OT-I–GFP TCR is expressed on the cell surface and is functional. (a) Flow cytometric analysis of GFP fluorescence in OT-I–GFP LN cell subsets expressing the indicated surface markers. (b) Flow cytometric analysis of GFP fluorescence and TCR surface expression in OT-I–GFP Rag2^−/− LN cells. Anti-Vα2 and anti-Vβ5 stain the OT-I–GFP α and β chains, respectively. Percentages are shown in each quadrant. (c and d) Naive WT OT-I or OT-I–GFP T cells were surface labeled with anti–TCR-β antibody, placed on a lipid bilayer presenting pMHC and ICAM-1, and imaged by TIRF microscopy. (c) Dynamics of microcluster and cSMAC formation. See Video 1 for full time lapse. Data are representative of two experiments. Bar, 10 µm. (d) Quantification of fluorescent intensity of surface labeling of TCR-β (H57) and GFP of OT-I–GFP TCR microclusters. (e) Proliferation of WT OT-I CD45.1 T cells versus OT-I–GFP Rag2^−/− T cells. CFSE dilution of cotransferred T cells 66 h after SL8 peptide + CFA immunization. (a, b, and e) Data are representative of at least three experiments. (f) TCR down-regulation in naive T cells after stimulation with antigen-pulsed DCs. Error bars represent SEM. (d and f) Data were pooled from two experiments.

TCR internalization does not require cSMAC formation. Naive or activated OT-I–GFP T cells were incubated with DCs pulsed with SL8 peptide and imaged by epifluorescence microscopy. Data were pooled from two experiments. For naive T cells, n > 60 cells per condition. For activated T cells, n > 30 cells per condition. P-values were determined using Fisher’s exact test. (a) TCR dynamics. The percentages of T cells in contact with DCs (contact), with a cSMAC, or with internalized TCR are indicated. Only T cells that contacted a DC during the time lapse are included. (b) Duration of cSMAC persistence. Only T cells that formed a cSMAC are included. Each dot represents one cell, and the bars represent the median. (c and d) Naive T cell–DC interactions. Contrast (top) and maximum intensity z projection of OT-I–GFP fluorescence (bottom) imaged by epifluorescence microscopy are shown. The dotted line indicates the DC border, and the arrowheads indicate internalized TCR vesicles. Bars, 10 µm. (c) The T cell contacts the antigen-bearing DC (1,000 ng/ml SL8), stops crawling and adopts a rounded morphology, forms a cSMAC, and then internalizes vesicles of TCR through the cSMAC. See Video 2 for full time lapse. (d) The T cell contacts the antigen-bearing DC (10 ng/ml SL8), forms microclusters at the T cell–DC interface, and then rapidly internalizes the TCR without forming a cSMAC. See Video 3 for full time lapse. (e) Frequency of TCR behaviors. (f) Frequency of cSMAC-independent TCR internalization. Only cells that internalized their TCR are included; those that did not form a cSMAC are designated cSMAC independent.

Quantification of OT-I–GFP T cell surface TCR fluorescence in vivo. (a) TCR expression on the surface of OT-I–GFP Rag2^−/− T cells versus WT T cells with endogenously rearranged Vα2⁺ or Vβ5⁺ TCRs. Data are representative of more than three experiments. (b–d) A localized active contour algorithm was used to identify the cell edges of OT-I–GFP Rag2^−/− cells in LNs from two-photon images. GFP images were masked in the cell surface regions. n = 27 cells from three experiments. (b) Representative cell surface identified by the local contour segmentation. Z-series images of an OT-I–GFP Rag2^−/− cell (top row) with a mask marking the surface-exposed pixels of the cell (green in bottom row). The interplane distance is 1 µm. Bar, 4 µm. (c) Surface volume rendering of an OT-I–GFP cell segmented using the active contour algorithm. (d) Quantification of green fluorescence of OT-I–GFP cells masked on the cell surface voxels.

OT-I–GFP fluorescence can be differentiated from tissue autofluorescence. (a, b, and d–f) CMTMR (red)-labeled OT-I or OT-I–GFP Rag2^−/− cells (green) were transferred into WT recipients, and LNs were visualized by two-photon microscopy. Data are representative of more than three experiments. (a) Green fluorescence in a single z plane. Arrows indicate the location of OT-I–GFP Rag2^−/− T cells, as determined by CMTMR fluorescence (not depicted). (b) Pixel intensities in the green channel from the image represented in panel a. (c) Flow cytometric analysis of green fluorescence and autofluorescence in WT LN containing OT-I–GFP Rag2^−/− cells. (d) Analysis of green fluorescence contributed by CMTMR bleed-through when imaged using 525/50- versus 510/20-nm filters. Background measurements were taken in regions that did not contain tissue. Fluorescence values in the red and green channels were measured in regions that were drawn around individual OT-I or OT-I–GFP Rag2^−/− cells based on CMTMR. Lines represent linear equations fit to the data. See Materials and methods for the equations. (e) Representative images of maximum intensity z projections showing unprocessed and processed images. (f) Representative maximum intensity z projection at various stages of image processing. Image was acquired with the 510/20-nm filter. See Video 4 for three-dimensional reconstruction of image processing. The region shown represents a tissue volume of 150 (x) × 150 (y) × 80 µm (z). Bars, 25 µm.

