Signal strength regulates antigen-mediated T-cell deceleration by distinct mechanisms to promote local exploration or arrest

Abstract

T lymphocytes are highly motile cells that decelerate upon antigen recognition. These cells can either completely stop or maintain a low level of motility, forming contacts referred to as synapses or kinapses, respectively. Whether similar or distinct molecular mechanisms regulate T-cell deceleration during synapses or kinapses is unclear. Here, we used microfabricated channels and intravital imaging to observe and manipulate T-cell kinapses and synapses. We report that high-affinity antigen induced a pronounced deceleration selectively dependent on Ca(2+) signals and actin-related protein 2/3 complex (Arp2/3) activity. In contrast, low-affinity antigens induced a switch of migration mode that promotes T-cell exploratory behavior, characterized by partial deceleration and frequent direction changes. This switch depended on T-cell receptor binding but was largely independent of downstream signaling. We propose that distinct mechanisms of T-cell deceleration can be triggered during antigenic recognition to favor local exploration and signal integration upon suboptimal stimulus and complete arrest on the best antigen-presenting cells.

Keywords: T cell; kinapse; migration; synapse.

Fig. 1.

Frequent direction changes and increased exploration during kinapses in vivo. Mice were transferred with GFP⁺ OT-I CD8⁺ T cells, were injected with the high-affinity (N4) or the low-affinity (Q4) peptide, and were subjected to intravital imaging of the popliteal lymph node. (A) Recognition of low-affinity peptide is associated with efficient exploration of T cell’s vicinity. T-cell images are projected over a 5-min trajectory. The maximal scanned radius was derived from the largest inscribed circle (white plain circle) in the projected track. (Scale bar: 20 μm.) (B) Turning angles measured in T-cell trajectories between two consecutive images (30 s apart). (C) Maximal scanned radius (μm) calculated for individual T cells. Data are representative of at least three independent experiments. ***P < 0.001.

Fig. S1.

T-cell dynamics in the lymph node are regulated by TCR ligand affinity. Mice were transferred with GFP⁺ OT-I CD8⁺ T cells, were injected with the high-affinity (N4) or the low-affinity (Q4) peptide, and were subjected to intravital imaging of the popliteal lymph node. (A) Representative two-photon images and overlaid T-cell trajectories (corresponding to 5 min of imaging) in uninjected mice or mice receiving the Q4 or N4 peptide. (B–D) Average cell speed (μm/min) (B), arrest coefficient (percentage of time during which a cell exhibited an instantaneous speed <2 μm/min) (C), and straightness (D) are graphed. Data are representative of at least three independent experiments. ***P < 0.001.

Fig. 2.

Frequent direction changes and increased exploration during kinapses in microchannels. The migration of preactivated GFP⁺ OT-I CD8⁺ T cells was analyzed in 6-µm-wide microchannels coated with either K^b-TRP2 (control peptide), K^b-Q4, or K^b-N4 antigenic complexes. (A) Sequential images acquired every 35 s are shown for the indicated pMHC. (Scale bar: 100 μm.) (B) Examples of exploration in microchannels. T-cell images were subjected to thresholding and then projected over a 30-min trajectory. The duration of exploration is calculated at each pixel along the microchannel axis from the projected image. The maximal scanned length is defined as the longest continuous zone explored for more than 2 min (dotted line). (C) Average maximal scanned length calculated for T cells in each of the indicated conditions. Data are representative of five independent experiments. ***P < 0.001; **P < 0.005.

Fig. S2.

Frequent direction changes and increased exploration during kinapses in microchannels. The migration of preactivated GFP⁺ OT-I CD8⁺ T cells was analyzed in 6-µm-wide microchannels coated with either K^b-TRP2 (control peptide), K^b-Q4, or K^b-N4 antigenic complexes. (A and B) The average arrest coefficient (A) and number of reversals per minute (B) are shown for the indicated conditions. Data are representative of five independent experiments. ***P < 0.001.

Fig. 3.

Differential role of calcium influx during synapse versus kinapse formation. The migration of GFP⁺ OT-I CD8⁺ T cells was analyzed in pMHC-coated microchannels in the absence (−) or in the presence (+) of EGTA (extracellular Ca²⁺ chelator). (A) Sequential images acquired every 5 min are shown for the different conditions. (Scale bar: 50 μm.) (B and C) Arrest coefficients (B) and reversals per minute (C) are shown. Data are representative of five independent experiments. ***P < 0.001; *P < 0.05; ns, nonsignificant (P > 0.05).

Fig. S3.

T-cell deceleration does not require calcium signaling. Migration of GFP⁺ OT-I CD8⁺ T cells in pMHC-coated microchannels in the presence of BAPTA-AM (intracellular Ca²⁺ chelator) and EGTA (extracellular Ca²⁺ chelator) (+) or DMSO (−) as a control. (A and B) Arrest coefficients (A) and reversals per minute (B) are shown. Data are representative of three independent experiments. ns, nonsignificant (P > 0.05). (C and D) Only high-affinity pMHC induces a strong calcium influx, which is abolished in the presence of EGTA. OT-I CD8⁺ T cells were stained with a calcium indicator (Fluo-3) and stimulated with the indicated coated pMHC, in the absence or in the presence of EGTA to chelate extracellular calcium. (C) Representative images. (D) Average Fluo-3 signal detected in OT-I T cells in the different conditions of stimulation. Maximal response was estimated by stimulating the T cells with ionomycin. Only significant differences are noted. ***P < 0.001; ns, nonsignificant (P > 0.05).

