Pathological alteration and therapeutic implications of sepsis-induced immune cell apoptosis

Sepsis is a life-threatening organ dysfunction syndrome caused by dysregulated host response to infection that leads to uncontrolled inflammatory response followed by immunosuppression. However, despite the high mortality rate, no specific treatment modality or drugs with high efficacy is available for sepsis to date. Although improved treatment strategies have increased the survival rate during the initial state of excessive inflammatory response, recent trends in sepsis show that mortality occurs at a period of continuous immunosuppressive state in which patients succumb to secondary infections within a few weeks or months due to post-sepsis “immune paralysis.” Immune cell alteration induced by uncontrolled apoptosis has been considered a major cause of significant immunosuppression. Particularly, apoptosis of lymphocytes, including innate immune cells and adaptive immune cells, is associated with a higher risk of secondary infections and poor outcomes. Multiple postmortem studies have confirmed that sepsis-induced immune cell apoptosis occurs in all age groups, including neonates, pediatric, and adult patients, and it is considered to be a primary contributing factor to the immunosuppressive pathophysiology of sepsis. Therapeutic perspectives targeting apoptosis through various strategies could improve survival in sepsis. In this review article, we will focus on describing the major apoptosis process of immune cells with respect to physiologic and molecular mechanisms. Further, advances in apoptosis-targeted treatment modalities for sepsis will also be discussed.

Immune cell alteration induced by uncontrolled apoptosis is a major cause of profound immunosuppression.
• Therapeutic perspectives targeting apoptosis through various strategies could improve survival in sepsis.

Open questions
• It is known that apoptotic depletion of immune cells is responsible for immune dysfunction; however, how do immune cells, including innate and adaptive immune cells, change in the course of sepsis?

Introduction
Sepsis is a major public health challenge worldwide owing to protracted inflammation, immune suppression, susceptibility to infections, and even death 1 . Despite improvements in the understanding of the pathophysiology of sepsis and therapeutic strategies, no specific therapeutic agent for the treatment of sepsis has been approved to date 2,3 .
Sepsis treatment, including antibiotic therapies, ventilator management, blood glucose maintenance, and resuscitative strategies, has rapidly progressed 4,5 , particularly the supportive therapies recommended by the Surviving Sepsis Campaign 6 . However, few novel effective therapies have been identified, and the incidence of sepsis has increased 7,8 , with approximately 31.5 million cases of sepsis, 19.4 million severe sepsis, and 5.3 million deaths reported annually 9 . Sepsis with subsequent multiple organ dysfunction remains the leading cause of mortality in hospitalized patients, and it is projected to increase at an alarming rate over the next two decades 10 . Even more alarming is the increasing rate of sepsis-associated intensive care unit (ICU) mortality, which is the most common cause of death in ICU patients; severe sepsis accounts for >50% of ICU mortality 11 . Patients with sepsis are also hospitalized longer and have an eight times higher risk of death during hospitalization than other inpatients 12 .
In general, the pathophysiology of sepsis is considered as an initial hyperinflammatory phase that lasts for several days followed by a more protracted immunosuppressive phase 13 . The current death distribution indicates peaks during the early phase, although at a lower magnitude, and another peak after 2-3 months that continues to increase over the next 3 years 14,15 (Fig. 1). However, varying definitions and ineffective clinical strategies have led to discrepancies in the incidence and mortality rates of sepsis 16 , and thus the criterion should be redefined 17 . In the third International Consensus for Sepsis and Septic Shock, sepsis was clinically defined as a life-threatening condition of organ dysfunction caused by a dysregulated immune response to infection 18 . These changes shift the focus of pathophysiology on sepsis-induced immune dysfunction and long-term outcomes of sepsis patients who survive from a fatal stage owing to sophisticated care in the ICU 19 . Sepsis can be considered as a race to death between the pathogens and host immune response, where the pathogens seek an advantage by incapacitating various aspects of host immunity 20 . Numerous studies have shown that a large number of patients who died of sepsis had unresolved opportunistic infections 21,22 and immunosuppressive features 10,23,24 (Fig. 2). Several mechanisms are responsible for sepsis-induced immunosuppression, including apoptotic depletion of immune cells, increased expression of negative costimulatory molecules, increased regulatory T (Treg) cell expression, expression of programmed cell death (PD)-1 on CD4 + T cells, and cellular exhaustion 23 . Of these, immune cell apoptosis has been increasingly recognized as a key factor in the pathophysiology of septic complications [25][26][27] .
Widespread lymphocyte apoptosis occurring in the lymphoid (spleen, thymus, and lymph nodes) and other organs 28 that results in impaired immune cell activity (including that of neutrophils, monocyte and macrophages, B cells, natural killer cells [NK cells], and Fig. 1 Immune response in sepsis. Early activation of both innate and adaptive immune response is involved in the pathogenesis of sepsis. The peak mortality rates during the early period (top red line) were due to overwhelming inflammatory response, also known as "cytokine storm," which comprises fever, refractory shock, inadequate resuscitation, and cardiac or pulmonary failure. Meanwhile, mortality at the later period is due to persistent immunosuppression with secondary infections that results in organ injury and/or failure. Although more sophisticated ICU care has improved mortality, patients still die at the later period or after several years owing to the persistent immunosuppression, immune dysfunction, or chronic catabolism dendritic cells [DCs]) is a crucial contributing factor to the development of the immunosuppressive phase of sepsis 29 . Multiple studies have reported that preventing immune cell apoptosis could markedly improve survival 10,30,31 . In this review, we will highlight the importance of sepsis-induced immune cell apoptosis, including the role of innate and adaptive immune cells in the pathogenesis of sepsis, alterations in their immune metabolic stage, and the clinical implications and potential therapeutic interventions.

