2057
The Journal of Experimental Biology 200, 2057–2062 (1997) Printed in Great Britain © The Company of Biologists Limited 1997 JEB0921
TRAFFICKING OF ADOPTIVELY TRANSFERRED B LYMPHOCYTES IN B-LYMPHOCYTE-DEFICIENT MICE RITA ROTH AND MARK J. MAMULA* Section of Rheumatology, Yale University School of Medicine, New Haven, CT 06520-8031, USA Accepted 23 May 1997 Summary intraperitoneal and intravenous injection, B cells were Many studies have investigated the fate of adoptively found in approximately equal numbers in the lymph nodes transferred lymphocytes in recipient mice, although little and the spleen. Two days later, the B-cell distribution in is known of the sites where these transferred cells reside at the lymphoid organs appeared to be independent of the particular time points. Using flow cytometry, we analyzed mode of B-cell transfer. A transient decrease in the the trafficking pattern of adoptively transferred naive B numbers of splenic and lymph node B cells occurred 9 days cells into the lymphoid organs of syngeneic B-cell-deficient µMT) mice. Within the first 24 h of transfer, the location after B-cell transfer (a decrease from 70 to 87 %) prior to (µ of B cells was highly dependent on the mode of B-cell the outgrowth of B cells that occurs 21 days after transfer. transfer. When B cells were injected subcutaneously into These studies are useful for understanding the numbers of µMT mice, they showed a different trafficking pattern B cells that may be required in adoptive transfer studies from cells administered into the peritoneal cavity or and their potential cellular interactions at particular injected intravenously. After subcutaneous transfer into physiological sites based on the route of cell transfer. the thigh, the greatest number of B cells was detected in the popliteal lymph node nearest to the injection site, whereas the lowest number was detected in the axillary lymph node Key words: B-lymphocyte trafficking, B-cell circulation, adoptive transfer, mouse. opposite to the injection side. Within the first 24 h of either
Introduction B lymphocytes recirculate in the body along defined routes as they patrol the body in search of invading antigen. Bonemarrow-derived naive B cells enter secondary lymphoid organs from the blood and migrate through the T-cell rich paracortical area into primary follicles, where they remain for up to 24 h before their recirculation. The B cells then circulate continuously between the blood and lymphoid tissues (Gowans and McGregor, 1965; Howard, 1972; Gutman and Weissman, 1973; Niewenhuis and Ford, 1976; Picker and Butcher, 1992). The circulating B-cell pool of a healthy mouse consists of approximately 200 million cells, most of which survive for more than 4 weeks (Sprent and Miller, 1972; Gray 1988; Forster and Rajewsky, 1990). The trafficking patterns and homing of lymphocytes are dictated by specific molecular signals which are responsible for the preferential recirculation patterns of distinct lymphocyte subpopulations (reviewed in Picker and Butcher, 1992; Springer, 1994). One approach to studying the trafficking of B cells has been by the transfer of radioactively labeled B cells into B-cell depleted (anti-µ-treated) mice (Ron and Sprent, 1987). The recent development of B-cell-deficient (µMT) mice by gene
*Author for correspondence (e-mail:
[email protected]).
