From dendritic cells to B cells dysfunctions during HIV‐1 infection: T follicular helper cells at the crossroads

Summary

T follicular helper (Tfh) cells are essential for B‐cell differentiation and the subsequent antibody responses. Their numbers and functions are altered during human and simian immunodeficiency virus (HIV/SIV) infections. In lymphoid tissues, Tfh cells are present in germinal centre, where they are the main source of replicative HIV‐1 and represent a major reservoir. Paradoxically, Tfh cell numbers are increased in chronically infected individuals. Understanding the fate of Tfh cells in the course of HIV‐1 infection is essential for the design of efficient strategies toward a protective HIV vaccine or a cure. The purpose of this review is to summarize the recent advance in our understanding of Tfh cell dynamics during HIV/SIV infection. In particular, to explore the possible causes of their expansion in lymphoid tissues by discussing the impact of HIV‐1 infection on dendritic cells, to identify the molecular players rendering Tfh cells highly susceptible to HIV‐1 infection, and to consider the contribution of regulatory follicular T cells in shaping Tfh cell functions.

Abbreviations

ART

antiretroviral therapy

Blimp‐1

B lymphocyte induced maturation protein 1

CCR

C‐C chemokine receptor

cDC

conventional dendritic cell

DCs

dendritic cells

FoxP3

forkhead box P3

GC

germinal center

HIV

human immunodeficiency virus

IFN

interferon

IL

interleukin

LCMV

lymphocytic choriomeningitis virus

LN

lymph node

pDC

plasmacytoid dendritic cell

PD‐L

programmed death‐ligand

PD

programmed death

SAMHD1

SAM domain and HD domain‐containing protein 1

SIV

simian immunodeficiency virus

Tfc

follicular cytotoxic T

Tfh

T follicular helper

Tfr

follicular regulatory T

TGF‐β

transforming growth factor β

Th

T helper

TLR

toll‐like receptor

Treg

regulatory T

Human immunodeficiency virus (HIV) cure and the development of an effective vaccine are dependent on a better understanding of the immune system in healthy individuals, as well as in HIV‐1‐infected subjects. Recent advance in the characterization of broadly neutralizing antibodies have proven their potency in preventing infection in animal models and also to decrease HIV‐1 viral load in humans.1 However, only a small proportion of HIV‐1‐infected individuals generate broadly neutralizing antibodies, and B‐cell functions are impaired during HIV‐1 infection.2

B‐cell maturation and differentiation toward memory and plasma cells is dependent on a subset of CD4⁺ T cells that express CXCR5 and migrate to B‐cell follicles, namely T follicular helper (Tfh) cells.3 The differentiation of naive CD4⁺ T cells toward Tfh cells is a multistep programme in which cell‐to‐cell contacts as well as cytokine environment are critical.4, 5 Mature antigen‐presenting cells migrate from the periphery to the lymphoid tissues to activate their cognate naive CD4⁺ T cells in the T‐cell zone. Upon activation, CD4⁺ T cells up‐regulate CXCR5 and migrate to the B‐cell zone where they will encounter cognate B cells. The T‐cell–B‐cell interactions through surface proteins provide further signals allowing pre‐Tfh cells to enter the B‐cell follicles and facilitate the germinal center (GC) reaction. Several cytokines, such as interleukin‐6 (IL‐6), IL‐21 and IL‐12 have been involved in the development of Tfh cells.4, 6

After considering the recent data on Tfh cell dynamics in HIV/simian immunodeficiency virus (SIV) infection, we herein discuss (i) the dynamics and functions of dendritic cells (DCs) during HIV‐1 infection, (ii) how lymph node (LN) B cells of HIV‐1‐infected individuals impair Tfh cell functions, and (iii) the involvement of newly described follicular regulatory and follicular cytotoxic T cells in shaping Tfh cells in HIV/SIV infection.