Apoptosis in sepsis
Apoptosis is a tightly regulated form of cell death that is vital in both embryo implantation and development and turnover of tissues during maturation 32 . During sepsisinduced immunosuppression, apoptosis plays a pivotal role in the selection of immune cell populations and maintenance of functional immune responses 33 . Cell death in innate and adaptive immune systems benefits the host by downregulating inflammatory response in sepsis, but the extensive loss of immune cells may compromise the ability of the host to eliminate invading pathogens. Immune cell apoptosis in lymphoid tissues and gutassociated lymphoid tissues could cause marked depletion of immune cells, including monocytes and macrophages, DCs, NK cells, and B cells 34 , contributing to immune suppression or secondary infection 35 . The depletion of these immune cells is a universal finding in all age groups and is particularly noteworthy because it occurs during life-threatening infection when clonal expansion of lymphocytes should be occurring. Although circulating lymphocytes undergo significant apoptosis, no apparent apoptosis in the heart, kidneys, lungs, and other substantive organs occurs during the progression of sepsis 36 . Sepsis-induced apoptosis of immune cells could undermine host immunity through anergy, latent infection reactivation, and susceptibility to secondary infections 37 . Importantly, the magnitude of lymphocyte apoptosis is possibly valuable for determining the severity of sepsis. However, the activity of immune cells differs under diverse apoptotic signals during sepsis. Cellular death through apoptosis directly leads to microvascular dysfunction and organ failure during sepsis, with apoptotic immune cells contributing to secondary infection or immune suppression.

Mechanisms and consequences of apoptosis in sepsis
Although immune cell depletion is a crucial event in the pathology of sepsis-induced immunosuppression, the mechanisms responsible for this are not fully understood 38 . One theory is that sepsis could affect apoptosis- Fig. 2 Alterations in innate and adaptive immunity in the pathophysiology of sepsis. Early activation of innate immunity is the first line of defense against infection and plays a central role in the initiation of adaptive immunity. However, in sepsis, excessive immune responses lead to several alterations in innate and adaptive immunity that contribute to protracted immunosuppression and increase the risk for opportunistic infection. In some way, sepsis can be considered as a race to the death between host immune response and pathogens that seek an advantage by impairing the host immune defenses induced decrease in the number of DCs, which are the most potent antigen-presenting cells (APCs). This then leads to impaired innate and adaptive immune response 25 . In addition, the profound decrease in the number of some critical cells could influence adaptive immune responses between the innate and adaptive systems. The other theory is that some factors, including steroids, cytokines (tumor necrosis factor [TNF]-α, high mobility group box-1 protein, FasL, and heat shock protein), could regulate apoptosis by directly modulating the activities of caspase-8 in the death-induced signaling complex or by changing the levels of death and survival factors that control the Fas apoptotic pathway. Conversely, the release of antiinflammatory cytokines, such as interleukin (IL)-10 and transforming growth factor beta, could accelerate apoptosis.
This process ultimately leads to major consequences. First, with respect to immune response, excessive apoptosis causes massive loss of immune cells. For example, the depletion of macrophages and NK cells impairs microorganism clearance 39,40 , which leads to protracted inflammatory responses. The second major consequence is that uncontrolled apoptosis of immune cells results in immunological tolerance with anti-inflammatory properties 41 . During apoptosis, the release of pro-inflammatory cytokines is inhibited, but the secretion of antiinflammatory factors is activated, indicating a shift from T helper type 1 (Th1) to Th2 cytokine production 25,42 . The sepsis-induced functional and quantitative changes in immune cells result in lymphopenia, with progression of immune paralysis.