targeting (Kitamura et al. 1991) allows the investigation of Bcell trafficking in an environment free of interfering anti-µ antibody and without the need for B-cell prelabeling. In this study, we analyzed the distribution of splenic B cells from syngeneic donor mice into the lymphoid organs of µMT mice following different modes of cell transfer. The identification and quantification of B cells were performed using flow cytometry. Our study shows that the initial distribution of the B cells is highly dependent on the mode of cell transfer. Analysis of recipient mice at later times demonstrated that the population of B cells in peripheral lymphoid organs was less dependent on the route of adoptive transfer. Finally, the recipient mice demonstrate a transient decrease in overall Bcell numbers prior to the outgrowth of adoptively transferred cells. Materials and methods Animals and B-cell adoptive transfer Female B10.A mice, 6–8 weeks of age (NIH, Bethesda, MD), were used as B-cell donors. B-cell-deficient mice (µMT
2058 R. ROTH
AND
M. J. MAMULA
mice; a kind gift of Dr C. A. Janeway, Jr) were generated by disrupting the µ chain constant region of IgM (Kitamura et al. 1991). µMT mice were backcrossed (tenth generation) and maintained on the B10.A background in the animal care facilities at Yale University. Mice made homozygous for the µMT mutation lack B cells expressing B220, sIgM or sIgD (see Fig. 1 and data not shown) and have no detectable serum IgM (data not shown). These mice, at 8–10 weeks of age, served as B-cell recipients. Purified splenic B cells (15×106 cells) in 100 µl of phosphate-buffered saline (PBS) from wild-type mice were subcutaneously injected into the left lower thigh, administered into the peritoneal cavity or intravenously administered into the tail vein of recipient µMT mice. All mice were used in accordance with approved procedures from the Yale University School of Medicine, Division of Animal Care. Yale University is a registered research facility with the US Department of Agriculture in compliance with the American Association for Accreditation of Laboratory Animal Care.
complement-mediated cell purifications were: anti-Thy-1 (Y19; Jones, 1983), anti-CD4 (GK1.5; Dialynas et al. 1983), anti-CD8 (TIB105 and TIB 210) and anti-Mac1 (TIB 128). FITC-conjugated anti-B220 monoclonal antibody (mAb) (RA3-6B2) (Coffman and Weissman, 1981) and FITCconjugated rat IgG2a were purchased from Pharmingen (San Diego, CA). Flow cytometry staining was performed with 5×105 cells in 50 µl of PBS, 0.1 % bovine serum albumin (BSA) and 0.02 % sodium azide in U-shaped microtiter plates. Culture supernatants of monoclonal antibodies were used undiluted in incubations with cells. FITC-conjugated antibodies were used according to the manufacturer’s instructions. Test samples and isotype controls were analyzed on a FACScan instrument (Becton-Dickinson, Mountain View, CA). Preparation of donor B cells B-cell suspensions from wild-type B10.A mice were prepared by gentle disruption of the spleen between frosted glass slides. Erythrocytes were depleted and the remaining cells were washed with Click’s EHAA medium (Irvine
Antibodies and flow cytometry The antibodies utilized in flow cytometric analysis and
A
106
Relative cell number
100
7.5×105 4.5×105
50 3×105 0 0 100
101
102
50
105 102 103 102 103 102 101 102 103 102 103 Fluorescence intensity
B
Fig. 1. Standard curve for the detection and semi-quantitative analysis of B cells. (A) Lymph node cells (LNCs; 0.5×106) from µMT (B-celldeficient) mice were mixed with increasing numbers of B cells, as indicated. Cell fractions were stained with anti-B220 antibody and analyzed by flow cytometry. The histogram on the left shows the staining of µMT LNCs lacking B cells. The individual peaks were spliced into a single figure and represent the B-cell staining with antiB220–FITC obtained from LN cell fractions titrated with B cells (105 to 106). (B) Relative cell numbers taken from A represent the fraction of B cells among total LNCs analyzed (y-axis). These percentages are plotted against the absolute number of B cells added on the y-axes. The results are the means of six titration experiments. Standard deviations were less than 8 %.
Relative cell number (%)
40
30
20
10
0 0
2×105
4×105 6×105 B cell number
8×105
106
B lymphocyte trafficking 2059 a million cells from µMT mice were supplemented with increasing numbers of purified B cells from wild-type B10.A mice (0–106 B cells). The cell populations were then stained with anti-B220–FITC. As illustrated in Fig. 1A, flow cytometry could clearly detect 105 cells within stained populations of LNCs. Fig. 1B shows the signal intensity of the added B cells (as determined in Fig. 1A) plotted as a percentage of total LNCs in the staining mixture. The lower limit of detection is approximately 0.5×105 B cells (as plotted in Fig. 1B). The standard curve is linear (correlation coefficient 0.997) up to 7.5×105 added B cells, where the signal reaches a plateau. This information allows for a semi-quantitative analysis of B cell numbers among LNCs in experimental mice.