Tfh cell differentiation and functions are altered during HIV‐1 infection

Replication of HIV occurs mainly in lymphoid tissues,7, 8, 9 particularly in CD4⁺ T cells, although some infected macrophages have been observed.10 Following their discovery in the early 2000s, Tfh cells in the context of HIV and SIV infection have been under intense investigation. These cells are depleted during acute SIV/HIV infection11 whereas their numbers are higher in the chronic stages of the disease compared with uninfected individuals,12, 13, 14, 15, 16, 17, 18 before decreasing in AIDS subjects.19, 20 HIV‐1 treatment induces the normalization of Tfh cell numbers in LNs from patients,13 and no statistical differences are observed between HIV‐1‐infected treated individuals and non‐infected controls.12, 21 Similarly, Tfh cell frequencies in gut biopsies did not differ from those observed in treated HIV‐1‐positive and ‐negative individuals.22 Importantly, Tfh cells form an important reservoir during HIV/SIV infection.23

Expansion of Tfh cells in LNs from chronically HIV‐1‐infected individuals correlates with the viraemia13 and is associated with gammaglobulinaemia and skewing of B‐cell phenotype.12, 13, 16 Importantly, Tfh cells exhibit the highest levels of HIV‐1/SIV DNA and RNA compared with other LN CD4⁺ T‐cell subsets,13, 14, 18, 24 and replicative HIV‐1 virions can be recovered from activated Tfh cells isolated from viraemic patients.13 Eventually, sorted Tfh cells were shown to be highly susceptible to HIV‐1 infection.13, 21, 24 The susceptibility of Tfh cells to HIV‐1 infection compared with other LN or tonsillar CD4⁺ T cells is associated with a lack of SAM domain and HD domain‐containing protein 1 (SAMHD1) expression in those cells.21 Moreover, we showed that the expression of the restriction factor SAMHD1 decreases with T‐cell proliferation and inversely correlates with CD4⁺ T‐cell susceptibility to HIV‐1 infection.21 SAMHD1 is a triphosphohydrolase enzyme that controls the intracellular level of deoxyribonucleoside triphosphates. SAMHD1 plays a role in innate immune sensing and autoimmune disease, and restricts HIV‐1 infection; its expression is regulated with cell cycling.25 Whether the low expression of SAMHD1 is an intrinsic characteristic of Tfh cells, or is a consequence of their cycling capacity remains to be determined.

The function of Tfh cells is usually measured by their capacity to produce IL‐21. Although a decreased IL‐21 production by Tfh cells has been observed in untreated HIV‐1 infection16 or SIV‐infection,19 others observed higher levels of IL‐21⁺ Tfh cells in LNs from HIV‐1‐untreated individuals compared with controls.13 In our cohort, we found comparable frequencies of IL‐21⁺ CD4⁺ T cells between treated HIV‐1‐positive subjects and controls.21 The production of IL‐21 by Tfh is crucial for their ability to induce B‐cell maturation and differentiation.26, 27 B cells, in turn, also impact IL‐21 production by Tfh cells in HIV‐1 infection, as further discussed below.16 The ability to produce IL‐21 by Tfh cells during chronic HIV‐1 infection was recently associated with a preferential differentiation of Tfh cells towards a T helper type 1 (Th1) ‐like phenotype.28 The Th1/Th17 balance of Tfh cells has been shown to be important in several diseases.29 Importantly, we showed that LN percentages of CD4⁺ T cells producing IL‐17A were lower in HIV‐1‐infected patients under antiretroviral therapy (ART) compared with healthy subjects.21 In line with those results, C‐C chemokine receptor 6 positive (CCR6⁺) Tfh cells, which were shown to produce IL‐17,30 are depleted in SIV infection.28 As IL‐17A promotes GC B‐cell migration toward CXCL12 and CXCL13 in vitro,31 changes in IL‐17A levels may also play a role in the aberrant B‐cell responses seen in HIV‐1‐infected individuals. Indeed, blood memory PD‐1⁺ CXCR5^hi CCR6^hi “Tfh17” cells capable of providing help to B cells are decreased in chronic HIV infection, but they recover after ART.32 Further studies are needed to better define the physiological role of IL‐17 signalling in normal human GC B cells.