Apoptosis-induced lymphopenia during sepsis
Sepsis could affect the function of virtually all types of immune cells. In the following section, we will discuss these various immunosuppressive effects on the different cells of the innate and adaptive immune systems.

Neutrophils
Neutrophils are produced in the bone marrow (BM) and released into the circulation as the most prevalent and integral innate cell population. They are fundamental components of innate immunity and are essential for microbial eradication and for sepsis survival 43 . Neutrophils are the most abundant leukocyte in systemic circulation. They are also present in small amounts in the spleen, liver, and lung 44,45 apart from the BM, and thus they are critical for early immune response 46,47 . In addition, neutrophils may function as APCs and mediate between innate and adaptive responses in various pathological infections 48 .
In physiological conditions, neutrophils are constitutively pro-apoptotic 49 and are short-lived granulocytes that undergo energy-and caspase-dependent apoptosis within 24 h 46 . However, in the early stage of sepsis, the level of neutrophils increase rapidly owing to delayed neutrophil apoptosis 25 , as evidenced by the high neutrophil count observed in animals and sepsis patients only during the first 24 h of sepsis initiation 50 . This then leads to persistent neutrophil dysfunction, compounded by the release of immature neutrophils from the BM that culminates in neutrophil deficits in oxidative burst 51 , cell migration 52,53 , complement activation, and bacterial clearance 54 , all of which contribute to immune dysfunction and persistent inflammation (Fig. 3). In addition, immature neutrophils are produced and released from the BM, and their circulation aggravates the delayed apoptosis 25 . Thus the inhibition of neutrophil apoptosis in sepsis could be harmful to host immunity, and neutrophil infiltration into tissues could damage organ function 23,55 . For example, neutrophil infiltration in the lungs is a pathological hallmark of sepsis-induced acute lung injury or acute respiratory distress syndrome 56 . Increased circulating neutrophils result in a dysregulated immune response by releasing cytokines and reactive oxygen species at sites distal to the infectious focus, leading to multiple organ failure (Fig. 4). However, accelerated apoptosis resulted in inflammation resolution in several preclinical models of injury 57,58 .
Various mechanisms governing neutrophil apoptosis have been implicated in sepsis. Intravenous lipopolysaccharide or endotoxin could increase the levels of various gene transcripts in neutrophils, which result in the suppression of neutrophil apoptosis 59 . The delayed apoptosis of neutrophils is associated with disease severity 60 , whereas increased apoptosis is beneficial for fighting infection in hemorrhagic shock 61 . Meanwhile, delayed neutrophil apoptosis but accelerated apoptosis of other immune cells might impair host immune system by increasing the dephosphorylation of epithelial cell caspase-8 62 . Although delayed neutrophil apoptosis results in increased neutrophil counts, some functional deficit in neutrophils are evident during the early and late time points in sepsis patients. Further, some functions decline with protracted sepsis 63 . More interestingly, neutrophil deficiency hastens the development of nosocomial and secondary infections 64 , which is probably due to impaired bacterial clearance and altered pulmonary cytokine response 65 . Currently, neutrophil apoptosisinduced lymphopenia is given more attention for its involvement in the development of secondary infection and as a potential predictor of mortality at 1 year after sepsis 66 . One potential mechanism by which apoptosisinduced lymphopenia occurs is that some neutrophil subsets during sepsis could secrete excessive amounts of IL-10, which restrains the proliferation of T lymphocyte 67 . Another mechanism is that complex interactions between the neutrophils and complement system could also cause complement-induced innate immune damage during sepsis 68 . In addition, von Gunten et al. reported that the apoptosis response of neutrophils to a death stimulus (Siglec-9 cross-linking) varies significantly between patients and at different stages of septic shock 69 . The feasibility of neutrophil function and immature neutrophil count as predictors of sepsis has also been evaluated based on the delayed neutrophil apoptosis observed in septic animals and patients 70 .

