Figure 1.
Illustration of cell migration experiments using microfluidic devices and data analysis methods.
Figure 2.
T cell migration in a gradient or a uniform field of CCL21.
(A) Angular histograms show T cells orient randomly in a 100 nM uniform CCL21 field, but toward a 100 nM CCL21 gradient (B) Comparison of chemotactic index (C.I.) and speed of cells in a 100 nM uniform CCL21 field or a 100 nM CCL21 gradient show random migration in the 100 nM uniform CCL21 field, but chemotaxis in the 100 nM CCL21 gradient with similar speed. The error bars represent the standard error of the mean (s.e.m.). The p values for each comparison from 2-sample t test are shown. Positive C.I. indicates cells migrate toward the gradients.
Figure 3.
T cell migration in CCL19 gradients.
(A) Angular histograms show T cells orient randomly in a 5 nM CCL19 gradient, but toward a 100 nM CCL19 gradient (B) Comparison of chemotactic index (C.I.) and speed of cells in a 5 nM CCL19 gradient or a 100 nM CCL19 gradient show random migration in the 5 nM CCL19 gradient, but chemotaxis in the 100 nM CCL19 gradient with higher speed in the 100 nM CCL19 gradient. The error bars represent the standard error of the mean (s.e.m.) The p values for each comparison from 2-sample t test are shown. Positive C.I. indicates cells migrate toward the gradients.
Figure 4.
T cell migration in “double-uniform” CCL19 and CCL21 fields.
(A) Angular histogram shows random orientation of T cells in superimposed 5 nM CCL19 and 100 nM CCL21 uniform fields. Chemotactic index (C.I.) and the speed of cells are shown with the errors represented as the standard error of the mean (s.e.m.) Positive C.I. indicates cells migrate toward the gradients. (B) Selected cell tracks from a representative experiment showing cells migrate randomly.
Figure 5.
T cell migration in a CCL19 gradient with a uniform background of CCL21.
(A) Angular histogram shows more T cells orient against a 5 nM CCL19 gradient with a uniform background of 100 nM CCL21. (B) Angular histogram shows T cells orient randomly in a 5 nM CCL19 gradient with a super-physiological 250 nM uniform CCL21 field. (C) Comparison of chemotactic index (C.I.) and the speed of cells between 5 nM CCL19 gradient with a uniform background of 100 nM CCL21 and 5 nM CCL19 gradient with a super-physiological 250 nM uniform CCL21 field. The error bars represent the standard error of the mean (s.e.m.). Positive C.I. indicates cells migrate toward the gradients. (D) Selected cell tracks from a representative experiment showing more cells migrate away from the 5 nM CCL19 gradient in a uniform background of 100 nM CCL21.
Figure 6.
Model predictions of cell orientation and migration in a single L2 gradient or a “double-uniform” L1 and L2 fields.
Orientation and migration of cells in a single L2 gradient (A, C) and in co-existing uniform L1 (0.88 nM) and L2 (17.6 nM) fields (B, D). The 20-fold concentration difference between L1 and L2 based on neutrophil parameters [28] simulates the scenario of CCL19 and CCL21 production in LNs. The cell orientation at steady state is represented by arrows in the figures and the length of the arrows indicates the strength of the orientation. The ligand gradient is represented by contour plot with the highest ligand concentration (17.6 nM) at the center of the contours for the gradient. The ligand concentration at the outmost contour circle is 0.03 nM, and the concentration difference between adjacent circles is 0.9 nM. Because of the magnitude difference between the orientation vector of the cell in different conditions, the length of the arrow is adjusted with a scaling factor of 0.07 for (A) and 15 for (B). The total time of cell migration in (C) and (D) is 150 minutes. Eight representative cell tracks are shown, and the starting positions of the tracks are consistent in all simulations. The end of the tracks is indicated by solid circles.
Figure 7.
Model predictions of cell orientation and migration in a single L1 gradient, a uniform L2 field, and a single L1 gradient with a uniform background of L2.