The reasons for Tfh cell accumulation during the chronic phase of HIV‐1 infection despite their high susceptibility to the virus are not clear. A better understanding of the dynamics of DC functions during HIV‐1 infection is necessary because those cells are the initiators of Tfh development, as we discuss below.

Impact of HIV‐1 on DCs

Due to their localization at the mucosa, where most HIV‐1 transmission occurs, DCs are thought to be crucial for the establishment of the infection.33 Upon uptake of HIV‐1 at the mucosa, DCs migrate to lymphoid tissues where they can transfer the virus to activated CD4⁺ T cells. DCs are sentinels distributed through the tissues and the blood. They are equipped with various receptors that render them able to sense HIV‐1 virions.34 Langerhans cells, present in the mucosa, are likely to play a role in the establishment of HIV‐1 infection.33 Plasmacytoid DCs (pDC) and conventional DCs (cDC1 and cDC2) are present in peripheral blood, and are recruited at the mucosal site upon inflammation. Both pDCs and cDCs express the class II major histocompatibility complex protein HLA‐DR, but pDCs display CD123 (IL‐3 receptor) whereas cDCs are defined phenotypically by the expression of CD11c. Functionally, cDCs act as antigen‐presenting cells, whereas pDCs respond to viral infection by secreting copious amount of type I interferon (IFN).

The necessary interaction between naive CD4⁺ T cells and DCs to drive Tfh cell differentiation comes essentially from studies in mice.35, 36 Recent studies in humans have shown that DCs have the capacity to induce Tfh‐like cells in vitro.37, 38 These studies point to a role for IL‐12, but also IL‐6 in driving the expression of Tfh cell markers, depending on the activation signals given to monocyte‐derived DCs. Transforming growth factor‐β (TGF‐β) acts also, together with IL‐12 and IL‐23, in the differentiation of Tfh‐like cells.39 In addition, a role for IL‐27 was reported in Tfh cell priming by DCs in vitro.40 Given the limited amount of human DCs in blood and the difficulties in accessing tissue DCs, data are scarce on the factors essential for the priming of human Tfh cells in vivo. A subset of DCs from human female reproductive tracts were shown to capture HIV‐1 virions ex vivo, which induced their activation and subsequently CD4⁺ T‐cell proliferation.41 The CD4⁺ T‐cell polarization by HIV‐1‐exposed and/or infected DCs remains to be evaluated.

Few reports have shown that the susceptibility of DCs to HIV‐1 infection is rather limited, but both CD11c⁺ cDCs and CD123⁺ pDCs could be infected ex vivo.42, 43 The low permissibility of DCs to HIV‐1 infection is most likely due to high expression of SAMHD1.44 The addition of the HIV‐2/SIV‐encoded protein Vpx in vitro, which inhibits SAMHD1, induces a much higher proportion of infection.

In vitro HIV‐1 culture, as well as in vivo HIV‐1 and SIV infections, lead to HIV‐1 uptake by pDCs and type I IFN release. The nucleic acids contained in HIV‐1 virions activate toll‐like receptor 7 (TLR7) in endosomes and induce the release of IFN‐α through interferon regulatory factor‐3 activation.34, 45 Plasmacytoid DCs are thought to be an important driver of immune activation through their release of type I IFN, and IFN‐α levels are elevated in HIV‐1‐infected individuals.46 The release of IFN‐α by pDCs upon culture with HIV‐142, 47 reflects the maturation of the cells and is accompanied by the expression of CD83 and CCR7, as well as the co‐stimulatory molecules CD80 and CD86.47 CCR7 expression allows pDCs to migrate toward lymphoid tissues. Although HIV‐1 does not directly induce cDC maturation in vitro, supernatant of pDCs cultured with HIV‐1 led to the increase in CD83, CD80 and CD86 expression on cDCs.47 Others have shown that HIV‐1 induces the production of IL‐6 and tumour necrosis factor‐α by cDCs,48 although the expression of maturation markers was only modestly increased.42, 48 Notably, studies in SIV models show that non‐pathogenic SIV infection of African green monkeys leads also to IFN‐α production, but is limited to the acute phase.45

