Memory CD4 T Cell Distribution and Recall Function in Nonlymphoid Organs

The CD4⁺ or Th cell subset has been studied predominantly in the context of its roles in promoting adaptive immune effector functions. The essential role of CD4⁺ Th cells for B cell differentiation, germinal center, and memory B cell formation was identified in the 1980s. The generation of CD4-deficient mice in the 1990s confirmed a key function for CD4⁺ T cells in protection from infection due to Ab production (1) and revealed their important role in promoting differentiation of CD8⁺ T cells into cytolytic effector cells and long-lived memory T cells (2). These Th cell interactions occurred in lymphoid organs during priming and involved interactions in specialized follicular structures. Although functional diversity of CD4⁺ T cells in the production of different cytokines was well appreciated, considerably less was known about the role of non-Th CD4⁺ T cell effector functions and their protective capacities.

The classic way to induce and study Ag-specific CD4⁺ T cell responses, as I learned during my postdoctoral training in the 1990s, was to immunize mice in the footpad with a peptide or protein Ag plus CFA, wait 10 d, then harvest the enlarged popliteal lymph nodes (LNs) in the leg, where the activated T cells had expanded. A typical mouse harvest at that time involved collecting the draining LN, which contained the majority of responding T cells, and to discard the remainder of the mouse. LN cells were cultured in vitro and stimulated with Ag to assess the specific immune response by ³H-thymidine incorporation and ELISA-based measurements of cytokines in the culture supernatants. Longer-term experiments to identify memory CD4⁺ T cells mostly involved harvesting the spleen, as well as the LNs, because their numbers were much lower and difficult to detect. We didn’t consider that CD4⁺ T cells would be functioning in any other sites.

The discovery by Sallusto and Lanzavecchia in the late 1990s of human memory CD4⁺ T cell subsets that differ in their expression of LN homing receptors revealed a memory subset with the potential to recirculate through nonlymphoid sites (3). However, tracking the migration and fate of Ag-specific CD4⁺ T cells in vivo following their activation was not possible in a polyclonal system, given the low-level expansion and low frequency of memory populations. The development of TCR-transgenic (Tg) mice provided a means by which to follow an Ag-specific response at a higher precursor frequency. Marc Jenkins and his group at the University of Minnesota developed a model for tracking TCR-Tg T cell activation to Ag stimulation using the transfer of graded numbers of TCR-Tg T cells into congenic mice, followed by Ag priming with protein or peptide plus LPS (4). In the first Pillars of Immunology article, Dr. Jenkins’s group applied innovative imaging to this system for investigating tissue homing of CD4⁺ effector and memory T cells within the whole mouse (5).

In the Jenkins study, mice received OVA-specific TCR-Tg (OT-II) CD4⁺ T cells and were immunized i.v. with OVA peptide alone or together with LPS. These mice were examined during the acute response (days 5 and 6) and following establishment of memory (60 d). Rather than using flow cytometry to assess the distribution of TCR-Tg T cells, which required digestion of multiple tissues and was done for more abundant memory CD8⁺ T cells in a concurrent study published by Leo Lefrançois’s group (6), Jenkins took an unbiased whole-mouse imaging approach for direct examination of potentially rare cells. They froze the mice, embedded them in optimal cutting temperature compound (as typically done for individual tissues), stained 10-μm whole-body sections, and computationally integrated results from multiple slices to generate whole-mouse immunofluorescence images. In nonimmunized control mice, the authors identified TCR-Tg cells (stained red) localized in the spleen. However, in mice immunized 5 d previously with peptide + LPS as an adjuvant, red Ag-specific T cells had significantly expanded and were detectable in multiple sites, including lung, intestines, liver, salivary gland, spleen, and even thymus (5). At day 60 postimmunization, Ag-specific T cells were still present in the intestines, spleen, and lung, albeit at reduced numbers compared with the acute response time point. Moreover, these memory T cells only persisted if the mice were primed with peptide + LPS and not peptide alone, confirming the requirement for a second inflammatory signal in memory formation.

This in vivo identification of effector and memory CD4⁺ T cells in multiple nonlymphoid sites provided compelling evidence that CD4⁺ T cells could function outside lymphoid organs. Interestingly, the study showed that nonlymphoid memory CD4⁺ T cells produced more IFN-γ than did lymphoid memory CD4⁺ T cells, which produced primarily IL-2 (5), suggesting protective roles for CD4⁺ T cells in diverse peripheral sites. Although the Jenkins model Ag system was not set up to examine protection, a study published that same year by David Woodland’s group at the Trudeau Institute, the second Pillars of Immunology article, demonstrated in vivo memory CD4⁺ T cell generation and protective capacity in an infection model (7). Woodland’s group showed that following respiratory infection with Sendai virus, memory CD4⁺ T cells could be found in the lungs, airways (bronchoalveolar lavage [BAL]), spleen, and LNs, with those in lungs and airways producing more IFN-γ than in lymphoid sites (7). To address whether nonlymphoid memory CD4⁺ T cells could mediate protection at the site, they isolated memory CD4⁺ T cells from the BAL of mice previously infected with Sendai or influenza virus, transferred them intratracheally (i.e., directly into the airways) into naive congenic mice, and subsequently challenged these mouse hosts intranasally with Sendai virus. Importantly, the mice that received intratracheally administered BAL memory T cells from Sendai virus–infected mice exhibited lower viral titers 4 d post-challenge compared with mice that received noncognate (flu-specific) or no T cells (7). These findings demonstrated that memory CD4⁺ T cells could mediate in situ protection within a nonlymphoid organ, and specifically in the lungs.

Together, these Pillars of Immunology articles revealed two new features of CD4⁺ T cells beyond their helper functions in lymphoid tissues: (1) their ability to localize in multiple sites, including lymphoid organs and nonlymphoid tissue, and (2) the potential for CD4⁺ T cells to coordinate protection in nonlymphoid sites. Although both studies interpreted the presence of memory CD4⁺ T cells in diverse anatomic sites as active migration and surveillance throughout the body, subsequent studies showed that in respiratory infection models, memory CD4⁺ T cells in the lung could take up residence and mediate protection at the site (8, 9). Accordingly, these studies provided a foundation for the discovery of tissue-resident memory CD4⁺ T cells in mice and humans (10). Mechanisms for how memory CD4⁺ T cells coordinate protection in nonlymphoid sites remain active areas of investigation. The pioneering work by Jenkins and Woodland and others moved the field toward a deeper understanding of T cell immunity as being intricately linked to tissue location and helped move the field beyond a lymphoid-centric view of the Th lineage. Current studies incorporate these more comprehensive assessments of immune memory across the body in mice and humans, and we are no longer throwing out major memory reservoirs with the mouse.