Research Article: Sepsis induces long-lasting impairments in CD4+ T-cell responses despite rapid numerical recovery of T-lymphocyte populations

Date Published: February 7, 2019

Publisher: Public Library of Science

Author(s): Christoph Ammer-Herrmenau, Upasana Kulkarni, Nico Andreas, Martin Ungelenk, Sarina Ravens, Christian Hübner, Angela Kather, Ingo Kurth, Michael Bauer, Thomas Kamradt, Francesco Ria.


Massive apoptosis of lymphocytes is a hallmark of sepsis. The resulting immunosuppression is associated with secondary infections, which are often lethal. Moreover, sepsis-survivors are burdened with increased morbidity and mortality for several years after the sepsis episode. The duration and clinical consequences of sepsis induced-immunosuppression are currently unknown. We have used the mouse model of peritoneal contamination and infection (PCI) to investigate the quantitative and qualitative recovery of T lymphocytes for 3.5 months after sepsis with or without IL-7 treatment. Thymic output and the numbers of naive and effector/memory CD4+ and CD8+ lymphocytes quickly recovered after sepsis. IL-7 treatment resulted in an accelerated recovery of CD8+ lymphocytes. Next generation sequencing revealed no significant narrowing of the T cell receptor repertoire 3.5 months after sepsis. In contrast, detailed functional analyses of T helper (Th)-cell responses towards a fungal antigen revealed a significant loss of Th cells. Whereas cytokine production was not impaired at the single cell level, the absolute number of Th cells specific for the fungal antigen was reduced. Our data indicate a clinically relevant loss of pathogen-specific T cell clones after sepsis. Given the small number of naive T lymphocytes specific for a given antigen, this decrement of T cell clones remains undetected even by sensitive methods such as deep sequencing. Taken together, our data are compatible with long lasting impairments in CD4+ T-cell responses after sepsis despite rapid recovery of T lymphocyte populations.

Partial Text

Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection [1]. Epidemiological studies suggest that more than 30 million cases of sepsis occur annually world-wide [2]. In the U.S., sepsis is the most expensive disease treated in hospitals with estimated hospital costs of > 20 billion US $ annually [3]. Mortality rates have been declining in high-income countries due to improved treatment and range from 20–50% depending on disease severity and other factors [4,5]. Immunologically, sepsis is characterised by concurrent proinflammatory and immunosuppressive alterations [6–10]. A prominent feature contributing to immunosuppression in sepsis is an early massive loss of lymphocytes due to apoptosis [8–12], which is recapitulated in mouse models of sepsis [8–10,13]. Profound or persistent lymphopenia in sepsis patients is associated with increased mortality [14,15]. Another important mechanism of sepsis-induced immune-suppression is the expansion of immunosuppressive cell populations including regulatory T lymphocytes, IL-10-producing B lymphocytes and myeloid-derived suppressor cells (MDSC) [8–10,16]. We have recently shown that numbers of IL-10 producing B lymphocytes and MDSC remain increased for months after sepsis [16]. Sepsis-induced immune-suppression renders patients susceptible to secondary opportunistic infections [17,18] and reactivation of latent viral infections [19], both of which contribute to late sepsis mortality [4,5,20]. It is currently unclear how long the sepsis-induced immunosuppression lasts and if an immunological restitutio ad integrum is reached in sepsis survivors. Most clinical and experimental studies to date have focussed on the immunopathology of acute sepsis. Clinical and epidemiological data indicate a massively increased morbidity and mortality of sepsis survivors for years after discharge from the hospital [4,21–23] and it is currently unknown how much persistent immunological alterations contribute to this disease burden. Boosting the immune response in sepsis patients is a promising approach to improve survival [4,9,10]. One candidate approach is the cytokine Interleukin (IL)-7, which is important for T-cell survival [24]. Early IL-7 treatment has been shown to improve survival in murine sepsis models [25,26] and to restore normal lymphocyte counts and functions in septic patients [27,28]. On the other hand, late IL-7 treatment prolongs the sepsis induced expansion of immunosuppressive IL-10 producing B-lymphocytes and MDSC after sepsis [16].

