Viral hepatitis is a significant global health issue, impacting hundreds of millions of individuals worldwide. Chronic hepatitis B, C, and D infections are strongly associated with liver cancer (77–79), as all three viruses infect hepatocytes. They are the reason that hepatocellular carcinoma (HCC) is the most frequently diagnosed malignancy in many regions worldwide (80). Despite notable advancements in treatment options against HBV (81), HCV (82), and HDV (83) in recent years, chronic viral hepatitis remain a wide-spread medical issue. Viral hepatitis, characterized by the persistence of the virus in the liver, is considered to be an immune-mediated disease, implying that the immune system plays a crucial role in the development and progression of chronic viral hepatitis (84), but also in virus control and resolvement of infection (85). However, our understanding of the mechanisms that regulate antiviral immunity during the chronic stage of hepatitis viruses remains insufficient.

It is widely accepted that CD4⁺ Th cells are protective during HBV and HCV infections (86–89). At the same time the role of CD4⁺ CTL during hepatitis infection continues to be a topic of investigation. It has been shown that hepatocytes infected with hepatitis viruses express MHC Class II molecules on their surface (90) and acquire antigen presenting cell function (91). Thus, they could potentially be targeted by CD4⁺ CTL for killing. Indeed, CD4⁺ CTL, defined as Perforin-expressing CD4⁺ T cells, were detected in chronic viral hepatitis, especially in HDV infection (72) ( Figure 1 ). Phenotypically, such CD4⁺ CTL exhibited a terminally differentiated effector phenotype (CD28⁻, CD27⁻) similar to that described for CD4⁺ CTL in other chronic viral infections (37). Even though direct cytotoxic killing of hepatocytes was not investigated ex vivo, authors provide indirect evidence that Perforin-expressing CD4⁺ CTL do kill infected cells and may accelerate fibrogenesis and hepatitis (92). Therefore, despite the fact that further studies are required to more precisely define the role of CD4⁺ CTL in viral hepatitis, this subset is very likely involved in immune-mediated pathology. For example, CD4⁺ CTL are known to mediated liver disease upon secondary infections with dengue virus (93). The authors of this study showed that dengue virus capsid-specific CD4⁺ CTL were responsible for liver cell killing through Fas/FasL interaction and also killed APC through Perforin expression (93). Thus, it is not surprising that MHC II-expressing hepatocytes infected with viruses may become targets of CD4⁺ CTL killing. Indeed, in the cohort of viral hepatitis patients there was a striking correlation between CD4 Perforin expression and aspartate aminotransferase levels that serves as a marker of hepatocyte damage (72).

On the other hand, the loss of CD4⁺ CTL in patients with HBV-mediated HCC was correlated with a higher mortality rate and a reduced survival time (94). Moreover, in tumor biopsies from the liver, CD107a degranulation marker (a surrogate marker for exocytosis in cytotoxic cells) on CD4⁺ CTL was significantly reduced in patients with an advanced cancer stage. This cytotoxic activity of CD4⁺ T cells was shown to be controlled by local Tregs (94). In a mouse model the beneficial therapeutic activity of CD4⁺ CTL was reported in a vaccination approach against HCC (95). In this study, tumor formation was controlled by vaccine-induced CD4⁺ T cells and this control was abrogated by anti-CD4 antibody administration. This preclinical model may suggest that CD4⁺ CTL inducing therapy in humans should be considered for further investigation.

First evidence for the existence of CD4⁺ T cells with direct anti-viral effector functions came from studies of chronic retroviral infections in monkeys in the ‘90s (96). Since then, CD4⁺ CTL have been described in simian immunodeficiency virus (SIV) infection of rhesus macaques. Here, viremia increase, caused by CD8⁺ T cell depletion in SIV-infected macaques, was efficiently controlled due to a combination of antibody responses and expansion of circulating CD45RA⁻ CD28⁺ CD95⁺ CCR7⁻ Granzyme B (GzmB)⁺ SIV-specific CD4⁺ CTL (76). Similarly, Gag- and Nef-specific CD4⁺ T cell responses were found in CD8-depleted macaques during virus control (97). Another group demonstrated cytotoxicity of an SIV Gag-specific CD4⁺ T cell clone that had the capability to control viral replication (98). The main surface marker to characterize such CD4⁺ CTL in SIV-infected rhesus macaques was found to be CD29 (99, 100). CD29^hi GzmB^hi T-Bet⁺ gag-specific CD4⁺ T cells were also capable of shrinking the SIV reservoir during ART (99). SIV as well as HIV infects macrophages as well as CD4⁺ T cells and those constitutively express MHC II as APCs (101). In addition, activated CD4⁺ T cells start to express MHC II, and HLA-DR is even used as common activation marker for human CD4⁺ T cells (102). Thus, both cell types that propagate viral infection are potential targets for CD4⁺ CTL.