Comparison of orientation and migration of cells in single L1 gradient (A, D), in a uniform L2 field (B, E), and in co-existing L1 gradient and uniform L2 fields (C, F). The 20-fold concentration difference between L1 and L2 based on neutrophil parameters [28] simulates the scenario of CCL19 and CCL21 production in LNs. The cell orientation at steady state is represented by arrows in the figures and the length of the arrows indicates the strength of the orientation. The ligand gradient is represented by contour plot with the highest ligand concentration (0.88 nM) at the center of the contours for each gradient. The ligand concentration at the outmost contour circle is 0.001 nM, and the concentration difference between adjacent circles is 0.044 nM. Because of the magnitude difference between the orientation vector of the cell in different conditions, the length of the arrow is adjusted with a scaling factor of 2.25 for (A), 3302 for (B) and 0.12 for (C). Simulation results show that cells migrate randomly in a low dose single L1 gradient (D); In a high dose uniform L2 field, cells migrate randomly as expected (E); In co-existing fields of a low dose L1 gradient and a high dose uniform L2, cells migrate away from the L1 gradient (F). The total time of cell migration is 150 minutes. Eight representative cell tracks are shown, and the starting positions of the tracks are consistent in all simulations. The end of the tracks is indicated by solid circles.
Figure 8.
Illustration of the model and its explanation for the repulsive T cell migration.
(A) Illustration of the model for receptor modulations by the 2 ligand L1 and L2. (B) The model provides an explanation for the repulsive migration of cells in a low dose desensitizing ligand gradient (L1 in the model and CCL19 in the experiment) with a high dose uniform background of a nondesensitizing ligand (L2 in the model and CCL21 in the experiment). The desensitizing ligand gradient (L1 in the model and CCL19 in the experiment) causes a differential receptor binding and activation between the front and the back of the cell with more activated receptors that does not lead to chemotaxis toward the gradient at low ligand dose. Although the difference of receptor activation across the cell does not lead to chemotaxis toward the gradient at the low ligand dose, it causes a difference of available free receptors between the front and the back of the cell with less free receptors in the front. As a result, when a nondesensitizing uniform ligand field (L2 in the model and CCL21 in the experiment) is superimposed to the desensitizing ligand gradient, the high dose nondesensitizing ligand binds and activates more receptors in the back than the front of the cell. Additionally, the nondesensitizing ligand activated receptors stay active on the cell surface for chemotactic signaling that reverses the difference of activated receptors between the front and the back of the cells with more activated receptors in the back facing the low concentration side of the desensitizing gradient. Thus, the model suggests that the differential ability of CCL19 and CCL21 for desensitizing CCR7 combined with the physiologically relevant configuration of superimposed CCL19 and CCL21 fields (possibly in the periphery of TCZ) enable the repulsive migration of T cells.
Figure 9.
The proposed combinatorial guidance mechanism.
(A) We propose a possible combinatorial guiding mechanism by different configurations of CCL19 and CCL21 gradient fields for T cell migration in different sub-regions of lymph nodes. CCL21 alone mediates the entry of T cells to the TCZ of LNs through HEV. Inside TCZ, T cells migrate randomly in uniform fields of CCL19 and CCL21 to maximize sampling efficiency with antigen presenting cells (APCs) for immune synaptic interactions. The exit of T cells from LNs is facilitated by first migrating out of TCZ through a CCR7-dependent mechanism. Specifically, T cells migrate away from TCZ when they reach (by random migration) the periphery region of TCZ wherein the gradient fields is likely to be a superposition of a low dose CCL19 gradient and a high dose uniform CCL21 field. This mechanism is enabled by the competition of CCR7 binding between CCL19 and CCL21, together with the differential ability of CCL19 and CCL21 for desensitizing CCR7 and the unique superimposed chemokine field profiles. Such combinatorial guiding mechanism argues the importance and necessity of co-expression of CCL19 and CCL21 in TCZ and the robust design for T cell entry to LNs, navigation within LNs, and exit from LNs using a united CCR7-dependent mechanism. (B–D) Schematic illustration of the hypothesized chemokine fields in different regions in the lymph nodes and the corresponding distributions of signalling CCR7 on the cell surface indicating the cell orientation and migration direction.