Flexibility accompanies commitment of memory CD4 lymphocytes derived from IL-4 locus-activated precursors

Abstract

Differentiation of T helper (Th) subset 2 effector lymphocytes is thought to foreclose on IFN-gamma gene expression. Using an IL-4 locus modified to detect transcriptional induction of this effector cytokine gene in developing Th2 cells, we show here that these cells contributed effectively to a long-term memory population. A memory CD4 subset formed efficiently from an activated population after transcriptional induction of the IL-4 locus and differentiation into an IL-4-producing subset with Th2 characteristics. Memory lymphocytes derived from Th2 cells with IL-4 locus activation remained committed to transcriptional competence of Th2 cytokine genes when reactivated and cultured under strong Th1-polarizing conditions. This commitment to transcriptional competence at Th2 cytokine gene loci upon recall activation indicates that linear differentiation is a substantial component of type 2 memory. Strikingly, however, descendants of the Th2 population could turn on IFN-gamma expression when reactivated after a quiescent period, revealing an unexpected flexibility allowing activation of the forbidden IFN-gamma gene after reactivation and growth.

Fig. 1.

Adoptive transfers using a knockin allele to detect IL-4 locus activation. (A) Schema of overall experimental design and donor cell preparation for sorting and transfer. (B) GFP profile of naive donor 4get⁺/WT DO11.10 cells. These cells were Ag activated, cultured under Th2 and Th-null (ThU) conditions, and purified as IL-4-locus-active (GFP⁺) and IL-4-locus-inactive (GFP⁻) populations. (C) Sorted cells were then expanded [Th2 (for IL-4-locus-active) or ThU (for IL-4-locus-inactive) conditions] and reanalyzed for purity. (D) FACS results for IL-4 and IFN-γ expression characteristics analyzed by staining reactivated sorted cells for intracellular IL-4 and IFN-γ. (E) Production of Th2 cytokines by cells restimulated by peptide Ag; shown are mean (± SEM) ELISA data (units/ml/100 cells), normalized to the number of KJ-26⁺ CD4 cells and compared to conventional DO11.10 Th2 cells. BLD, below limit of detection.

Fig. 2.

A memory CD4 population derived from a Th2 population purified for cells with an activated IL-4 locus. (A and B) A memory CD4 population forms from Th2 cells purified according to the activation of the IL-4 locus. (A) Kinetics of the memory population are shown for samples across the indicated time course; the graphs show mean (± SEM) frequencies of donor-derived cells. (B) Aggregate data from analyses of the indicated organs of recipient mice 6–11 weeks after transfer are shown in six experiments involving independent sorting of GFP⁺ and GFP⁻ CD4 T cells. Each square represents an independent sample; horizontal lines denote mean values. (C) GFP⁺ donor cells at the time of transfer are CD27^lo and Bcl-2^lo relative to GFP⁻ CD4 cells. GFP⁺ (Th2) and GFP⁻ (Th-null) CD4⁺ T effector cells were purified and expanded, and a portion of the population was analyzed before transfer into recipients. Shown are representative (two to three repeats) histograms of immunofluorescent staining for the indicated proteins gated on KJ1–26⁺ CD4⁺ cells.

Fig. 3.

Memory cells derived from purified IL-4 locus-on (GFP⁺) cells in Th2 cultures are committed to the expression of Th2 cytokine genes, but need prolonged recall activation for full productivity. Memory cells derived from previously activated populations (as in Fig. 2) were analyzed after recall stimulation ex vivo using peptide Ag. (A) Short-term (40-h) cytokine production. Ag-induced IL-4 and IFN-γ production was measured in culture supernatants by ELISA and divided by the number of donor (KJ1–26⁺ CD4⁺) cells in the culture. Shown are mean (± SEM) values (four independent experiments) for the indicated types of donor cell. (B) A schema of the experimental design to test commitment of the Th2 locus under non-Th2 recall conditions. (C) High-level Th2 cytokine production later after recall activation independent of the cytokine milieu. After recall activation with Ag, 6-day culture under the indicated conditions (Th1, Th-null, and Th2), and restimulation with Ag-loaded APCs, IL-4, IL-5, IL-10, and IL-13 produced per donor-derived cell were determined as for A. Shown are mean (± SEM) values (four independent experiments).

Fig. 4.

A facultative capacity to induce the IFN-γ gene after recall activation of memory cells derived from IL-4 locus-on Th2 effectors. (A) Shown are amounts of IFN-γ produced per donor-derived cell (determined as for Fig. 3) after 6-day culture under the indicated conditions (Th1, Th-null, and Th2) and restimulation with Ag (OVA323–339)-loaded APCs. (B) IL-4⁺, IFN-γ⁺ CD4 T cells can arise from Th2 cytokine locus-committed memory Th2 subset. After 6-day culture under the indicated conditions (Th-null, Th1, and Th2), donor-derived cells were restimulated (anti-CD3⁺ anti-CD28), processed for detection of intracellular IL-4 and IFN-γ, and analyzed by flow cytometry. Shown are cells in the KJ1–26⁺ CD4⁺ gate from one experiment representative of three repeats. (C–E) Memory cells derived from IL-4 locus-active Th2 effectors generate both IL-4 and IFN-γ producing memory effectors after recall activation in vivo. Purified GFP⁺ and GFP⁻ CD4 T cells were made as in Fig. 2 and transferred into naive recipients. Draining LN cells were analyzed 8–11 weeks after transfer, 6 days after immunization under Th1-biasing conditions. Shown are the ELISA data for cytokine production (as in Fig. 1; one representative experiment of three with comparable result) (C), GFP expression in donor cells upon harvest (D), and cytokine staining of cells in the KJ1–26⁺ CD4⁺ γατε (E). As in Fig. 3, values for IL-4 and IL-13 and IFN-γ, elicited from BALB/c controls by Ag, were all below limits of detection (<0.1 units/ml/100 cells), as were %IL-4⁺ and %IFN-γ⁺.