Scientific, Santa Ana, CA, USA) supplemented with 2 mmol l−1 L-glutamine, 0.1 mmol l−1 β-mercaptoethanol, 100 u ml−1 penicillin and 100 µg ml−1 streptomycin. Cells were incubated with a cocktail of undiluted culture supernatants of anti-Thy-1 (Y19), anti-CD4 (GK1.5), anti-CD8 (TIB105 and TIB 210) and anti-Mac1 (TIB 128) antibodies at 4 °C for 30 min. Washed cells were then treated with rabbit complement (Low-Tox-M, Cedarlane, Accurate Chemical and Scientific, Westbury, NY, USA) at 37 °C for 30 min. Purified B cells were washed and resuspended in sterile PBS. The purity of B cells was greater than 95 % as analyzed by staining with FITC-conjugated anti-B220 mAb followed by flow cytometry. Quantification of B cells by flow cytometry Lymph nodes (LNs) were removed from µMT mice and gently disrupted between frosted glass slides. Constant numbers of washed LN cells (0.5×106) were supplemented with different numbers of purified splenic B cells from wildtype mice. Cell mixtures were stained with FITC-conjugated anti-B220 mAb and analyzed by flow cytometry as described above.
Trafficking of B cells in B-cell-deficient mice Purified B cells were first injected subcutaneously into the left thigh of µMT mice. After 24 h, single cell suspensions of lymph nodes and the spleen were analyzed by flow cytometry. As shown in Fig. 2, the popliteal LN removed from the side of B-cell injection (left-side cell transfer) was found to contain the greatest number of B cells (up to 35 % of total cells). Fewer B cells (5–12 % of the total) were detected in the inguinal LN (left) isolated from the same side. All lymph nodes examined on the same side as the cell transfer consistently retained a greater number of B cells at early times points than corresponding nodes on the opposite side. Popliteal and inguinal LNCs opposite to the transfer side contained 7 % or fewer B cells, while periaortic nodes contained approximately
Results Detection limits of B cells within lymph node cells in B-celldeficient mice We first examined the sensitivity of flow cytometry in detecting B cells within lymph node cells from µMT mice. Half
Popliteal LN Left 100
Relative cell number
Left
100
35 %
50
Fig. 2. Flow cytometric detection of B cells subcutaneously transferred into µMT recipient mice. The plots of fluorescence intensity histograms illustrate the relative numbers of B cells stained with anti-B220 monoclonal antibody in single cell suspensions of different lymph nodes (LN) and the spleen 24 h after B-cell transfer. Unstained samples from the same cell preparation (overlayed heavy lines) were used to define the area of positive staining marked by the bars.
Inguinal LN Right
0 100
101
102
Right
100
7%
50
0 100
101
102
100
11 %
50
0 100
5%
50
0 101
100
102
101
102
Axillary LN Left 100
100
13 %
50
0 100
0 101
102
100
100
50
100
2%
Spleen
Periaortal LN
Right
50
0 101 102 100 101 Fluorescence intensity
8%
6%
50
0 102
100
101
102
2060 R. ROTH
AND
M. J. MAMULA