Monocytes and macrophages
The impact of sepsis on monocyte subpopulations has been extensively studied. The most notable change during sepsis is that the impairment of bold monocytes from the patients initiates "endotoxin tolerance" and leads to poor outcome. Monocytes and macrophages are major components of the innate immune system and play pivotal roles in orchestrating host immune response during sepsis 71 . At the early phase of sepsis, two different lymphocytes secrete an increased level of pro-inflammatory factors and chemokines, which aggravate the inflammatory response 72,73 and could contribute to increased mortality rate 74 . As sepsis progresses, the excessive apoptosis of monocytes and macrophages could result in immunosuppression and the higher risk of secondary infection or mortality 22,75 . Furthermore, monocyte and macrophage dysfunction leads to decreased release of proinflammatory cytokines, such as TNF-α, IL-1β, IL-6, and IL-12, whereas the release of anti-inflammatory mediators, such as IL-1 receptor antagonist (IL-1ra) and IL-10, Fig. 3 Sepsis alters innate and adaptive immune cells. Sepsis-induced immune paralysis is characterized by immunological defects that impair host immunity. Lymphoid cell loss, often resulting in the diminished capacity to fight and eliminate pathogens, is a primary feature of immune suppression during sepsis. Altered immune cell function induced by uncontrolled apoptosis is a major cause of profound immunosuppression. Lymphocyte apoptosis, including that of innate immune cells and adaptive immune cells, is associated with a higher risk of secondary infections and poor outcome in various diseases. As shown here, sepsis rapidly triggers profound apoptosis in macrophages/monocytes, dendritic cells, NK cells, γδ T cells, CD4 + T cells, and B cells. However, apoptosis of neutrophils is delayed, and Treg cells are more resistant to sepsis-induced apoptosis. Immune cell depletion due to apoptosis is the primary mechanism of sepsis-induced immune suppression is neither impaired nor enhanced (Fig. 3) [76][77][78] . These changes indicate that intracellular signaling has shifted from production of pro-inflammatory cytokines toward anti-inflammatory mediators associated with nosocomial infections and increased mortality 14 .
Aside from accelerated apoptosis during sepsis, functional defects in monocytes also contribute to the pathophysiology of sepsis-induced immunosuppression, which is characterized by suppressed mononuclear cell HLA-DR expression in monocytes 79 . The decreased level of monocyte HLA-DR could be considered as a marker of monocyte anergy 80 , as it impairs the function of monocytes 81 and decreases lymphocyte proliferation in response to invading pathogens 82 . Notably, HLA-DR expression is a reliable predictor for the development of nosocomial infections and mortality during sepsis [83][84][85][86] , and immunotherapies targeting monocyte function in patients with sepsis are being developed 87,88 .
Macrophages are essential in maintaining and activating host inflammatory responses 89 . Traditionally, macrophages perform two most well-known polarizations: M1 (i.e., expression of inflammatory cytokines) and M2 (i.e., expression of factors that resolve or suppress inflammatory responses) 90 . Macrophage phagocytosis leads to an imbalance of pro-and anti-inflammatory cytokines, resulting in polarization of macrophages to an M2 phenotype. This is characterized by predominant release of anti-inflammatory cytokines, such as IL-10 and IL-1ra, and decrease of pro-inflammatory cytokines 91 . Collectively, sepsis-induced extensive depletion of macrophages Fig. 4 Sepsis-induced delayed apoptosis and recruitment of neutrophils into tissues lead to multiple organ dysfunction syndrome. The apoptosis of neutrophils is delayed during the first 24 h after the initiation of sepsis. Then neutrophils are recruited and infiltrate into tissues, aggravating the ongoing neutrophil dysfunction with persistent immune dysfunction and inflammation persistence. Neutrophil infiltration in the lungs is a pathological hallmark of sepsis-induced acute lung injury or acute respiratory distress syndrome as well as of organ dysfunction in the liver and heart. Representative histological changes in tissues are shown in hematoxylin and eosin-stained samples (original magnification ×400). Arrows denote the recruitment and infiltration of neutrophils into tissues could seriously impair host antimicrobial defenses, but anti-apoptotic therapies targeting macrophages through anti-apoptotic proteins, such as modulation of Bcl-2 family members, could effectively ameliorate the host response and decrease sepsis-induced morbidity and mortality 92 .

