We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

## Results: 8

1.

**Cognate peptide interactions increase CTL division in the meninges.**P14 CTL motility and division were examined after administration of cognate (GP

_{33-41}) or irrelevant (OVA

_{257-264}) peptide into the subarachnoid space of P14 seeded mice at day 6 after infection. Mean track velocities (A) and motility coefficients (B) were calculated for GFP

^{+}P14 CTL imaged in the meninges within the first hour after peptide administration. (*, P < 0.001, Mann-Whitney rank sum test). Each dot represents an individual P14 CTL quantified from mice receiving GP33 (

*n*= 1,888 tracks,

*n*= 11 mice) or OVA (

*n*= 1,016 tracks,

*n*= 8 mice). Data are compiled from six to eight independent experiments. Red line denotes the mean ± SEM for each group. The normalized frequency (C) and absolute number (D) of BrdU

^{+}Thy1.1

^{+}P14 CTL were calculated 90 min after administration of BrdU and peptide into the subarachnoid space. (*, P < 0.05, Student’s

*t*test). Data are plotted as mean ± SEM and are representative of 15–16 mice compiled over four independent experiments.

2.

**Virus-specific CTLs divide in the meninges of the CNS.**To examine whether virus-specific CTLs divided locally within the CNS, brains from mice seeded with GFP

^{+}P14 cells and then infected i.c. with 10

^{3}PFU LCMV Arm were harvested on day 6 after infection. Frozen sections were cut and stained with Ki-67 (red) to detect dividing CNS-infiltrating GFP

^{+}P14 cells (green). Shown are representative panels to reflect anatomical location and division events. The top two rows show lower magnification images and the dashed lines represent the border between the meninges and brain parenchyma. Bar, 30 µm. The asterisks denote the parenchymal portion of the tissue. The white boxes in the second row highlight a mitotic CTL that is enlarged in the third row. Both the third and fourth rows show representative examples of individual virus-specific P14 cells undergoing mitosis within the meningeal space. Bar, 5 µm. Cell nuclei were counterstained with DAPI (blue). Data are representative of two independent experiments (

*n*= 8 mice).

3.

**Virus-specific CTLs divide in lymphoid and nonlymphoid sites.**To determine the cell cycle stage of virus-specific CTL, blood (BL), CNS, liver (LIV), cervical LNs, and spleen (SPL) of mice seeded with Thy1.1

^{+}P14 cells were harvested 15 min after i.p. BrdU administration on day 6 after LCMV infection. (A) Representative flow cytometric plots of BrdU versus 7AAD are shown for Thy1.1

^{+}P14 cells extracted from tissues of infected mice. (B–E) The ratio of Thy1.1

^{+}P14 cells found in G0-G1, S, or G2-M stages of cell cycle in the CNS (B), liver (C), LN (D), or spleen (E) relative to the blood was calculated for individual mice (

*n*= 16–24 mice per group). Data were compiled from six independent experiments. The red line is set at a ratio of 1 and indicates no difference relative to the blood. The ratio of cells in S or G2-M compared with G0-G1 was significantly different in all tissues examined (P < 0.05, one way ANOVA).

4.

**Virus-specific CTLs in cell cycle are activated effector cells.**To determine whether the CD8

^{+}T cells undergoing cell division in the CNS were naive or activated effectors, the CNS, liver (LIV), blood (BL), cervical LNs, and spleen (SPL) of mice seeded with Thy1.1

^{+}P14 cells were harvested 15 min after i.p. BrdU administration on day 6 after LCMV infection. (A) Representative flow cytometric plots of CD44 versus CD62L for Thy1.1

^{+}P14 cells. Note the absence of naive P14 cells (identified as CD62L

^{+}CD44

^{−}) in the tissues of infected animals. (B) Quantification of CD44 expression on P14 cells at all stages of cell cycle. Data are plotted as mean ± SD and are representative of

*n*= 14 mice per group collected over four independent experiments. (C) Representative flow cytometric plots of granzyme B or isotype control staining versus CD8 for Thy1.1

^{+}P14 cells. (D) Granzyme B expression was quantified in P14 cells at all stages of cell cycle. Data are plotted as mean ± SD and are representative of

*n*= 19 mice per group acquired from five independent experiments.

5.

**Anatomy and dynamics of virus-specific CTL division in the CNS.**(A) A representative maximal projection of a 3D z-stack is shown for a mouse at day 6 after infection. The projection represents one time point in a 51-min film. The white arrows denote a GFP

^{+}P14 CTL undergoing mitosis (See Video 2). The strip beneath the maximal projection shows a view through z-stack at the position of the dividing cell (arrow). Note that the dividing cell is positioned in the meninges above a blood vessel (red) and beneath the skull bone (blue). (B) A cropped area is shown depicting the dividing CTL in A. Different 3D projections were generated by rotating the z-stack around the x axis. The dividing CTL is always denoted with an arrow. Bars, 20 µm. (C) A representative time lapse is shown for a P14 CTL (green) dividing in the meninges (arrow; see Video 2). The magenta track shows the position of the dividing CTL at each point in the time lapse. Note that the CTL is stationary when compared with three motile nondividing CTLs (blue tracks). (D) The mean track velocity and motility coefficient were calculated for dividing (red) and nondividing (blue) P14 CTLs in the meninges. Means are denoted by horizontal bars. (E) Instantaneous track velocities are plotted versus time for three representative dividing (top graph) and nondividing (bottom graph) P14 CTLs. Brackets in the top graph denote velocities in three phases of CTL division: arrest, mitosis, and the split into two daughter cells. Data are representative of five independent experiments (

*n*= 13 mice).