Dynamics of blood and tissue DCs during HIV‐1 infection

Phenotypical studies of peripheral blood DCs have revealed that the levels of both cDCs (HLA‐DR⁺ CD11c⁺) and pDCs (HLA‐DR⁺ CD123⁺) are decreased in HIV‐1‐infected subjects.49, 50, 51, 52, 53, 54, 55, 56 Others showed that pDC levels were increased in non‐treated HIV‐1‐infected individuals with CD4 counts > 400 cells/μl, whereas they declined strongly in patients with AIDS.55 Blood dendritic cell antigen positive cDC1 levels were also found to be lower in infected subjects compared with HIV‐1‐negative controls, whereas similar levels of total CD11c⁺ cDCs were observed in the two groups.52 In most studies, low levels of CD11c⁺ and CD123⁺ DCs inversely correlated with viral load and/or CD4 decline.50, 51, 52, 56, 57 Longitudinal studies showed that ART initiation leads to an increase of both cDC and pDC subsets, although not reaching those of HIV‐1‐negative controls for the latter.58 Others, however, did not observe a normalization of peripheral DC numbers in HIV‐1‐infected individuals under ART.50, 51 Some reported an increase of cDC levels in HIV‐1‐infected individuals with CD4 T‐cell counts > 500 cells/μl compared with controls.59

Studies of SIV infection showed a similar decrease in pDC levels in peripheral blood,60, 61 whereas CD1c⁺ cDCs were at higher numbers compared with non‐infected animals60 but were also depleted in animals with AIDS.61 Longitudinal studies of SIV‐infected macaques showed a rapid increase of blood cDC and pDC subsets during the first week post‐infection in peripheral blood.62 Thereafter, during the advanced stages of the disease, DC proportions declined to lower levels compared with non‐infected animals.62

Lower numbers of circulating DCs during HIV/SIV infection are also associated with altered functions. Blood cDCs from viraemic HIV‐1‐infected individuals spontaneously secrete IL‐6 and IL‐12 ex vivo.57 Co‐stimulatory molecule expression is also increased. CD86 and CD40 expression was found at higher levels on blood HLA‐DR⁺ cells from HIV‐1‐positive individuals compared with controls, with the highest CD40 expression observed on cDCs from non‐treated HIV‐1‐positive individuals.51 Importantly, ART induced normalization of both CD86 and CD40 expression on total HLA‐DR⁺ cells.51 The ability of cDCs to stimulate allogeneic T cells is not altered between HIV‐1‐infected and negative individuals.52, 63 Similarly, in line with the decreased levels of pDCs, IFN‐α production in response to viruses measured in PBMCs is lower in HIV‐1‐infected individuals compared with controls.52, 64

While DC blood levels decrease during HIV/SIV infection these numbers were found at higher levels in lymphoid tissues from infected monkeys62, 65 and humans,53, 54, 66, 67 pointing to their recruitment into the lymphoid organs. As the disease progresses towards AIDS, however, SIV macaques display a depletion of DCs in LNs.61 The pDCs that are recruited to LNs, form clusters in the interfollicular regions (Fig. 1).66, 67 Clustering of pDCs was shown to inversely correlate with the CD4⁺ T‐cell count and to increase with progressing HIV‐associated lymphadenopathy.67

Figure 1.

Follicular helper T (Tfh) cell dysregulations during chronic HIV/SIV infection.