Dendritic cells
DCs arise from BM progenitors and seed the peripheral tissues as immature cells. They are the most potent APCs and play an essential role in pathogen recognition, regulation of immune response, and inflammation 93,94 by linking the innate and adaptive immunity 95,96 . DCs are classified into two types: conventional DCs (cDCs) and plasmacytoid DCs (pDCs) 97 . cDCs, such as monocytes, can also secrete IL-12, while pDCs are similar to plasma cells and secrete large amounts of interferon (IFN)-α. DCs, including plasmacytoid and myeloid DCs, are markedly reduced in the spleen and circulation during sepsis 98 . Notably, the loss of DCs seems associated with worse clinical outcomes in sepsis including death and nosocomial infections [99][100][101] . For example, the number of DCs was markedly reduced in both the spleen and mesenteric nodes at 12 h in septic models created via cecal ligation puncture 102 . Although the excessive depletion of DCs could induce accelerated differentiation of monocytes into DCs 103,104 , it does not compensate for the excessive depletion of DCs, and the number of DCs continue to decrease along with impaired functional capability 105 . The levels of surface molecules associated with the function of DCs, including CD40 and CD86 as well as major histocompatibility complex (MHC)-DR, markedly decrease, but the release of IL-10 tends to increase (Fig. 3) 96,106 . These changes lead surviving DCs to transform to tolerogenic DCs, which are DCs that are unable to induce an allogeneic T cell activation but lead to either T cell anergy or Treg cell proliferation 107 . Under this condition, DCs with immunosuppressive properties that undermine immune responses are lost, thus leading to failure of activation of the immune response of effector T cells and ultimately results to an immunosuppressive state and organ injury 105,108 . Thus the profound depletion of DCs has been considered as an early predictive biomarker for the outcome of sepsis 109 .
There are several mechanisms involved in the induction of DC apoptosis, including abnormal activation of the nuclear factor of activated T cells (NFAT) and an influx of extracellular Ca 2+ and calcineurin-dependent nuclear NFAT translocation 110 . Further investigation found that DCs profoundly affected the systemic impact of severe sepsis through Toll-like receptor (TLR) signaling pathways by increasing the expression of MHC class II antigen and costimulatory molecules CD80 and CD8 63 . Multiple reports have shown that blocking sepsis-induced DC apoptosis or augmenting DC function could enhance sepsis survival [111][112][113] , indicating that preventing DC apoptosis could be a promising treatment strategy for sepsis.