Similar to SIV, CD4⁺ CTL were found to be beneficial for people living with HIV. Despite the fact that CD4⁺ T cells are the main targets of HIV, including the crucial role of CD4⁺ T follicular helper cells in reservoir formation and maintenance (103, 104), studies support the vital role of CD4⁺ CTL in HIV control (70, 105). CD4⁺ CTL limit HIV pathogenesis in elite controllers (106). In acute HIV infection CD4⁺ CTL were characterized as GzmA⁺, IFN γ⁺, and CD40L⁺ and were associated with the reduction in viral loads (71). In another study similar observations were made and the phenotype of CD4⁺ CTL during acute HIV infection were described as Perforin⁺, GzmB⁺, and Eomes⁺ (70). Additionally, ex vivo studies revealed that virus-specific CD4⁺ CTL can kill HIV-infected macrophages and T cells (73). Moreover, an HIV vaccine candidate induced CD4⁺ CTL with lytic functions (107, 108). These lines of evidence suggest that CD4⁺ CTL play an important role in HIV control and therefore could be targeted as effectors in vaccine development and treatment interventions.

At the same time, inducing strong CD4⁺ T cell activation is a debatable issue in the field of HIV vaccination. It can be a double-edged sword since these cells may be favorable virus targets. In fact, it has been reported that GzmB⁺ cells harbored more HIV than GzmB^- cells in gut CD4⁺ T cells stimulated with enteric bacteria in the lamina propria aggregate culture model (109, 110). Nevertheless, the Thai HIV phase III prime-boost vaccination trial with ALVAC and AIDSVAX reported successful induction of HIV-specific CD4⁺ CTL (111). This vaccination regimen, combined a recombinant canarypox vector vaccine (ALVAC) and a recombinant glycoprotein 120 subunit vaccine (AIDSVAX), exhibited a moderate level of protection against HIV-1 infection partly correlating with the induction of polyfunctional effector CD4⁺ T cell responses (112, 113). Although stimulation of CD4⁺ CTL responses in HIV infection remains controversial, recent studies suggest that CD4⁺ CTL can compensate for reduced CD8⁺ T cell cytolytic activity against HIV in the setting of CD8⁺ T cell exhaustion (114), HIV-mediated downregulation of HLA I molecules (115), or CD8-associated HIV mutational escape (116). Therefore, while HIV-specific CD4⁺ CTL may be targeted by the virus and experience depletion during the early stages of infection, the remaining cells might play an important role in controlling viral loads (105).

Another well-known murine model of persistent viral infection is the lymphocytic choriomeningitis virus (LCMV) infection. LCMV is a member of the Arenaviridae family and is commonly found in rodents, particularly mice. LCMV has been extensively studied to understand the mechanisms of persistent viral infections and immune responses. The virus infects a broad spectrum of cellular targets, including dendritic cells (131), macrophages (132), endothelial cells (133), epithelial cells (134), fibroblasts (135), and neurons (136). As professional APC, dendritic cells, and macrophages constitutively express MHS class II molecules on their surface and therefore can serve as CD4⁺ CTL targets. CD4⁺-mediated killing of target cells in LCMV-infected mice was reported in β2-microglobulin-deficient (β2m⁻) mice (137). Mice lacking β2m do not effectively express MHC class I complexes, resulting in a deficiency of CD8⁺ T cell responses (138). In this setting, LCMV-infected animals do not develop CD8⁺ cytotoxicity, but instead generate MHC class II-mediated CD4⁺ CTL (139, 140). These LCMV-specific CD4⁺ CTL express FasL and utilized the Fas-dependent killing pathway (141). However, other group showed that in LCMV infection both the FasL- and a Perforin-dependent pathway can contribute to CD4⁺ CTL killing (75). That again demonstrates that impaired CD8⁺ CTL responses are compensated in vivo by cytotoxic CD4⁺ T cells.