Massive apoptosis of lymphocytes is a hallmark of sepsis. We performed a long-term examination of the quantitative and qualitative T-lymphocyte recovery after sepsis. In accordance with earlier studies [4,9,10] we found rapid, complete and stable recovery of T cell numbers after sepsis. Similar to what has been observed in recent clinical trials [28] administration of recombinant human IL-7 accelerated and enhanced the recovery of T lymphocytes. Although CD4+ and CD8+ T cells express the IL-7R at similar levels [24], we found that the beneficial effect of IL-7 treatment on T cell recovery was more pronounced for CD8+ T cells than for CD4+ T cells, which confirms and extends earlier reports on other sepsis models [25,26,44]. Recovery of T cell numbers could occur via two different, not mutually exclusive, mechanisms. The apoptotic loss of lymphocytes could be overcome by homeostatic proliferation of the surviving clones, which depends largely on IL-7 and IL-15 together with low-affinity interactions of the TCR with p/MHC complexes [39,45]. Homeostatic proliferation is thought to be the main mechanism for T cell replenishment in adult humans [39,40]. Homeostatic proliferation of surviving clones would fail to generate new TCR specificities and result in a narrowed TCR repertoire [39,45]. Moreover, data from mouse models suggest that homeostatic proliferation favours T cells with TCRs displaying higher affinity for self-peptide/MHC complexes [46]. Therefore, over the course of many years, homeostatic proliferation could result in an increased risk for autoimmune responses. Alternatively, thymic output could be increased to replenish the peripheral T cell pool with newly generated naive T cells. To date, very little is known about the relative contributions of homeostatic proliferation and increased thymic output to T cell recovery after sepsis. To estimate thymic output after sepsis we used RAG2p-GFP transgenic mice [30] in which the RAG2 promoter drives GFP expression and the GFP signal remains detectable in recent thymic emigrants for approximately three weeks after the RAG gene is no longer expressed [41]. Compatible with earlier reports on sepsis-induced thymic atrophy [47,48] we found a decreased number of GFP+ recent thymic emigrants (RTEs) up to 8 days after sepsis. Thereafter, the frequency of RTEs was similar in control mice, septic mice and septic mice treated with IL-7, compatible with earlier reports that had indicated that thymic output is not influenced by the number of T cells in the periphery [41]. When comparing thymectomized and euthymic mice in the CLP model, Cabrera-Perez et al. also found no influence of the thymus on quantitative T-cell recovery after sepsis [49]. IL-7 treatment did not influence the frequency of CD4+ or CD8+ of GFP+ RTEs, which is consistent with earlier reports on altered IL-7 signaling and responsiveness in RTEs as compared with naive T cells undergoing homeostatic proliferation [24,41,50]. In the absence of increased thymic output the recovery of T cells after sepsis-induced lymphopenia seems to occur largely via homeostatic proliferation, which might result in a diminished diversity of the TCR repertoire [39,45]. Indeed, a decreased TCR repertoire diversity was reported in sepsis patients [51]. We used next generation sequencing to determine TCR diversity in post-septic mice. Despite the fact that there was no increased thymic output these analyses did not reveal a narrowed TCR repertoire. Sepsis-survivors are burdened with an increased risk of secondary infections [4,9,10,17–19]. Similarly, post-septic mice are more susceptible to secondary infections [26,44,49,52–57]. This could be explained by the loss of particular T-cell clones after sepsis, which might be undetectable even by the analysis of large numbers of TCR sequences [49]. Alternatively, sepsis could result in functional T-cell defects [58] [12] [56]. Thirdly the sepsis-induced expansion of immunosuppressive cell populations such as regulatory T cells, IL-10 producing B-cells and myeloid derived suppressor cells (MDSC) [16] could inhibit T cell effector functions. Finally, severe infections can result in a loss or functional impairment of antigen-presenting cells, resulting in altered T-cell activation and consequently an increased susceptibility to secondary infections [59] [60] [61] [58] [62]. Which (combination) of these mechanisms is relevant for post-septic immunosuppression is currently unknown. We asked whether sepsis-induced immunosuppression was long-lasting and aimed at determining the underlying mechanisms. Immunization of mice with the fungal antigen GliT 30 days after sepsis induction revealed a strongly reduced number of GLiT-specific Th-cells in post-septic mice, regardless of whether they had been treated with IL-7 or not. Importantly, on a per-cell basis the effector functions were not different between septic mice and control mice. The GliT specific Th cells from all three groups of mice were similarly capable of producing TNF-α, IFN- γ, IL-17, GM-CSF or IL-10. Moreover, within the GliT-specific Th cell population, the frequency of cells that were capable of producing two or three cytokines simultaneously was similar in all three groups of mice. These findings support an earlier report by Markwart et al. [63], who assayed the intrinsic functions of TCR-transgenic T cells in septic mice. Our data are in contrast to earlier reports on sepsis-induced T-cell intrinsic functional impairments [58] [12] [56]. This difference may be explained by the fact that we analyzed Th cell responses 30 days after sepsis rather than during the acute sepsis episode. Our data are best compatible with a loss of GliT-specific T cell clones, which was undetectable by NGS analyses of the TCR repertoire. It is currently thought that in a naive repertoire, each individual T cell clone occurs in about 102 copies. Our data are compatible with a loss of approximately 50% of the GliT-specific Th cells after sepsis. Such a loss would not be detectable with bulk methods, including NGS. Our data would also be compatible with immune suppression by IL-10 producing B cells and MDSC which are still increased in numbers 30 days after sepsis [16].

Our study demonstrates long-lasting impairments in CD4+ T-cell responses despite rapid recovery of T-lymphocyte populations after sepsis. Whereas numerical recovery of T lymphocytes occurred quickly after sepsis and was accelerated by IL-7 treatment, the Th cell response against a fungal antigen was impaired 1 month after sepsis, regardless of IL-7 treatment. The impaired response was due to a reduced number of antigen-specific Th cells, the remaining clones were functionally intact. Our findings show that functionally relevant losses of T cell clones may go unnoticed even by sensitive global assays such as next generation sequencing of T cell receptors. Moreover, our data indicate decreased T cell responses as long-term sequelae of sepsis and suggest that this lasting immunodeficiency contributes to the increased morbidity and mortality observed in sepsis survivors.