NK cells
NK cells have been heavily studied in sepsis due to their low number in the circulation but high numbers in tissues 114 ; thus their contribution to the pathophysiology of sepsis remains unclear 115 . Only 5-15% of blood lymphocytes do not express specific receptors 116 . Traditionally, NK cells are recognized as immune regulators and are divided into different subpopulations based on CD16 and CD56 expression 117 . In humans, NK cells are characterized as CD3 − NKp46 + CD56 + cells 118 . Under normal conditions, NK cells can induce a rapid, non-specific innate immune response against intracellular bacteria, pyogenic bacteria, fungi, protozoa 119 , and viral infections 120,121 through direct cytotoxicity against virusinfected cells and early production of cytokines that can inhibit viral replication. Further, NK cells are the major producers of IFN-γ 122 . In addition, NK cells play an important role in initiating host defense and coordinating innate and adaptive immune responses during sepsis 123 .
The most important change is that the accelerated apoptosis of NK cells leads to the decreased number of circulating NK cells 124,125 that lasts for several weeks during sepsis 126 . Both CD56 hi and CD56 lo NK cell subpopulations are altered, and this is associated with increased mortality in patients with sepsis 127,128 . More importantly, the cytotoxic function of NK cell is decreased 40 , which contributes to sepsis-induced immunosuppression (Fig. 3). Furthermore, the decreased level of IFN-γ caused by excessive apoptosis of NK cells increases the risk of secondary infection 129 .
The loss of NK cells directly affects the immune responses in sepsis patients and thus may be a potential target for therapeutic intervention 130,131 . For example, PD-1/programmed cell death ligand-1 (PD-L1) blockadebased immunotherapy can be used during sepsisassociated immunosuppression that develops due to the loss of protective function of NK cells 132,133 . Thus NK cell-based immunotherapy and immunomodulatory molecules specifically targeting NK cells and their immunometabolism may open a future avenue to target sepsis 134 .
Gamma delta (γδ) T cells γδ T cells are a distinct lymphocyte population with a unique and expansive function. They play a pivotal role in protecting tissues against bacterial, viral, and parasitic pathogen damage 135 . γδ T cells comprise a small subset of T cells that possess a distinct T cell receptor (TCR) on their cell surface, that is, a TCR comprised of one γ chain and one δ chain. These cells exhibit features of both innate and adaptive immunity and play an indispensable role in host defense, immune surveillance, and homeostasis 136 . γδ T cells contribute to both innate and acquired immune responses during sepsis, with IFN-γ, IL-17, and other chemokines being released after γδ T cell activation.
In sepsis patients, γδ T cells are activated with increased surface expression of CD69 and HLA-DR, but the number of circulating γδ T cells is significantly lower than that of healthy subjects 137,138 , and the reductions correlate with the severity of illness and highest mortality rates 139,140 . This is probably because the apoptosis-induced loss of γδ T cells in the intestinal mucosa results in higher susceptibility to secondary infections as pathogens invade the circulation or peritoneal cavity (Fig. 3) 13 . The number of γδ T cells in the peripheral circulation decreases by as much as 80% 141 . Liao et al. found that changes in the function-related markers of γδ T cells in the blood of sepsis patients and impaired IFN-γ expression by γδ T cells after antigen stimulation are associated with mortality in sepsis patients 142 . Furthermore, γδ T cell deficiency impairs the immune defenses and increases the mortality risk in sepsis 143 . Expansion of the γδ T cell population increased the resistance of immunodeficient mice to bacterial infection 144 . Collectively, these findings show that preventing γδ T cell apoptosis could attenuate inflammatory responses 145 , providing new insight in the understanding of the functions of γδ T cells in sepsis.

CD4 + T cells and associated subpopulations
T lymphocytes are key elements in all adaptive immune responses. CD4 + T cells are among the most important peripheral lymphocyte subsets in terms of modulating successful immune responses, influencing innate and adaptive immune cells through cytokine production, and cell-to-cell interaction 146,147 . When presented with peptide antigens, mature CD4 + T cells become activated and rapidly divide into several subsets that facilitate various immune responses by differing cytokine generation and secretion upon activation 148 . Among these cell subsets, Th1, Th2, and Th17 cells are major representatives of Th cells and are further explored here.
Apoptosis of T lymphocytes is critical in the pathophysiology of sepsis, and CD4 + T cells could directly mediate the host response to sepsis 149 . The number of T lymphocytes undergoing apoptosis is significantly reduced during sepsis and is even higher in non-survivors than survivors 150 . One of the most notable T cell defects induced by sepsis is the development of apoptosis, which destroys the CD4 + T cell population (Fig. 3) 151,152 . Uncontrolled apoptosis of CD4 + T cells causes marked lymphocytopenia, which is particularly serious because clonal expansions are critical to overcome potentially lethal infections. Patients who died of sepsis were found to have a much greater magnitude of CD4 + T cell apoptosis than survivors 23 .
Several treatments to prevent immunosuppression have targeted the apoptosis of CD4 + T cells in animal models 153,154 . Of note, IL-7 has emerged as a promising therapeutic agent because it has been found to be essential in preventing T cell depletion 155 . IL-7 administration has been shown to effectively improve T cell viability and trafficking and release of IFN-γ and restore the delayedtype hypersensitivity response to recall antigens 156 , which improves survival in sepsis. Another promising approach in reversing immunosuppression in sepsis involves blockade of the co-inhibitory molecules PD-1 and PD-L1. PD-1 blockade has been reported to increase the release of IFN-γ and prevent T cell apoptosis in patients with active infections 157 . However, the mechanisms inhibiting T cell apoptosis during sepsis are complex and yet to be completely understood 158 .
Not only the number of CD4 + T cell populations but also the function of the remaining lymphocytes is reduced during sepsis. Sepsis leads to uncontrolled apoptosisinduced depletion of CD4 + T cells, and some remaining cells are rendered dysfunctional or exhausted due to the prolonged exposure to excessive pro-and antiinflammatory cytokines. CD4 + T cell exhaustion has been typified by T cells that have severely impaired effector functions and has been found in patients with sepsis. The prolonged duration of sepsis is characterized by high antigen load and elevated pro-and antiinflammatory cytokines, which is conducive for T cell exhaustion 159 . The association between T cell exhaustion and increased mortality in sepsis is due to immune paralysis and secondary nosocomial infections 160,161 . Multiple independent studies have reported that blockade of the PD-1/PD-L1 pathway could attenuate T cell exhaustion, increase IFN-γ production, prevent apoptosis, and improve survival in various pathologic models of sepsis [162][163][164] . Collectively, these findings indicate that T cell exhaustion is also a major etiology of immune dysfunction in sepsis and that reversal of putative T cell exhaustion is a promising modality in the treatment of sepsis.