Other authors compared the in vivo killing mediated by CD4⁺ versus CD8⁺ T cells utilizing an in vivo CTL assay during LCMV infection (142). They detected substantial CD4⁺ CTL-mediated killing of target cells loaded with the immunodominant peptide LCMV-GP_64–80 (143) in mice infected with LCMV that was measured at 16 hours after target cell infusion. This killing appeared less efficient as compared to the remarkably fast CD8-mediated in vivo killing of target cells, however researchers concluded that CD4⁺ and CD8⁺ CTL responses were similar in magnitude and were only slower due to the FasL-dependent pathway of CD4⁺ T cell killing (144). Additionally, a careful transcriptional investigation helped to discover a new marker for CD4⁺ CTL in LCMV infection (145). These cytotoxic cells express Eomes and GzmK together with uniquely high expression of the signaling lymphocytic activation molecule family member 7 (SLAMF7), a surface protein that was already described to characterize CD4⁺ T cells with cytotoxic potential in tumor (146) and autoimmune diseases (147).

Interestingly, experimental induction of CD4⁺ T cells during chronic LCMV caused lethal immunopathology in mice (148). Administration of vaccines to selectively induce CD4⁺ T cell responses resulted in severe generalized inflammation, a cytokine storm, and mortality. Furthermore, adoptive transfer of LCMV-specific CD4⁺ T cells following acute infection induced lethal inflammation (149). These results demonstrate the fine balance between anti-viral immunity and immunopathology for CD4⁺ CTL that has to be taken into account when designing immunotherapies or vaccines to induce such cells.

Viruses of the Herpesviridae family affect the majority of the human population. They establish lifelong infections, however are largely asymptomatic in healthy individuals, while causing severe disease in the immunocompromised hosts (150). CD4⁺ CTL were described in mouse and human herpesvirus infections. CD4⁺ T cells with cytotoxic effector functions were found in murine cytomegalovirus (MCMV) (151) and chronic infection of mice with γ-herpesvirus 68 (152, 153). Accordingly, CD4⁺ CTL were isolated from peripheral blood mononuclear cells of humans infected with human cytomegalovirus (HCMV) (36, 154). Other authors demonstrated HCMV-specific effector CD4⁺ T cells that expressed GzmA, GzmB, and Perforin with antiviral activity (67). Moreover, HCMV-pp65-specific CD4⁺ CTL were described in the cohort of older adults (155). CD4⁺ CTL have been observed in individuals infected with human herpesvirus-6B, suggesting their role in the long-term control of the disease (156).

The presence of CD4⁺ T cells exhibiting cytotoxic potential has also been identified in patients and mice infected with Epstein-Barr virus (EBV) (157). EBV is highly immunogenic virus which can be associated with the emergence of various types of cancer affecting B cells and epithelial cells (158). In individuals with infectious mononucleosis, blood samples reveal the presence of CD4⁺ T cells that express GzmB with potential anti-viral activity (159). CD4⁺ T cells isolated from tonsils, the hotspot of EBV infection, demonstrated cytotoxic potential in vitro (160). Cytotoxicity of EBV-specific CD4⁺ T cells was shown indirectly through the expansion of lytic cells from peripheral blood mononuclear cells, acquired from EBV-seropositive donors. The lytic activity was demonstrated to be facilitated through different pathways: the secretion of the cytotoxic molecules Perforin and Granzyme (161, 162) or via the Fas/FasL pathway (163). Moreover, lytic activity of EBNA1-specific CD4⁺ T cells against virus-transformed tumor cells was observed in all EBV-mediated malignancies, including Burkitt lymphoma (BL). BL cell lines serve as targets for the cytotoxic activity of CD4⁺ T cells specific against EBNA1 (164). In a mouse model, BL were eliminated in the absence of any CD8⁺ T lymphocytes, however no direct lytic CD4⁺ CTL activity could be detected in that model (165).

