The patterns of Ag-induced cytokine coexpression in normal, in vivo-primed CD4 memory T cells has remained controversial because the low frequency at which these cells occur has effectively prevented direct ex vivo measurements. We have overcome this limitation by using two-color cytokine enzyme-linked immunospot assays and computer-assisted image analysis. We found CD4 memory cells that simultaneously expressed IL-2, IL-3, IL-4, IL-5, and IFN-γ to be rare (0–10%). This cytokine segregation was seen in adjuvant-induced type 1, type 2, and mixed immunity to OVA, in Leishmania infection regardless of the Ag dose used or how long after immunization the assay was performed. The data suggest that type 1 and type 2 immunity in vivo is not mediated by classic Th1 or Th2 cells but by single-cytokine-producing memory cells.

At present it is unclear how Ag dose-dependent T cell functions, such as cytokine production, reflect TCR affinity and how the signal strength afforded by the Ag dose affects the kinetics of cytokine production by the individual T cell. We used a computer-assisted ELISPOT approach to address these issues. IFN-γ release by a clonal population of CD4 T cells was monitored on a clonal population of APC while titrating the nominal peptide. The frequency of cytokine-producing cells, the net per-cell output of cytokine, and the onset of cytokine production were each found to be functions of the signal strength. Sigmoidal dose-response curves were seen at the clonal population level, but the activation thresholds for the individual T cells followed a Gaussian distribution. Moreover, the overall dose-response curve of the T cell clone revealed cyclic changes, becoming increasingly shifted toward lower Ag concentrations with the duration of time that elapsed since the last restimulation with Ag. Therefore, responsiveness to Ag (“functional avidity”) is not a constant parameter of a T cell clone but a function of the T cell’s history of last Ag encounter. The implications of such shifting activation thresholds are discussed for autoimmune disease.

Direct measurements of the frequency and the cytokine signature of the neuroantigen-specific effector cells in experimental allergic encephalomyelitis (EAE) are a continuing challenge. This is true for lymphoid tissues, and more importantly, for the CNS itself. Using enzyme-linked immunospot analysis (ELISPOT) assays, we followed proteolipid protein (PLP) 139–151-specific T cells engaged by active immunization of SJL mice. The total numbers of PLP139–151-specific CD4 cells were highest before disease onset. At this time, these cells resided in lymphoid and nonlymphoid tissues, but were not detected in the CNS. While the PLP139–151-specific cells reached high frequencies in the CNS during clinical EAE, in absolute numbers, less than 20% of them were present in the target organ, with the majority residing in the periphery throughout all stages of the disease. The numbers of PLP139–151-specific cells gradually declined in both compartments with time. While eventually this first wave of effector cells completely disappeared from the CNS, PLP178–191-specific cells became engaged, being detected first in the CNS. These data suggest that throughout all stages of EAE, the effector cells in the CNS are recruited from a vast peripheral reservoir, and that the second wave of effector cells is engaged while the first wave undergoes exhaustion.