Memory cells sound the alarm

In a finding that could have implications for vaccine design, memory immune cells at mucosal surfaces have been shown to respond to encounters with pathogens by issuing signals that recruit other memory cells to the site.

The immune system clears pathogens it has encountered before more efficiently than those of an initial infection because specialized cells, known as memory cells, can remember and rapidly eliminate them. This process is the bedrock of all vaccines. Writing in Nature Immunology, Schenkel et al.¹ show that, in mice, a small number of memory cells that reside near sites of pathogen entry sound an alarm that recruits more memory cells from the blood to rapidly boost defences at the front line of a subsequent infection.

Most infections are initiated by pathogens that breach epithelial surfaces, such as the skin and the mucous membranes that line the genital, respiratory and gastrointestinal tracts. The immune system responds through the orchestrated actions of innate and adaptive immune cells. Innate cells are distributed throughout the body in large numbers; they can immediately identify and respond to pathogens, but they lack specificity and often fail to control an infection on their own. However, the B cells and T cells of the adaptive immune system have surface receptors that allow them to respond in a specific manner; they also have diverse mechanisms to eliminate infections. For example, in the case of CD8⁺ T cells, receptor binding by a specific pathogen activates the cells such that they start producing immunomodulatory cytokine proteins and acquire a killing function. However, adaptive cells that are specific to any one pathogen are exceedingly rare and, to function, they first need to proliferate. This takes time, and so, in the meantime, innate immune cells attempt to control pathogen spread.

An important function of innate cells is to produce chemokines, a subset of cytokines that attracts other immune cells to the site of infection (Fig. 1a). In addition, some innate cells travel through the 'highways' of the lymphatic system to bring pathogens or pathogen debris to nearby lymph nodes and other lymphoid organs, where naive B and T cells (those that have not yet been activated) normally reside. Here, the few B and T cells that recognize the pathogen undergo massive proliferation, become activated and migrate to the site of initial infection. For pathogens that remain at this site, such as papillomaviruses or the influenza virus, this response may be enough to eliminate the infection. However, many pathogens, including the viruses that cause chickenpox, smallpox and measles, can proliferate in the lymph nodes and escape through lymph vessels to access the circulatory system and invade other organs^2,3. In such cases, CD8⁺ T cells must enter those organs to eliminate the pathogen.

Figure 1: Cytokine alarms accelerate memory T-cell recruitment.

a, During a primary immune response, innate immune cells in the body's epithelial surfaces attempt to control the pathogen, and secrete chemokine proteins that sound an alarm to cells of the adaptive immune system. If the pathogen breaches this first line of defence and enters the lymph nodes, it will activate the few naive CD8⁺ T cells that specifically recognize it. These cells proliferate and are attracted to the site of infection by the chemokines. b, After infection, some pathogen-specific CD8⁺ T cells survive in lymphoid organs and as circulating or tissue-resident memory (TRM) T cells. During a secondary response, TRM cells act rapidly at the site of infection and, if the pathogen enters the lymph node, resident memory CD8⁺ T cells are activated, and rapidly proliferate to curb further pathogen spread from the lymph node and to migrate to infected tissues. Schenkel et al.¹ show that, in addition to their direct function on infected cells, TRM cells send out an alarm. They secrete the cytokine IFN-γ, which induces chemokine production by the surrounding cells to draw circulating memory CD8⁺ T cells into the infected tissue before the response in the lymph node occurs.

Although most of the responding CD8⁺ T cells die after an infection subsides, a fraction of pathogen-specific, but deactivated, cells persists at frequencies that are substantially higher than before the infection. Some of these memory cells take up residence in lymphoid organs, others circulate in the blood and yet others remain in non-lymphoid tissues as resident memory T cells (T_RM cells)^4,5. T_RM cells can be found in skin and mucosal tissues, and are able to rapidly secrete cytokines and kill infected cells on recognition of a pathogen⁴ (Fig. 1b). Thus, these memory cells are spatially and temporally poised to be the first line of defence against reinfection. Studies in mice have shown^6,7 that T_RM cells in the skin and vaginal tract can limit viral spread during localized infection. Now, Schenkel et al. show how T_RM cells communicate with circulating memory cells to swell the cellular army for battle at the site of a recall infection.

Schenkel and colleagues provide strong evidence that, like innate immune cells, T_RM cells in the vaginal mucosa have an alarm function — by rapidly producing the cytokine IFN-γ, the T_RM cells induce surrounding innate immune cells and vascular cells to express chemokines. One of these chemokines is CXCL9, a molecule that has been shown to attract T cells to the vaginal tract⁶. The authors observed that, by sounding the IFN-γ alarm, T_RM cells rapidly recruit large numbers of additional memory CD8⁺ T cells from the blood to the site of infection. Notably, the T_RM-cell-induced chemokine response was faster and larger than that of an innate response. Furthermore, only memory CD8⁺ T cells, which are poised for a rapid response, were recruited; naive CD8⁺ T cells were not. It should be noted, however, that after immunization and rechallenge of mice with the vaccinia virus, T_RM cells in the skin reduce viral proliferation in the absence of circulating T cells⁷; this suggests that the alarm function is absent or dispensable in some infection circumstances. Thus, future studies should determine the relevance of this alarm function in different tissues and disease settings.

By activating their own effector functions and calling additional memory CD8⁺ T cells to the site of infection, it is possible that T_RM cells not only directly and indirectly decrease infection at the primary site, but also reduce pathogen spread to the lymph nodes. Memory CD8⁺ T cells in the lymph nodes curb the spread of viruses to the blood and other organs^8,9. By reducing spread to lymph nodes, T_RM cells might reduce some of their burden. In the end, the more roadblocks the immune system creates to prevent pathogen spread, the fewer chances there are for a pathogen to cause disease.

Vaccines that induce T_RM cells could prove to be effective against pathogens that are resistant to existing vaccine strategies. Indeed, work in mice has shown⁶ that, after vaccination, a population of T_RM cells specific for herpes simplex virus 2 can be established in the vaginal mucosa by topical application of CXCL9, and that this results in reduced spread of this virus after subsequent infection. Schenkel and colleagues' findings suggest that, for an optimal front-line response, vaccines should induce both a T_RM-cell population and circulating memory T cells that can respond to the T_RM cells' alarm.