B cells
B cells play an important role in both adaptive and innate immune response 165 . Under physiological circumstances, activated effector B cells could differentiate into plasma cells or memory B cells and drive the humoral immune response and act as APCs to activate effector T cells 166 . B cell function is relegated to the production of antibodies (Abs) and the development of memory plasma B cells 165 . Multiple studies have documented that impaired B cell function 167,168 , along with the production of IL-10 169,170 , the presentation of microorganism antigens to T lymphocytes 171 , and interaction of several bacterial products with B cells 165,172 , weakens immunity during sepsis. The relative proportion of B cell subsets is decreased on admission day in critically ill patients with sepsis, indicating that the apoptosis of B cells was induced 173,174 . In addition, endotoxemia causes a transient depletion of memory B cells and regulatory B cells from the circulation, but the functional capacity of B cells to produce IL-10 is maintained (Fig. 3) 175 .
Excessive apoptosis of B cells in sepsis likely involves multiple pathogen-sensing receptors and redundant signaling pathways. For example, Octavia et al. found that myeloid differentiation primary response gene 88 (MyD88) was effective in blocking the apoptosis of B cells, and MyD88 deficiency could markedly decrease B cells apoptosis, but the mortality rate significantly increased 176 . However, administration of Tubastatin A, a selective inhibitor of histone deacetylase 6, restored the percentage of B lymphocytes and significantly increased the percentages of innate immune cells and macrophages 177 . Therefore, therapies targeted at reversing B cell depletion should be actively investigated.