Limited data are available on the functions of LN DCs. Conventional DCs isolated from LNs of HIV‐1‐infected individuals were shown to spontaneously produce IL‐12 ex vivo, which increased upon TLR stimulation. Ex vivo tumour necrosis factor‐α production by cDCs and pDCs was low but also increased following TLR activation. The LN pDCs needed TLR stimulation to produce measurable levels of IFN‐α.54 In this study, the authors did not compare the functions of LN DCs from HIV‐1‐infected subjects with non‐infected controls, so limiting the extent of the findings. In contrast to circulating DCs, the ability of LN DCs to respond to TLR stimulation and supporting T‐cell responses ex vivo has been shown to be altered during SIV infection.68 This outcome results from a lower ability of pDCs and cDCs isolated from LNs of SIV‐infected macaques to produce IFN‐α and IL‐12 upon TLR stimulation compared with naive controls.68

In a recent study, skin DC numbers were similar in HIV‐positive and HIV‐negative individuals, confirming that the lower levels of DCs measured in the blood are not general.69 When measuring the infiltration of DCs in the skin after challenge with antigen, however, DC numbers were lower in skin from HIV‐1‐infected individuals compared with control, with the lowest numbers found in individuals with AIDS,69 suggesting that DC recruitment to inflamed tissue might be impaired during HIV‐1 infection.

More work is still needed to decipher the interplay between HIV‐1 and primary DCs from blood and tissues, especially in term of maturation, migration and T‐cell priming. Indeed, with the recent advances in DC biology and subset functions,70 it would be crucial to evaluate the susceptibility of these different DCs to HIV‐1 infection and their ability to induce Tfh cells upon HIV‐1 exposure.

How DCs may drive Tfh differentiation during HIV‐1 infection

Type I IFN production by pDCs as well as their recruitment to lymphoid tissues is a hallmark of HIV‐1 infection (Fig. 1). Interestingly, type I IFN induces expression of bcl‐6, CXCR5 and programmed death 1 (PD‐1) during naive CD4⁺ T‐cell activation, through the activation of signal transducer and activator of transcription 1.71 The authors showed, however, that IL‐21 production was not induced by IFN‐α/β.

It remains to be established whether the increased numbers of Tfh observed during chronic HIV‐1 infection are associated with pDC recruitment to the lymphoid tissues. Importantly, mice with persistent viral infection also display an accumulation of Tfh cells in lymphoid tissues.72 Indeed, a preferential differentiation of lymphocytic choriomeningitis virus (LCMV) ‐specific naive CD4⁺ T cells into Tfh cells is observed in the context of persistent LCMV infection. This imbalance toward Tfh cell differentiation is dependent on type I IFN, IL‐10 and programmed death ligand 1 (PD‐L1). Type I IFN stimulates the expression of PD‐L1 and the production of IL‐10 by DCs, which in turn drives naive CD4⁺ T cells towards a Tfh phenotype.72 Of note, IFN‐α has also been recently involved in B‐cell apoptosis and differentiation toward short‐lived plasmablast,73, 74 in a similar way to what has been described in HIV‐1‐infected patients.

Other cytokines that are found at increased levels in HIV‐1‐positive individuals favour Tfh differentiation. That is the case for IL‐6, which is required for the early stages of Tfh development.75 As mentioned, TGF‐β can participate to the differentiation of naive CD4⁺ T cells into Tfh cells in humans.39 The authors showed TGF‐β production by tonsillar CD11c⁺ cells in the T‐cell zone. Notably, TGF‐β levels are higher during HIV‐1/SIV infection and are associated with regulatory T‐cell accumulation and disease progression (Fig. 1).76 Furthermore, TGF‐β is an important cytokine driving the fibrosis of lymphoid organs during HIV‐1/SIV infection.77 It would therefore be important to analyse Tfh cell dynamics and functions in association with TGF‐β levels in HIV‐1‐infected individuals.

