C57BL/6 mice received 10⁵ WT or receptor-deficient OT-I cells and were infected with VV-OVAp. Spleen cells were analyzed on days 5 and 7 post-infection for expression of IL-7 receptor α chain. Each histogram is a representative of those obtained for the three or four mice in each group.

C57BL/6, IL-12 p40-deficient, and IL-12 p35-deficient mice received 10⁵ WT or receptor deficient OT-I cells and were infected with VV-OVAp. Spleen cells were analyzed on day 30 and the numbers of OT-I cells determined. (A) Day 30 OT-I cell numbers in C57BL/6 (B6) and IL-12 p40-deficient hosts. Results shown are average and SEM for 3 mice/group. (B) Day 30 OT-I cell numbers in C57BL/6 (B6) and IL-12 p35-deficient hosts. Results shown are average and SEM for 4 mice/group. (C) CD62L expression on day 30 of WT and receptor-deficient cells from panel B. Too few OT-I.DKO cells were detected in C57BL/6 hosts to determine CD62L expression. Histograms shown are representative for each group.

Purified, naïve (CD44^lo) (A) OT-I.WT or (B) OT-I.DKO cells were placed in culture with aAPC bearing H-2K^b/OVA_257–264 and B7-1 on the surface, as previously described (8). Cultures received no cytokine, IL-12 (2U/ml), IFN-a (1500U/ml) or IL-21 (100ng/ml). Three days later cytolytic activity was determined in triplicate in a standard 4hr ⁵¹Cr-release assay using OVA_257-264-peptide pulsed EL-4 targets. Results are representative of three independent experiments.

C57BL/6 mice received 10⁵ cells of either WT or one of the receptor-deficient OT-I cells and one day later were infected with VV-OVAp (i.p.). Thirty days later, mice were challenged with 10⁵ LM-OVA (i.v.). Three days later cells from LNs, spleens and lungs were isolated, and LM was cultured from spleen as described in Methods. (A) LM was cultured from spleens and PFU per spleen determined. Values shown are averages and SD for each group (7 to 8 mice per group). (B) Numbers of OT-I cells in mice three days after challenge with LM-OVA. Dots indicate values for individual mice in each group, and the bars are the average. Based on Students T test, the number of WT cells was significantly greater than DKO cells (p < 0.001) or IFNARKO cells (p < 0.001), but did not differ significantly from IFNARKO cells (p = 0.68).

OT-I cells were stimulated in vitro for 3 days in the presence of IL-12 as described in Materials and Methods and adoptively transferred into C57BL/6 mice (3SigOT-I). A separate group of mice received naïve OT-I cells by adoptive transfer and were challenged with VV-OVAp. Day 0 is the day that in vitro stimulation was initiated. VV-OVAp challenge was done on day –1, as we had previously shown that initial OT-I response to VV-OVAp begins about 24hr after infection (24). CD62L expression was determined on days 8, 18 and 33. Representative histograms are shown on the left (including CD62L expression on endogenous CD8 T cells for reference), and the percent of CD62L^hi cells in the populations is shown on the right. Values are average and SD for 4 mice/group. In three additional independent experiments, both CD62L^hi and CD62L^lo cells were found in the 3SigOT-I cells as they transitioned to memory, although there was some variation in the fraction of cells that expressed a CD62L^hi phenotype.

(A) Naïve WT and receptor-deficient OT-I cells were adoptively transferred into C57BL/6 mice (10⁵/mouse) and the mice challenged one day later with VV-OVAp, and KLRG1 expression levels were determined on the indicated days. Representative histograms are shown, and the numbers indicate the percent of cells (average ± SD) expressing high KLRG1 levels, determined using gate shown on the bottom panels. Similar results were obtained in a separate experiment. (B) Naïve OT-I cells were stimulated for 3 days in vitro in the presence of IL-12 as described in the legend of Fig. 7, washed, and adoptively transferred into C57BL/6 mice. KLRG1 expression on the OT-I cells was determined on day 9, and KLRG1 expression on the endogenous CD8 T cells is shown for comparison. Representative histograms are shown (4 mice/group). Additional experiments examining KLRG1 expression on 3SigOT-I cells at the time of transfer and on days 9, 18 and 33 confirmed that KLRG1 expression remains low on these cells as they transition to memory.