8 % B cells after 1 day. Surprisingly, the spleen contained no more than 6 % B cells after 24 h, indicating that this peripheral lymphoid organ is not an early reservoir enriched in adoptively transferred cells. The axillary LN on the injection side contained approximately four times more B cells than that on the opposite side. The transient increase in B-cell number in the popliteal LN on day 3 opposite the subcutaneous transfer site (see Fig. 3) probably reflects the exit of cells from the popliteal node nearest to the site of transfer. Background staining of µMT LNCs or spleen cells with anti-B220 was less than 1 % of the total staining (Figs 1, 2). All plots in Fig. 2 have an overlay showing the background staining of µMT LNCs (heavy lines). We next performed a systematic examination of where B cells traffic at various times under different routes of adoptive transfer into µMT mice (Fig. 3). B cell staining is plotted as a percentage of the number of resident LNCs. As described
above, the greatest number of B cells (35 %) migrated to the draining popliteal LN nearest to the B-cell injection site within 1 day of subcutaneous injection and remained there for the first 3 days. In contrast, the lowest number of B cells (2.4 %) was detected in the axillary LN opposite to the B-cell injection side. One consistent phenomenon of adoptive transfer by the subcutaneous route was that all lymph nodes (popliteal, inguinal and axillary) on the side of cell transfer maintained significantly greater B cell numbers (as much as tenfold greater) than the corresponding node on the opposite side. Other routes of adoptive transfer did not display such polarized trafficking patterns. In contrast to other sites, the periaortal lymph nodes are only transient reservoirs, with cells detectable after 24 h leaving again by day 3. B cells injected intraperitoneally showed a similar distribution in all LNs and in the spleen. The percentage of B cells detected ranged between 8.7±0.8 % (inguinal LN, left) % 20
Popliteal LN (left)
10
0
Popliteal LN (right)
10
0
Percentage of stained cells
Periaortal LN
10
0 Inguinal LN (left)
10
0 Inguinal LN (right)
10
Axillary LN (left)
10
0 Fig. 3. Trafficking of B cells in µMT mice. B cells were injected subcutaneously (A), intraperitoneally (B) and intravenously (C) into µMT mice. The lymph node (LN) indicated in the figure and the spleen were removed 1, 3 and 9 days after B-cell transfer. Single cell suspensions were stained with anti-B220 monoclonal antibody and analyzed by flow cytometry. The percentages of stained B cells (+1 S.E.M.) among the total LNCs, shown by the horizontal bars in Fig. 2, are plotted on the y-axes. Data represent an average of 4–7 experiments.
0 Axillary LN (right)
10
Spleen
10
0
0
A
A AA AA AA AA AA AA AA A
B
C
AA AA A AA A A AA AA AA AA
A AA AA AA AA A A AA AA A
Day 1 Day 3 Day 9
B lymphocyte trafficking 2061 Table 1. Absolute numbers of B cells detected in the lymphoid organs of µMT mice after adoptive transfer Day 1
Day 3
Day 9
0.89×106
0.59×106
(5.9±1.2 %)
(3.9±1.0 %)
0.25×106 (1.7±0.6 %)
Intraperitoneal
1.2×106 (8.0±0.7 %)
0.73×106 (4.9±0.2 %)
0.36×106 (2.4±0.4 %)
Intraveneous
0.87×106 (5.8±0.3 %)
0.69×106 (4.6±0.9 %)
0.11×106 (0.73±0.3 %)
Subcutaneous
Semi-quantitative analysis of B cells in recipient µMT mice was determined 1, 3 or 9 days after transfer by the route indicated. B cell numbers represent the sum of B cells detected in all lymphoid organs. Total B cell numbers were extrapolated from the signal in flow cytometry using a standard curve as illustrated in Fig. 1. Percentages detected among the total number of transferred B cells (15×106) are expressed in parentheses as mean ± S.E.M., N=5.