Alterations of GC B‐cell impact on Tfh cells during HIV‐1 infection

Despite higher levels of Tfh cells, B‐cell responses are impaired in HIV‐1‐infected individuals.2 Cognate Tfh–B‐cell interactions are essential for the GC reaction and PD‐1 expression by Tfh cells is important for appropriate GC B‐cell development. Although no difference in PD‐1 expression was found on Tfh cells from HIV‐1‐infected individuals compared with healthy controls,16 in vitro HIV‐1 infection of Tfh cells leads to a decrease in PD‐1 surface expression.24

The ligands for PD‐1, namely PD‐L1 and PD‐L2, are expressed by DCs, macrophages and also B cells. PD‐L1 expression by cDCs in the LNs of HIV‐1‐infected individuals was observed,54 as well as macrophages.68 Studies of SIV‐infected macaques have shown that while PD‐L1 is increased on LN B cells, their expression of PD‐L2 is decreased compared with non‐infected animals.19 PD‐L1 was also reported to be expressed at higher levels on GC B cells during HIV‐1 infection.16 Importantly, autologous Tfh‐B‐cell co‐cultures from SIV‐infected macaques or HIV‐1‐infected patients led to a defect in IgG production compared with co‐cultures of cells isolated from healthy subjects (Fig. 1).16, 19 In those studies, however, Tfh cells from infected subjects showed high levels of apoptosis, and it cannot be excluded that Tfh viability impacts the in vitro co‐culture experiments. Nevertheless, blocking PD‐L1/‐L2 improves IgG secretion of B cells co‐cultured with Tfh cells from HIV‐1‐infected individuals.16 The authors showed that the triggering of PD‐1 on tonsillar Tfh cells, leads to decreased cell proliferation, to a lower expression of inducible T‐cell co‐stimulator as well as a lower production of IL‐21 upon T‐cell stimulation (Fig. 1).16 Importantly, as anti‐PD‐1 treatment is thought to be a good candidate for immunotherapy in HIV‐1‐infected individuals, it would be critical to evaluate its effects on Tfh cells in this context.78

Dysfunction of Tfh cells is also characterized by decreased IL‐21 production in HIV‐1 infection16 or SIV infection.19 In contrast, we and others did not observe defects in IL‐21 production by LN CD4⁺ T cells.13, 21 Of note, IL‐21 treatment of SIV‐1‐infected macaques induced an increase in circulating memory B cells and in SIV‐specific IgG antibody levels in the serum compared with controls, without, however, affecting viral load.79

We reviewed previously the possible reasons for B‐cell dysfunctions during HIV‐1 infection.2, 80 Further investigations of GC B cells during HIV‐1 infection are needed, as the crosstalk between Tfh and B cells is crucial for an effective humoral response. In addition, an increase in plasma cells has also been observed during HIV‐1 infection81, 82 where they accumulate in the interfollicular regions.66 As recent evidence in mice showed an effect of plasma cells on Tfh cell differentiation,83, 84 it would also be important to evaluate the impact of plasma cells on Tfh cells in humans and during HIV‐1 infection.

Follicular regulatory T cells during HIV‐1 infection

Germinal centre reaction needs to be highly regulated to prevent the development of systemic autoimmune diseases, chronic inflammation, allergic responses and B‐cell malignancies.85, 86 Recent advances in the field have revealed specialized subsets of T cells necessary for the control of B‐cell responses in the follicle. These cells are called follicular regulatory T cells (Tfr) and share phenotypic characteristics with Tfh and conventional regulatory T (Treg) cells but are distinct from both. The Tfr cells originate from natural Treg precursors, unlike Tfh cells, which originate from naive CD4⁺ T cells. Tfr cells also differ from Tfh cells by expressing forkhead box P3 (FoxP3) and B lymphocyte induced maturation protein 1 (Blimp‐1). Tfr cells express high levels of CXCR5 (which directs them to the GC), PD‐1 and inducible T‐cell co‐stimulator, and also Bcl6 but at lower level than Tfh cells.87 Importantly, Tfr cells potently suppress both Tfh and B cells in the GC reaction (Fig. 1).88, 89, 90 After initial imprinting by DCs, Tfr cells either leave the LN and enter the circulation destined to become a memory‐like Tfr cell, or migrate to the B‐cell zone to become effector Tfr cells that suppress Tfh and GC B cells.91

