundamental research on EVD, as well as the development of vaccines and large-scale screening of effective antiviral compounds. Over the past two decades, laboratory diagnostics and biological research have undergone a technological revolution. However, conducting research using primary samples obtained from patients with EVD-induced disease remains a challenge.

The method by which researchers have circumvented the difficulties associated with studying samples containing or suspected of containing EVD using FC is the exposure of EVD to immune cells in vitro. That is, in compliance with biosafety rules, PBMC are isolated from blood, cryopreserved, and stored in liquid nitrogen, extracting them as needed [3, 4, 6, 8, 14, 15, 17, 18, 21, 30, 80].

When preparing samples for blood culture and other biological materials for the purpose of chemical inactivation of a particularly dangerous pathogen, which is EVD, paraformaldehyde [84–86] and formaldehyde [85] are used for sample preparation for blood FC and other biological materials for the purpose of chemical inactivation of particularly dangerous pathogens, such as EVD, since they do not alter the light-scattering and fluorescent properties of cells, and their virus-inactivating properties are widely known.

The work with EVD presented in the review was carried out in laboratories with a biosafety level (BSL) of 4 or a sanitary zone of 3, and sample preparation for FC in these conditions was performed in accordance with specific recommendations [87]. The WhBl-BSL-3/4 protocol [86] was validated to FC standards for whole blood and PBMC analysis. The procedures were divided into two protocols: a protocol with paraformaldehyde inactivation for less than 30 minutes (so-called short exposure) and a procedure with paraformaldehyde inactivation overnight (long exposure) [87].

In general, blood sample preparation for FC in BSL-4 conditions when performing phenotyping and intracellular staining of cytokines and other elements is performed as follows. Peripheral venous blood is collected in accordance with appropriate safety rules into Vacutainer tubes with heparin or ethylenediaminetetraacetic acid. If necessary, PBMC are isolated from blood using gradient centrifugation [88] and suspended in culture medium. Whole blood is diluted in half with culture medium. The cells are then placed in culture plates and incubated overnight in a 5% CO₂ atmosphere in the presence of stimulating antigen(s) (cell virus lysate, pool of control peptides) and costimulating antibodies. One hour after the start of cultivation, brefeldin A is added to block intracellular transport and ensure the accumulation of cytokines in the Golgi apparatus. Stimulated whole blood cultures are stained for surface markers, treated with a red blood cell lysing solution, washed in phosphate-buffered saline, inactivated with 4% paraformaldehyde (according to the short or long protocol), and transferred from BSL-4. When working with PBMC, the lysing solution is not used. Under BSL-2 conditions, fixed samples are incubated with mAbs for intracellular FC. The samples are then analyzed on a cytometer according to standard procedures.

If only the expression of surface markers needs to be assessed, the stimulation and brefeldin treatment stages are omitted, the cells are immediately stained with antibodies to surface markers, treated with a lysing solution under BSL-4 conditions, washed in phosphate-saline buffer, inactivated with 4% paraformaldehyde, and transferred from BSL-4. Under BSL-2 conditions, the cells are immediately analyzed on a cytometer or washed in buffer and resuspended in phosphate-saline buffer with paraformaldehyde or formaldehyde at a concentration of 0.1–1.0% for storage for several days at 4°C [86].

Summarizing the information presented in the review, it can be said that the use of both FC and methods based on FC (Luminex xMAP, CyTOF) has revealed numerous facts about the immunopathogenesis of the disease caused by EVD:

• Cytotoxic lymphocytes play a leading role in protecting against EVD;
• Dendritic cells are an early target of EVD;
• The elimination of NK cells in the early stages of the disease may be a critical reason for the host's inability to mount a sustained immune response;
• EVD evades the immune response.

Fatal outcomes in EVD infections are associated with an aberrant innate immune response and global suppression of adaptive immunity. Immune responses in EVD-induced disease are characterized by a “cytokine storm” with hypersecretion of proinflammatory cytokines, chemokines, growth factors, and a lack of antiviral IFN-α2. Immunosuppression in EVD-induced disease is manifested by low levels of circulating cytokines and massive loss of peripheral CD4⁺ and CD8⁺ lymphocytes through Fas/FasL-mediated apoptosis. The key factors for the outcome of the infection are the timing and kinetics of viral replication and the immune response, since a late or incomplete immune response is unable to suppress viral replication, leading to death. Conversely, an early immune response reduces viral replication and ensures recovery. In recovered patients, T cells are activated, proliferate, and form specific antiviral cytotoxicity, cellular and humoral immunity, and immunological memory. FC also helped establish that an effective criterion for evaluating the antigen-specific T-cell immune response formed upon administration of vaccines against EVD is the proportion of polyfunctional T-lymphocytes.

Using FC, the mechanism of EVD penetration and spread in the body was clarified and the disorders caused by the virus were explained. It was found that the phosphatidylserine receptor Tim-1 plays a central role in EVD penetration into the body, the spread of the virus through intercellular connections, and the development of a “cytokine storm.” This type of EVD transmission can be blocked by mAbs targeting GP MPER. Inhibition of intercellular EVD transmission depends on the host protein BST2/teeterin/CD317.

FC has pioneered a new high-throughput and sensitive method that allows the detection, quantification, and characterization of viral particles based on the biophysical properties of the virus and the expression of proteins on its surface. In particular, flow virometry allows rapid detection of viral particles damaged during vaccine production and identification of process parameters that contribute to damage.

Immunological systemic innate and adaptive responses to EVD introduction remain poorly understood in many areas, and antigen presentation together with T-cell responses remains crucial for elucidating the future dynamics of host immune responses to EVD. Using FC in conjunction with virological, immunological, and molecular biological methods, including transcriptomic technologies and bioimaging, most stages of the pathogenesis of EVD-induced disease have been characterized, and the boundaries of understanding of complex immune responses induced by the pathogen itself and by antiviral drugs, polyclonal/monoclonal antibodies, and small interfering RNA molecules designed to prevent or cure the disease. Currently, 371 differentiation cluster molecules that can be expressed by immune cells have been classified. There are 10 Toll-like receptors in the differentiation cluster nomenclature, whose functions remain unclear in terms of how or why EVD causes such diverse immunological changes in affected organisms. Using the capabilities of FC, it remains to be studied the innate and adaptive responses of the immune system relevant to the pathogenesis of EVD-induced disease at the level of the whole organism, cells, and molecules.

 

¹ WHO. Ebola disease. URL: https://www.who.int/news-room/fact-sheets/detail/ebola-disease

² Centers for Disease Control and Prevention (CDC). (2006d).

URL: www.bt.cdc.gov/documents/PPTResponse/table3bbiosafety.pdf

About the authors

Galina V. Borisevich

48 Central Scientific Research Institute

Author for correspondence.
Email: 48cnii@mil.ru
ORCID iD: 0000-0002-0843-9427

Сand. Sci. (Biol.), senior researcher

Russian Federation, Sergiev Posad

Aleksander V. Ovchinnikov

48 Central Scientific Research Institute

Email: 48cnii@mil.ru
ORCID iD: 0000-0003-2309-3572

Cand. Sci. (Tech.), senior researcher

Russian Federation, Sergiev Posad

Svetlana L. Kirillova

48 Central Scientific Research Institute

Email: 48cnii@mil.ru
ORCID iD: 0000-0003-1245-9225

Dr. Sci. (Biol.), leading researcher

Russian Federation, Sergiev Posad

Anna S. Turavinina

48 Central Scientific Research Institute

Email: 48cnii@mil.ru
ORCID iD: 0009-0000-5662-6799

laboratory researche

Russian Federation, Sergiev Posad

References

Supplementary files