and 11.4±1.2 % (inguinal LN, right, and the spleen; means ± N=4–7) (Fig. 3). A similar trafficking pattern with an almost equal distribution of the B cells in the LNs and the spleen, ranging between 6.5 and 10 %, was observed with intravenously injected B cells 1 day after the transfer. Three days after the subcutaneous B-cell injection, the number of B cells detected in the popliteal LN nearest to the injection side (left) had dropped by 64 % (Fig. 3). A large decrease (71–87 %) in B cell numbers was also noted in the periaortal LN of all mice, irrespective of the mode of B-cell transfer. Also, in all mice, the B-cell number had decreased in the cell suspensions of the inguinal LN (left) by 20–28 % and in the axillary LN (left and right) by 37–60 % (Fig. 3). This loss of B cells was maintained or decreased further by day 9 in all lymph nodes and the spleen. By day 21, B cells were present in the circulation and numbers began to rise in the peripheral lymphoid organs, indicating that the transfer populations were now multiplying after this transient death of cells (data not shown). Using the standard curve shown in Fig. 1B, we determined the absolute number of B cells detected by flow cytometry (Figs 2, 3) in the lymphoid organs of the µMT recipient mice. Table 1 shows the total number of B cells (from all harvested lymph nodes) discovered after subcutaneous, intraperitoneal or intravenous injection on day 1, 3 and 9. These numbers are also presented as a percentage of the original number of B cells injected. Similar B-cell recovery was obtained with all three types of B-cell transfer, ranging between 5.8 % and 8 % at 1 day post-injection, and between 3.9 % and 4.9 % at 3 days postinjection. Between 0.7 % and 2.4 % of the transferred B cells were detected 9 days after B-cell injection. Other organs, such as kidney and liver, were not examined for B-cell populations. S.E.M.,
Discussion The present study shows that B-cell trafficking and homing in the µMT mouse model can be studied by flow cytometry in
a semi-quantitative manner. Prelabeling of the B cells was not required and thus did not impair the migration or localization pattern of the transferred cells. Major differences were noted in the migration pattern of B cells transferred by different routes after the first day of cell transfer. Subcutaneously injected B cells localized preferentially to the popliteal LN nearest the injection site while intraperitoneally or intravenously injected B cells distributed evenly in the host lymphoid organs. After 24 h, the distribution of subcutaneously injected B cells was similar to that of intraperitoneally and intravenously injected B cells. Our results contrast with an earlier study in which five times more B cells (approximately 10 % of the initially injected radiolabeled cells) were detected in the popliteal LN, while five times fewer B cells (0.2–0.6 %) were detected in the pooled mesenteric, inguinal, axillary and brachial LN 2 days after the subcutaneous transfer of radiolabeled LN B cells into both hind footpads (Ron and Sprent, 1987). The same authors also report a tenfold higher B-cell recovery (8 %) in the spleen and a two- to fourfold lower recovery in the LNs 1 day after intravenous injection of radiolabeled B cells. It is important to note that, in the study by Ron and Sprent (1987), mice were immunized with antigen/CFA 16 h after the B-cell transfer. Previous studies have demonstrated that antigen/CFA induces selective recruitment of circulating lymphocytes (Sprent et al. 1971). Thus, antigen treatment may have led to B-cell accumulation in specific organs because of the inflammatory responses caused by immunization. It remains to be investigated whether other differences in the experimental systems used, such as the source or the labeling of the B cells or the different B-cell detection techniques applied, may account for the observed discrepancies. Nevertheless, both studies demonstrate that intravenously injected B cells (compared with subcutaneously injected cells) localized in much lower numbers, 99 % lower (Ron and Sprent, 1987) and approximately 60 % lower (the present study), in the popliteal LN at 1–2 days post-injection. The principal distinction between the present study and those previously reported is in the phenotype of the recipient mouse. The present study utilized mice made genetically deficient in endogenous B lymphocytes. This phenotype in the recipient mouse makes the identification of transferred cells accurate and unambiguous. No prior studies of B-lymphocyte transfer have utilized this unique recipient mouse. However, we cannot also address the potential effects of cell-to-cell contact (B cell to B cell) in influencing migration patterns observed in other studies. As previously reported by others, large numbers of transferred B cells (40–80 %) fail to become established in recipient lymphoid organs (Sprent and Miller, 1972; Ron and Sprent, 1987). We detected 6–8 % of the initially transferred B cells in the host lymph nodes and the spleen 1 day after cell transfer. A slight decrease (1–3 %) was noticed on day 3, whereas drastically reduced B-cell numbers (0.7–2.4 % of the initially transferred cells) were detected 9 days after transfer. The fate of the ‘missing’ B cells is not yet known. Some of the
2062 R. ROTH
AND
M. J. MAMULA
transferred B cells may have died, although it has been reported that resting B cells survive for at least 2 months when transferred into recipient mice (Gray, 1988). Space within secondary lymphoid organs has been suggested to be a major reason for the more rapid decay of donor LN B cells in nonirradiated recipients as opposed to irradiated hosts (Gray, 1988). However, extreme atrophy of lymphoid tissue appears not to hinder cell trafficking in nude mice (Sprent and Miller, 1972). It is also unlikely that transferred B cells from syngeneic donor mice were eliminated by phagocytic cells, such as macrophages or dendritic cells, although some B cells may have migrated to other organs such as the liver. There is no doubt that the µMT mouse model offers a wide spectrum of biological applications including studies of mechanistic and functional aspects of B-cell trafficking, homing and recirculation, or of the fate of B cells in primary and secondary immune responses. The relevance of the present work lies in understanding the potential cognate B-cell interactions that may occur when different routes of adoptive transfer are used and in predicting the fraction of transferred cells that may be expected to occur in selected nodes. We would like to thank Dr Charlie Janeway for his review of this work and for the contribution of µMT mice. We are also grateful to Renelle Gee for her technical assistance and to Dr Mark Shlomchik for the important monoclonal antibody reagents used in this study. This study was supported by grants from the Lupus Foundation of America and the NIH (AR41032 and AI36529) to M.J.M.