EBV is not only linked to the malignancies, but is associated with the development of autoimmune diseases, such as systemic lupus erythematosus (166), myasthenia gravis (167), multiple sclerosis (168), rheumatoid arthritis (169), celiac disease (170), and Sjögren’s syndrome (SS) (171). The development of SS has been linked to EBV infection, as salivary gland biopsies taken from SS patients demonstrate elevated levels of EBV DNA compared to healthy salivary glands. This suggests viral reactivation and an impaired immune system’s ability to control EBV latency (172). Interestingly, EBV-specific T cells show cross-reactivity to endogenous peptides from tears and saliva (173). While EBV is commonly found in salivary gland epithelial cells of healthy individuals, SS patients exhibit increased levels of HLA-DR expression in their salivary gland epithelial cells (172). This allows them to present EBV antigens to T cells and become targets of CD4⁺ CTL-mediated killing, which contributes to tissue damage. Indeed, a study investigating the cytotoxic immune response in ectopic lymphoid structures with persistent EBV infection in SS salivary glands revealed an increase in CD4⁺ GzmB⁺ CTL that was a risk factor for organ lesions (174). In contrast, CD8⁺ GzmB⁺ lymphocytes were impaired and did not correlate with the damage of the salivary glands. Moreover, a positive correlation has been observed between elevated levels of CD4⁺ CTL in the peripheral blood and their increased infiltration into the salivary glands, which is associated with disease progression and severity (175). Thus, CD4⁺ CTL might play an important role in the pathogenesis of EBV-induced disease.

Taken together, anti-viral CD4⁺ CTL are found in almost every chronic viral infection. They seem to compensate for CD8⁺ CTL responses when they are poorly induced or become exhausted. Their induction and differentiation are counter-regulated by Tregs, most likely because they can be dangerous inducers of immunopathology. CD4⁺ CTL often kill via the Fas/FasL pathway, but they can also facilitate the exocytosis pathway of killing. They are able to keep persistent virus in check, but especially their FasL pathway of killing is not sufficient to eliminate chronic viruses. CD4⁺ CTL are interesting tools for cure approaches in chronic viral infections, but their potential to induce immunopathology has to be carefully taken into account.

Due to the fact that CD4⁺ CTL attracted attention in solid cancers only recently, there is still limited but growing data about their cytotoxic activity against tumor cells. So far, CD4⁺ T cell-mediated killing of either MHC Class II-matched or autologous tumor cells has been demonstrated in melanoma, bladder cancer and glioblastoma (12, 146, 191, 195, 196). Kitano et al. were the first to described cytolysis of human melanoma cells by NY-ESO-specific CD4⁺ T cells, that they found induced or enhanced in peripheral blood of patients treated with the anti-CTLA-4 blocking antibody ipilimumab. Those T cells expressed Perforin and GzmB and efficiently killed autologous melanoma cells in an MHC Class II-dependent manner (12). To achieve MHC Class II antigen presentation tumor cells were either transduced with CIITA the transcriptional activator of genes encoding the MHC Class II antigen presentation machinery (146, 197), or pretreated with IFNγ, known as potent inducer of MHC Class II expression (12).

Important to note, a sizable fraction of melanomas shows IFNγ-independent constitutive MHC class II surface expression (194, 195, 198–200), which in general is considered a specific feature of professional APC as dendritic cells, macrophages or B cells. So far, the mechanisms driving constitutive MHC class II expression in melanoma are poorly understood. Recently, we demonstrated that JAK1/2 signaling is involved in both IFNγ-induced and IFNγ-independent constitutive MHC Class II expression (195). The pathways triggering aberrant JAK1/2 activation in the absence of interferons remain to be determined. In line with this regulation, patient-derived JAK1/2-deficient melanoma cells displayed a stable MHC Class II-negative phenotype, resistant to CD4⁺ CTL (195). Interestingly, prior studies showed that ERK signaling negatively regulates constitutive but also IFNγ-induced CIITA expression (201, 202), suggesting that oncogenic Ras-RAF-MEK-ERK pathway activation in tumor cells counteracts MHC class II antigen presentation. So far, constitutive MHC class II expression has mainly been studied in melanoma, but seems to be present also in other cancers like glioma and lung cancer (195, 199, 203, 204), indicating a broader relevance of the MHC class II-positive tumor cell immunophenotype.