Motility of T cells during antigen recognition. CMTMR-labeled OT-I–GFP Rag2^−/− cells and CFP⁺ polyclonal CD8 T cells were cotransferred into CD11c-YFP recipients, and recipients were immunized with CFA in the footpads. 18 h after T cell transfer and immunization, popliteal LNs were explanted and visualized by two-photon microscopy before and after the addition of SL8 peptide antigen. Imaging was performed in the most superficial 80 µm of the LN to replicate imaging conditions during TCR visualization. Data were pooled from four experiments. See Video 5 for representative video. For all data with antigen, time = 0 represents the time at which the antigen was added. (a) Representative tracks displaying cell paths during 10 min of imaging. (b) Mean T cell crawling velocity. (c) Duration of T cell arrest. (d) Duration of T cell–DC contact. (b–d) Statistical analysis was performed using a one-way analysis of variance with Tukey’s post test. Each dot represents one cell, and the bars represent the mean.

Rapid TCR clustering and internalization during the initial stages of antigen recognition. CMTMR-labeled OT-I–GFP Rag2^−/− cells were transferred into WT recipients and immunized as in Fig. 5. LNs were explanted and visualized by two-photon microscopy. Data are representative of more than five experiments. For all data with antigen, time = 0 represents the time at which the antigen was added. (a, b, and d) Maximum intensity z projections of OT-I–GFP fluorescence represented in pseudocolor (top) and overlaid OT-I–GFP fluorescence in green and CMTMR fluorescence in red (bottom). (a) Representative time-lapse image of OT-I–GFP TCR localization on a crawling T cell in a nondraining LN in the absence of antigen. White arrowheads indicate the leading edge of the cell of interest. Image was collected using the 510/20-nm filter. (b) Representative time lapse of sustained OT-I–GFP TCR clustering after antigen addition. Image was collected using the 525/50-nm filter. For full video, see Video 6. (c) TCR clustering associated with the cell surface. Orthogonal views of OT-I–GFP fluorescence of the T cell represented in b showing a 1.6 (x) × 1.6 (y) × 2–µm (z) slice through the cell at the location indicated by the crosshairs. (d) Representative time lapse of OT-I–GFP TCR internalization after antigen addition. Image was collected using the 525/50-nm filter. For full video, see Video 7. (e) TCR internalization. Orthogonal views of OT-I–GFP fluorescence of the T cell represented in d showing a 1.6 (x) × 1.6 (y) × 2–µm (z) slice through the cell at the location indicated by the crosshairs. Bars, 10 µm.

Transient antigen-independent TCR clustering and rapid antigen-dependent TCR clustering and internalization in vivo. CMTMR-labeled OT-I–GFP Rag2^−/− cells were transferred into WT recipients and immunized, and LNs were imaged by two-photon microscopy as in Figs. 5 and 6. Data analyzed were combined from images collected using the 510/20- and 525/50-nm filters. For all data with antigen, time = 0 represents the time at which the antigen was added. (a) Bulk dynamics of T cell motility arrest, TCR clustering, and TCR internalization in the presence and absence of antigen. For no antigen, n = 225 combined from four experiments. For + antigen, n = 82 combined from three experiments. (b) Dynamics of TCR clustering and TCR internalization in individual cells in the presence and absence of antigen. Each line corresponds to a single randomly selected representative cell. (c) Frequency of T cell motility arrest, TCR clustering, and TCR internalization. For no antigen, n = 225 combined from four experiments. For + antigen, n = 82 combined from three experiments. Statistical analysis was performed using Fisher’s exact test. (d) Duration of TCR clustering. Only cells with clustered TCR are represented. Each dot represents one cell, and the bars represent the median. For no antigen, n = 21 combined from four experiments. For + antigen, n = 30 combined from three experiments. Statistical analysis was performed using a two-tailed t test with Welch’s correction.

TCR signaling does not require motility arrest. (a–c) CMTMR-labeled OT-I–GFP Rag2^−/− cells were transferred into WT or CD11c-YFP recipients and immunized, and LNs were imaged by two-photon microscopy as in Figs. 5 and 6. (a) Contingency of TCR clustering and internalization on motility arrest in vivo. Only cells that clustered or internalized their TCR are represented. For clustering no antigen (Ag), n = 21 combined from four experiments. For clustering + antigen, n = 30. For internalization + antigen, n = 13 combined from three experiments. (b and c) Single z-plane time lapse images of in vivo T cell–DC interactions in the presence of antigen. See Video 8 for the full time lapse. Data are representative of four experiments. (b) Motile in vivo T cell–DC interaction. (c) Arrested in vivo T cell–DC interaction. (d) Representative in vitro time lapse of a naive OT-I–GFP T cell internalizing TCR while crawling along an antigen-bearing DC. The T cell internalizes its TCR through microclusters as it maintains an active leading edge and continually crawls along a DC. Contrast (top) and maximum intensity z projection of OT-I–GFP fluorescence (bottom) are shown. Arrowheads point to the T cell of interest. The white lines define the xy space rendered in the xz rendering below. See Video 9 for the full time lapse. (e–h) Naive OT-I T cells were labeled with Fura-2AM and placed on lipid bilayers presenting pMHC + ICAM-1. Data are representative of two experiments. (e and f) Contrast (grayscale background image) and calcium imaging (pseudocolor overlay) visualized by epifluorescence microscopy. Low–high calcium is represented by a blue–white pseudocolor scale. See Video 10 for the full time lapses. (e) Representative time lapse of calcium signaling while crawling along a lipid bilayer with oscillations in calcium levels and motility. (f) Representative time lapse of a T cell fluxing calcium while arrested. (b, c, e, and f) White lines show cell tracks. (g and h) Quantification of motility with relation to calcium signaling. At each time point, the instantaneous velocity of the cell was correlated with its calcium level. Time points with a Fura-2AM 340-nm/380-nm emission ratio of >1.4× baseline were scored as having elevated calcium. Time points with an instantaneous velocity <2 µm/min were scored as arrested. The dashed line indicates the cutoff value below which cells were termed nonmotile (2 µm/min). n = 1,970 time points from 40 cells pooled from two experiments. Bars: (b and d–f) 10 µm; (c) 5 µm.