References COFFMAN, R. L. AND WEISSMAN, I. L. (1981). A monoclonal antibody that recognizes B cells and B cell precursors in mice. J. exp. Med. 153, 269–279. DIALYNAS, D. P., QUAN, Z. S., WALL, K. A., PIERRES, A., QUINTANS, J., LOKEN, M. R., PIERRES, M. AND FITCH, F. W. (1983).
Characterization of the murine T cell surface molecule, designated L3T4, identified by monoclonal antibody GK1.5: Similarity of L3T4 to the human Leu-3/T4 molecule. J. Immunol. 131, 2445–2451. FORSTER, I. AND RAJEWSKY, K. (1990). The bulk of the peripheral Bcell pool in mice is stable and not rapidly renewed from the bone marrow. Proc. natn. Acad. Sci. U.S.A. 87, 4781–4784. GOWANS, J. L. AND MCGREGOR, D. D. (1965). The immunological activities of lymphocytes. Prog. Allergy 9, 1–8. GRAY, D. (1988). Population kinetics of rat peripheral B cells. J. exp. Med. 167, 805–816. GUTMAN, G. A. AND WEISSMAN, I. L. (1973). Homing properties of thymus-independent follicular lymphocytes. Transplantation 16, 621–629. HOWARD, J. C. (1972). The life-span and recirculation of marrowderived small lymphocytes from the rat thoracic duct. J. exp. Med. 135, 185–199. JONES, B. (1983). Evidence that the Thy-1 molecule is the target for T cell mitogenic antibody against brain-associated antigens. Eur. J. Immunol. 13, 678–684. KITAMURA, D., ROES, J., KUHN, R. AND RAJEWSKY, K. (1991). A B cell-deficient mouse by targeted disruption of the membrane exon of the immunoglobulin µ chain gene. Nature 350, 423–426. NIEUWENHUIS, P. AND FORD, W. L. (1976). Comparative migration of B- and T-lymphocytes in the rat spleen and lymph nodes. Cell. Immunol. 23, 254–267. PICKER, L. J. AND BUTCHER, E. C. (1992). Physiological and molecular mechanisms of lymphocyte homing. A. Rev. Immunol. 10, 561–591. RON, Y. AND SPRENT, J. (1987). T cell priming in vivo: A major role for B cells in presenting antigen to T cells in lymph nodes. J. Immunol. 138, 2848–2856. SPRENT, J. AND MILLER, J. F. A. P. (1972). Thoracic duct lymphocytes from nude mice: migratory properties and life-span. Eur. J. Immunol. 2, 384–387. SPRENT, J., MILLER, J. F. A. P. AND MITCHELL, G. F. (1971). Antigeninduced selective recruitment of circulating lymphocytes. Cell. Immunol. 2, 171–181. SPRINGER, T. A. (1994). Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 76, 301–314.