According to its role in CD4⁺ CTL activation, melanoma cell-intrinsic expression of MHC class II molecules has been associated with improved patient prognosis and response to immunotherapy with immune checkpoint blocking antibodies (200, 205). Similar data has been obtained for lung adenocarcinoma (206), but is lacking for most other tumors. This should encourage research to deepen our understanding on MHC class II regulation in different cancers as a basis for its therapeutic manipulation and killing of tumor cells by cytotoxic CD4⁺ CTL. As MHC class II-positive tumor cells can stimulate also tumor antigen-specific Treg (194), it is might be necessary to combine therapeutic MHC class II upregulation on cancer cells with Treg depleting strategies.

As CD4⁺ CTL have been intensively studied in the context of viral infections, this led to the idea of exploiting those cells also in therapy on solid cancers. The concept is based on the observation that virus-specific T cells have been detected among the infiltrates of different cancers. For instance, TILs isolated from both lymph node and subcutaneous tumors of melanoma patients contained CD8⁺ T cells with specificity for viral antigen epitopes originating from CMV, EBV or influenza A (207). CD8⁺ T cells specific for epitopes from those viruses were present also among TILs from glioblastoma, colorectal and lung cancer (208, 209). In line with the clinical observations, a preclinical study in B16 melanoma demonstrated that virus-specific CD8⁺ T cells infiltrated cutaneous tumors not only upon acute infection with CMV or poxvirus, but were resident in lesions after poxvirus elimination and during the chronic state of CMV infection as well (210). The therapeutic potential of tumor-resident virus-specific memory T cells has already been demonstrated in different murine tumor transplant models. Activation of virus-induced T cells by intralesional injection of viral antigen peptides delayed tumor growth (209, 211) ( Figure 2A ). Alternatively, immunocojugates have been proposed for delivery of viral epitopes into tumors ( Figure 2A ). In this case, viral peptides were coupled to antibodies targeting a cell surface protein expressed on tumor cells. Upon immunoconjugate binding the surface complex was internalized and viral peptides were shuttled to the ER for loading onto MHC molecules. In a xenograft tumor model, systemic application of the immunoconjugates mediated recruitment of adoptively transferred virus-specific T cells into the tumor and combined administration of immunoconjugates with immune checkpoint blocking antibody suppressed tumor growth (212). Although the aforementioned studies focused on CD8⁺ T cells, it is expected that tumor infiltrates contain also virus-specific CD4⁺ CTL that could be exploited for therapy of solid cancers by applying similar therapeutic strategies.

Notably, activation of tumor-resident virus-specific T cells might even be applicable to cancer patients who not yet encountered an infection with the specific pathogen. Analyses of the CD4⁺ T cell repertoire from adults detected HIV-1-, HSV- and CMV-specific CD4⁺ T cells in blood from unexposed individuals (213). Surprisingly, these CD4⁺ T cells showed features of memory T cells even without direct antigen contact, namely expression of memory-associated genes, clonal expansion and rapid cytokine production. Further analyses on HIV-1-specific CD4⁺ T cells from unexposed individuals revealed TCR cross-reactivity towards similar environmental microbial peptides (213). Although this study did not specifically focus on cytotoxic CD4⁺ T cells, it provided important insights into cross-reactivity of virus-specific CD4⁺ T cells towards similar MHC class II-presented peptides.

Currently, T cell cross-reactivity towards viral and tumor antigens is under intense investigation, as referenced in (214) ( Figure 2B ). A recent study demonstrated very broad specificity of a MHC Class II-restricted CD4 TCR isolated form TILs of a glioblastoma patient. Those T cells recognized different peptides derived from pathogenic bacteria, commensal gut microbiota and also glioblastoma-associated tumor antigens (215).

Overall, these finding suggest that redirecting pathogen-specific CD4⁺ CTL towards tumor cells could be a promising mean to enhance the efficacy of immunotherapies. However, further investigations are needed to define the antigen cross-reactivity of CD4⁺ CTL, to develop approaches that recruit pathogen-specific, cytotoxic CD4⁺ T cells into tumors with low immunogenicity and to unleash strong cytotoxic T cell responses against tumor cells in vivo without causing severe immune-related adverse events.