6.

**CTLs form stable and dynamic contacts with DCs during LCMV meningitis.**(A–E) Representative time lapses of 3D reconstructions show the position of dividing CFP

^{+}P14 CTL (green) in relation to CD11c-YFP

^{+}DCs (red) in the meninges at day 6 after infection (see Video 3). Panel A shows CTL swarming around and contacting a large YFP

^{+}meningeal DC. One P14 CTL (circle) divides after contact with this DC. An enlarged view of the dividing cell, as well as the CTL-DC contact point (arrow), is shown in B. C and D show two additional examples of CTL mitosis after DC contact (arrows), whereas E illustrates a P14 CTL that divides when not in contact with a DC. Bars, 10 µm. (F) Contact durations (in minutes) between P14 CTL and meningeal DCs were calculated from 30-min time lapses. Contacts were divided into two categories with respect to YFP

^{+}DCs: mobile contact (>2 µm/min) and stable contact (≤2 µm/min). Each dot represents an individual P14 CTL. Red dots denote P14 CTL that divided while contacting YFP

^{+}DCs. The mean ± SD contact durations were 11.0 ± 9.3 min (all), 6.3 ± 5.9 min (mobile), and 17.9 ± 9.3 min (stable). Asterisk denotes a statistically significant (P < 0.00001, Mann-Whitney rank sum test) difference between mobile and stabile contacts. Means are denoted by horizontal bars. (G) The frequency of mobile and stable P14 CTL contacts represented as a histogram (bin size = 2 min). Data are compiled from

*n*= 4 mice (nine different fields).

7.

**MHC I but not CD80/CD86 expression influences the ability of virus-specific CTL to enter cell cycle in the CNS.**(A and B) 3.0 × 10

^{7}CD8

^{+}T cells isolated from the spleens of day-6 mice seeded with Thy1.1

^{+}P14 cells were transferred i.v. into day-6 LCMV-infected WT or H-2D

^{b−/−}recipients. Approximately 2–3 × 10

^{6}of the transferred cells were CD8

^{+}Thy1.1

^{+}P14 cells. 90 min after adoptive transfer, recipient mice received 2 mg BrdU i.p. and were harvested 15 min later. (A) Representative dot plots of BrdU versus 7AAD staining shown for CD8

^{+}Thy1.1

^{+}P14 cells that infiltrated the CNS of recipient mice. (B) The percentage of CD8

^{+}Thy1.1

^{+}P14 cells in G0-G1, S, and G2-M is shown for WT or H2D

^{b−/−}mice. Data are shown as mean ± SD for a single experiment and are representative of

*n*= 14 mice per group collected over four independent experiments. The asterisk denotes statistical significance (P ≤ 0.006, Student’s

*t*test). (C) To block CD80 and CD86, WT B6 mice seeded with Thy1.1

^{+}P14 cells were injected i.p. on days 4 and 5 after LCMV infection with 200 µg CTLA-4-Fc (red) or PBS (black). The percentage of CNS Thy1.1

^{+}P14 cells (mean ± SD for a single experiment) in the denoted stages of cell cycle was analyzed at day 6 after infection. Data are representative of

*n*= 22–24 mice per group acquired over six independent experiments.

8.

**Peptide–MHC I interactions influence the division program of virus-specific CTL.**(A) Representative flow cytometric plots depict the non–T cell component of CNS mononuclear cells in mock versus day-6 LCMV-infected mice. Plots are gated on Thy1.2-negative cells. The schematic plot on the far left shows the relative position of 1, microglia; 2, neutrophils; and 3, macrophages/monocytes/DCs. To further delineate macrophages/monocytes and DCs, the cells in region 3 were subdivided into CD11c

^{−}CD45.2

^{hi}CD11b

^{+}(macrophages/monocytes) and CD11c

^{+}CD45.2

^{hi}CD11b

^{+}(DCs) as shown in the far right plot. Data are representative of

*n*= 16–24 mice per group accumulated over four to six experiments. (B) Representative histograms show the expression of MHC class I (K

^{b}D

^{b}), MHC class II (I-A/I-E), CD80, and CD86 (red) relative to isotype controls (blue) on microglia (Thy1.2

^{−}CD45.2

^{lo}CD11b

^{+}), macrophages/monocytes (Thy1.2

^{−}CD45.2

^{hi}CD11b

^{+}CD11c

^{−}), and DCs (Thy1.2

^{−}CD45.2

^{hi}CD11b

^{+}CD11c

^{+}) at day 6 after infection. Data are representative of

*n*= 16 mice collected in three separate experiments. (C) Sorted effector virus-specific CTL pooled from four mice were stimulated at a 2:1 ratio with sorted GP33 peptide-pulsed CNS APCs isolated from day 6–infected animals (

*n*= 20). BrdU was added to the in vitro culture for the last 30 min of the 90-min incubation. Representative BrdU versus 7AAD staining of CD8

^{+}Thy1.1

^{+}P14 cells incubated alone or with CNS APCs is shown. These data are representative of four separate experiments. (D) Normalized percentages of T cells in various stages of cell cycle are shown. Data were normalized from four independent experiments. (*, P < 0.05, one-way ANOVA). (E) Sorted effector Thy1.1

^{+}P14 cells pooled from two to four mice were stimulated at a 2:1 ratio with splenic DCs isolated from day 6–infected animals (

*n*= 2–3) in the presence of OVA or GP33 peptide. The BrdU incubation is the same as in C. Normalized percentages of T cells in various stages of cell cycle are shown (*, P < 0.05, one way ANOVA). These results are compiled from three independent experiments. Data in D and E are plotted as mean ± SD.