Studies on the role of Tfr cells in viral infections have started to emerge. In response to influenza virus or LCMV infection, the percentage of Tfr cells contained within the CD4⁺ CXCR5⁺ gate in lymphoid organs corresponds approximately to 20–30% upon immunization. In contrast, the Tfr‐cell percentage drops to 6–8% in lung‐draining LNs after influenza virus infection or in the spleen after LCMV infection.91 Increased Tfr cell numbers were seen also during chronic hepatitis B and hepatitis C virus infection in humans.92 Tfr cell expansion seems to be a feature of many chronic viral infections as Tfr frequency appeared to increase in the spleens of chronically HIV‐infected untreated individuals.93 In a more recent study, however, it has been shown that Tfr cell numbers may actually decrease in SIV infection.94 The discrepancies in these findings might be due to the different gating strategies used to identify Tfh and Tfr cells.

The capacity of Tfr cells to suppress Tfh or B cells during HIV or SIV infection is still unclear but it has been reported that Tfr cells suppress Tfh‐cell activation as well as IL‐21 and IL‐4 production in in vitro experiments.95 In SIV infection studies, greater avidity of anti‐gp120 antibodies correlated with lower frequencies of Tfr cells,94 suggesting that during chronic infection Tfr cells may inhibit somatic hypermutation of B cells. Altogether, these data demonstrate an important role for Tfr cells in controlling antibody production during viral infection. More direct experiments are needed to assess the suppressive capacity of Tfr cells in B‐cell responses.

As mentioned above, Tfh cells serve as the main reservoir for HIV or SIV.23 Recent studies reported that Tfr cells can also be infected with HIV or SIV, possibly due to CXCR4 or CCR5 expression on these cells, and blockade of these chemokine receptors inhibited HIV infection of these cells in vitro.15, 94, 95 Hence, Tfr cells (as well as Tfh cells) may influence HIV and SIV infections by harbouring virus as well as by regulating B‐cell responses.

More recently, another CXCR5⁺ T‐cell subset that migrates to the GC and plays an important role in eradicating infected Tfh cells, has been reported.96, 97 These cells are called “follicular cytotoxic T cells” (Tfc cells) and exhibit high expression of molecules linked to memory development such as CD62L, IL‐7 receptor and T‐cell‐specific transcription factor 1, and low expression of granzymes and perforin relative to that of conventional cytotoxic T cells. The Tfc cells are able to control infections in B‐cell follicles and their lack results in exaggerated infection (Fig. 1).97 HIV‐specific CXCR5⁺ CD8⁺ T‐cell numbers were inversely correlated with viral load in HIV‐infected patients.96, 97 Similar data have been reported in SIV‐infected animals.98, 99 Hence, understanding how to direct the differentiation and function of Tfc cells might help in the design of new strategies for eliminating HIV reservoirs.

Conclusion

The Tfh cells are central to mount protective humoral responses against HIV‐1, as has recently been confirmed in an SIV model.100 The recent discovery of Tfr cells has changed our view about how antibody responses are regulated, which will help in the design of new strategies for controlling humoral immune responses. In light of the recent findings on passive immunotherapy by using broadly neutralizing antibodies that can boost host immunity to HIV‐1,101 the impact of such immune interventions on DC priming, and Tfh and Tfr cell functions should be considered. Uncovering the molecular players necessary for Tfh differentiation may help in the design of innovative vaccines and curative strategies against HIV‐1.

Disclosure

The authors declare that they have no competing interests.