(A) C57BL/6 mice received 5 × 10⁵ naïve (CD44^lo) cells of either WT or one of the receptor-deficient OT-I T cells. One day later, mice were infected with 5 × 10⁶ PFU (i.p.) of VV-OVAp. At the indicated times were harvested (3 or 4 mice per group) and the number of OT-I cells determined as described in Methods. Results shown are the average and SD of the total OT-I cells from spleens, LN and lungs of each group. The dotted line indicates the limit of detection of OT-I T cells in the samples. (B) Mice were adoptively transferred and infected as in A, and the numbers of OT-I cells determined on day 30 in LN, lungs and spleen. (C) OT-I cells on day 5 from spleens of mice primed as described in (A) were re-stimulated in vitro with OVA_257–264 peptide and IFN-g production was determined by as described in Methods. Representative histograms are shown, and numbers show the average and standard deviation (3-4 mice/group) of the percent of cells expressing high IFN-γ. The mean fluorescence intensities (MFI) for the IFN-γ^high cells were: WT, 396+/−28; IL12RKO, 360+/−15; IFNARKO, 297+/−25; DKO, 259+/−14. (D) C57BL/6 mice received 10⁵ naïve WT or OT-I.IL12RKO cells and one day later were not challenged (None) or were challenged s.c. with OVA_257–264 peptide and poly I:C as described in Methods. The numbers of WT and OT-I.IL12RKO cells were determined days 5 and 33, as indicated. Values shown are averages and SD for each group (3 mice per group). Similar results were obtained in two independent experiments.

(A) C57BL/6 mice received 10⁶ (open squares) or 5×10³ (closed circles) purified naive cells of both WT and one of the receptor-deficient OT-I cells by co-adoptive transfer. One day later, mice were infected with LM-OVA (i.p.). At day 7 after infection, spleens were removed from 13-16 mice of each group, and the number of OT-I cells of each type was determined by flow cytometery. The number of receptor-deficient OT-I cells is expressed as a percentage of co-transferred WT cells. Based on Students T test, the number of DKO cells was significantly reduced in comparison to IFNARKO cells (p = 0.026) and IL12RKO cells (p = 0.001), while IL12RKO did not differ significantly from IFNARKO (p = 0.09). (B) OT-I cells from day 7 spleens were re-stimulated in vitro with OVA_257–264 peptide and production of IFN-γ was determined by intracellular staining. Representative histograms are shown, and numbers show the average and standard deviation of the percent of cells expressing high IFN-γ (3 to 4 mice/group). The mean fluorescence intensities (MFI) for the IFN-γ^high cells were: WT, 266+/−36; IL12RKO, 231+/−13; IFNARKO, 283+/−9; DKO, 206+/−18. (C) C57BL/6 mice received 10⁵ WT or one of the receptor-deficient OT1 cells, and were infected one day later with 10⁴ LM-OVA (i.p.). At day 30 after infection cells from LN, spleen and lung were isolated. Each bar represents the average and SD of total OT-I cells (LNs, spleen and lungs) from 3 or 4 mice/group. (D) Distribution of memory OT-I CD8 T cells in tissues from mice in panel B. Each bar represents the average and SD for 3 or 4 mice/group.

Naïve (CD44^lo) OT-I T cells were stimulated for 3 days in microtiter wells coated with H-2K^b/OVA_257-264 , B7-1 and IL-2 in the absence (2sigOT-I) or presence of IL-12 (3SigOT-I). On day 3, cells were harvested, washed and adoptively transferred into C57BL/6 mice (10⁶/mouse). At the same time, one group received naïve OT-I T cells by adoptive transfer (Naïve). (A) OT-I cell numbers in spleens were determined on the indicated days post-transfer. Results shown are average and SD for 4 mice/group. Solid and dashed lines show results obtained from two independent experiments. Additional experiments gave similar results for numbers of OT-I cells on days 4, 6 and 30. The horizontal dashed lines indicate the limit of detection of OT-I cells in the samples. (B) Mice were adoptively transferred with in vitro generated 2SigOT-I or 3SigOT-I cells. As a control, one group received naïve OT-I cells by adoptive transfer and were primed on day 0 by infection with VV-OVAp (VV-OVAp prime). Thirty days later, the numbers of OT-I cells in the spleens were determined (gray bars). Groups of mice were also challenged with LM-OVA (10⁵ i.v.) on day 30, and the numbers of OT-I cells in the spleens determined 3 days later (black bars). Results shown are averages and SD for groups of 4 mice each. (C) LM was cultured from spleens and PFU per spleen determined for the groups of mice described in (B). One group of mice received naïve OT-I cells by adoptive transfer on day 0 and was left unchallenged until the day 30 infection with LM (Naïve OT-I). Values shown are averages and SD for each group (4 mice per group). Protection by 3SigOT-I cells but not 2SigOT-I cells was confirmed in two additional independent experiments.