Treg cells
Treg cells are a component of adaptive immunity that suppress responses of other effector T cell subsets, helping to maintain tolerance to self-antigens and suppression of autoimmune disease 178 . Treg cells are involved in the maintenance of peripheral tolerance and control of the immune processes 179,180 . Furthermore, Treg cells play a key role in controlling inflammation in many infectious diseases, including sepsis 180,181 . In sepsis and critical illness, Treg cells are detrimental to the proliferation and functional activity of effector T cells and other immune cells; for example, Treg cells inhibit both monocyte and neutrophil function 82 .
The number of circulating Treg cells increase in septic shock, particularly in the early stages after initiation. Further, although this increase was observed immediately after the onset of sepsis, it persisted only in those who subsequently died 182 . Further study revealed that the increase in Treg cells is caused by the loss of effector T cells rather than an absolute increase in Treg cells 183 . The involvement of these cells in long-term sepsisinduced immune dysfunction 184 may be attributed to inhibition of monocyte function through a pro-apoptotic mechanism involving the Fas/FasL pathway 82 . Treg cells are less vulnerable to sepsis-induced apoptosis; therefore, the percentage of Treg cells increases in patients with sepsis 185,186 . Cell apoptosis is a continuous state, but Treg cells are more resistant to sepsis-induced apoptosis than other T cell subpopulations 1,187 , thus the increased ratio of Treg cells during the early period after sepsis. Increased Treg cell population would prevent recovery of the immune system from excessive immune responses. Importantly, these resistant effects are possibly due to TLR4 deficiency and associated NF-κB signal pathways 188 .
The higher number of Treg cells in sepsis patients impairs immunity and contributes to secondary infections and mortality by acting both on innate and adaptive immune cells (Fig. 3). However, the role of Treg cells in septic injury is yet to be clarified. Although prevention of Treg cell apoptosis inhibited immune responses or decreased survival in animal models of sepsis, the removal of Treg cells via anti-CD25 monoclonal Ab administration did not improve survival in animal model 189 . This may be because not only Treg cells but also other T cell Table 1 Immune modulators and proposed benefits for sepsis-induced apoptosis therapy   Immune modulator Effects Reference G-CSF Improve neutrophils and monocyte production and release 194 GM-CSF Activate and induce production of neutrophils and monocytes or macrophages and reduce cell death 195,196 IFN-γ Increase monocyte expression of HLA-DR, increase numbers of IL-17 producing CD4 + T cells 197,198 PD-1/PD-L1 Anti-apoptotic effects to prevent loss of protective function of NK cells Prevent lymphocyte apoptosis and reverse monocyte dysfunction 132,133,199,200 IL-7 Blockade of sepsis-induced apoptosis depletion, increase production of CD4 + T and CD8 T cells Tim-2-specific antibody Decrease lymphocyte apoptosis and reverse the macrophage function 202,203 Ulinastatin Increase apoptotic rate of Treg cells and reduce the percentage through NF-κB pathway, ameliorate mortality 188 CTLA-4-specific antibody Improve overall sepsis-induced lymphocyte apoptosis and survival of secondary fungal infections 163,204 G-CSF granulocyte colony-stimulating factor, GM-CSF granulocyte-macrophage colony-stimulating factor, IFN-γ interferon gamma, PD-1 programmed cell death-1, IL interleukin, CTLA-4 cytotoxic T lymphocyte antigen-4 subpopulations are depleted. In conclusion, Treg cells are more resistant to apoptosis in sepsis, thus potentially serving as targets for immune modulation.

Clinical perspectives
Given the profound immunosuppression induced by depletion of immune cells that occurs during sepsis, the ability to sequentially follow the uncontrolled lymphocyte apoptosis as a means to evaluate the efficacy of immuneadjuvant therapies provides promising novel therapeutic opportunities 190 . The potential of preventing lymphocyte apoptosis as a treatment strategy in sepsis has been supported by many animal model studies. The first study to report that prevention of apoptosis improved the survival in sepsis showed that Bcl-2 overexpression is more resistant to sepsis-induced apoptosis 191 . Furthermore, an increasing number of immunoadjuvant therapies to prevent sepsis-induced immune paralysis have been identified as apoptosis dependent. IL-7 and anti-PD-L1 have been found to have potent effects to prevent lymphocyte apoptosis. IL-7 administration is an attractive therapy in sepsis, because it blocks sepsis-induced apoptosis of immune effector cells and increases IFN-γ, a cytokine that is critical for an effective host response against invading pathogens. This presents a potential novel strategy in the treatment of patients with sepsis by restoring adaptive immunity. Such immune-based therapy should be broadly protective against numerous bacterial and fungal pathogens 192 . More specifically, other agents, such as granulocyte colony-stimulating factor, granulocyte-macrophage colony-stimulating factor, IFN-γ, IL-15, anti-PD-1/PD-L1, and anti-B and T lymphocyte attenuator will target the immunosuppressed state in critically ill patients (Table 1) 193 . Therefore, development of a non-invasive imaging modality to detect and serially follow apoptotic cell death could be helpful in evaluating the efficacy of immunoadjuvant therapies in sepsis patients. This provides a bright prospect of autophagic modulation in clinical application.

Conclusion
Impaired apoptosis aggravates sepsis-induced immunosuppression in both innate and adaptive immune systems. Thus manipulating apoptosis could be a new therapeutic approach in sepsis. Further, exploring potential therapeutic targets related to apoptosis will be valuable in reversing sepsis-induced immunosuppression.