Fig. 4.

Arp2/3 activity is required for T-cell arrest during synapses. The migration of GFP⁺ OT-I CD8⁺ T cells in pMHC-coated microchannels was analyzed in the presence of CK666 (inhibitor of Arp2/3) (+) or DMSO (−) as a control. (A) Sequential images acquired every 3 min are shown for the different conditions. (Scale bar: 50 μm.) (B and C) Arrest coefficients (B) and reversals per minute (C) are shown. Data are representative of five independent experiments. *P < 0.05; ns, nonsignificant (P > 0.05).

Fig. 5.

Switch of migration mode during kinapse formation. (A and B) Recognition of low-affinity pMHC complexes results in T-cell migration in the absence of a marked uropod. The migration of preactivated GFP⁺ OT-I CD8⁺ T-cell was analyzed in pMHC-coated microchannels. (A) Sequential images acquired every 35 s are shown for the indicated pMHC. (Scale bar: 20 μm.) For each time point, cells were scored for the presence of a uropod (white arrow). (B) The presence of a uropod was analyzed for individual T cells. Each row represents one cell followed over time, and one square represents one time point. Filled squares correspond to time points in which a uropod was evident. The average percentage of time during which a uropod is detected is shown (Right). Data are representative of four independent experiments. (C–E) T-cell migration during recognition of low-affinity TCR ligands results in the relocation of the MTOC. OT-I CD8⁺ T cells were retrovirally transduced to express GFP-centrin and labeled with Hoechst. T-cell migration was analyzed in pMHC-coated microchannels. (C) Sequential images acquired every 35 s are shown for the indicated pMHC. Green indicates GFP-centrin; red indicates Hoechst. (Scale bar: 20 μm.) (D) Sequential images of MTOC and nucleus tracking. Green indicates MTOC; red indicates nucleus. (Scale bar: 20 μm.) The average distance between the MTOC and the nucleus is graphed (Right). (E) Representative trajectories of MTOC in a microchannel (gray and green lines). The black lines represent the microchannel sides, and the dotted line represents the microchannel axis. The average distance between the MTOC and the side of the microchannel is graphed (Right). Data are representative of four independent experiments. (F) T-cell migration during recognition of low-affinity TCR ligands results in the repolarization of LAT toward the front of the cell. (Left) Representative images of LAT-GFP–expressing OT-I T cells migrating in K^b-TRP2– or K^b-Q4–coated microchannels. (Scale bar: 10 μm.) Line scans of GFP intensity along the central axis of the microchannels are shown for different time points. (Right) Graph shows the average LAT polarity (−1: LAT at the rear of the cell; +1: LAT at the front of the cell). Data are representative of three independent experiments. ***P < 0.001; **P < 0.005.

Fig. 6.

Kinapse formation requires TCR binding to cognate pMHC but appears largely independent of downstream signaling. The migration of GFP⁺ OT-I CD8⁺ T cells in pMHC-coated microchannels was analyzed in the presence of PP2 (Src-family kinases inhibitor) (+) or DMSO (−) as a control. (A) Sequential images acquired every 3 min are shown for the different conditions. (Scale bar: 40 μm.) (B and C) Average speed (B) and arrest coefficients (C) are shown. Data are representative of three independent experiments. ***P < 0.001; ns, nonsignificant (P > 0.05).

Fig. S4.

PP2 effectively inhibits T-cell activation induced by K^b-Q4 complexes. OT-I T cells recovered from the K^b-Q4–coated microchannel entrance shows down-regulation of CD62L and up-regulation of CD69 and CD25 compared with the K^b-TRP2 control. This activation signature was abolished by the addition of PP2.

Fig. S5.

Recognition of low- and high-affinity TCR ligands trigger LAT accumulations. (A) Representative images of LAT-GFP–expressing OT-I T cells migrating in K^b-TRP2–, K^b-Q4–, or K^b-N4–coated microchannels. LAT spots (bright punctuated GFP accumulations) can be observed on K^b-Q4 and K^b-N4 images. (Scale bar: 10 μm.) (B) Graphs show the average percentage of time during which LAT-punctuated accumulations were detected and the average persistence of the spots. Data are representative of three independent experiments. ***P < 0.001; *P < 0.05; ns, nonsignificant.

Fig. S6.

The hallmarks of kinapse motility require TCR binding to pMHC but are largely independent of downstream TCR signaling. The migration of GFP⁺ OT-I CD8⁺ T cells in pMHC-coated microchannels was analyzed in the presence of PP2 (Src-family kinases inhibitor) (+) or DMSO (−) as a control. (A and B) The frequency of reversals (A) and the average percentage of time during which a uropod is detected (B) are shown. Data are representative of three independent experiments. ns, nonsignificant (P > 0.05).

Fig. S7.

Model of kinapse and synapse formation triggered by low- and high-affinity TCR ligands.