- Email: [email protected]
Splenic migration of xid and non-xid splenic B cells
CLINICAL
IMMUNOLOGY
Splenic
AND
IMMUNOPATHOLOGY
Migration
of xid
39, 379-393 (1986)
and Non-xid
Splenic
HOWARD R. SMITH,NANCYLITTELL,ANDALFRED Cellular
Immunology Diabetes,
Section, Arthritis and Rheumatism and Digestir,e and Kidney Diseases, Bethesda, Maryland
Branch, National 20205
B Cells
D. STEINBERG National Institates
Institute of ArthritiJ, of Health,
The X-linked immune deficiency gene (xid) has been viewed as inducing either a deficiency in a B-cell subset or in B-cell maturation. The present experiments were performed in an attempt to better understand whether (a) xid B cells migrate differently from mature and immature non-xid B cells, and (b) whether the failure of mature B cells to be found in xid spleen, but not Peyer’s patches, is secondary to migratory differences, especially the possible inability of xid spleens to receive mature B cells. We employed the technique of internally fluorescein labeling donor cells and subsequent injection into recipients. Double labeling permitted analysis of B or T cells by two-color flow microfluorometry. Functional studies revealed that labeled cells appropriately responded to TI-1 and TI-2 antigens after migration. We found that (a) adult xid splenic B cells do not migrate to spleen as well as adult non-xid splenic B cells, (b) the migration of adult xid splenic B cells to spleen resembles that of neonatal (xid and non-lid) B cells, (c) equal masses of xid and non-xid spleens have an equal capacity to receive either xid or non-xid splenic B cells, and (d) the migration of xid and non-xid T cells is similar. These results suggest that xid B cells do not migrate normally. but that migratory differences cannot account for the failure of mature B cells to be found in xid spleens. 0 1986 Academic Press. Inc
INTRODUCTION
The X-linked immune deficiency gene (xid)’ has provided a unique opportunity to examine B lymphocytes. Comparisons of B lymphocytes from normal mice to those bearing the xid gene have furthered our understanding of B-cell development, maturation, heterogeneity, and activation requirements. Although the functional consequences of xid have been studied in detail, the underlying reasons for the observed differences continue to be an unsettled issue. It is generally thought that the presence of xid results in either (a) the absence of a B-cell lineage, or (b) the relative failure of B-cell maturation (l-3). Thus, xid can be viewed as a genetic marker for either B-cell subsets or maturation (l-3). The precise nature of the xid defect has yet to be fully elucidated. Although xid B cells have many similarities to both Lyb 5- B cells and neonatal B cells (4, 5), controversy exists as to whether xid B cells are equivalent to the Lyb 5- B cells found in non-xid mice. Recent investigations have shown xid B cells to have both phenotypic (6) and functional (7- 11) properties which have no known normal counterparts. Fur-
’ Abbreviations used: BA, Bruce/la abortus; FITC, fluorescein isothiocyanate; Hepes, N-2 hydroxyethylpiperazine-N’-2-ethanesulfonic acid; HEV, high endothelial venules; LPS, lipopolysaccharide; PFC, plaque-forming cells; SRBC, sheep red blood cells; TNP, trinitrophenyl; xid, X-linked immune deficiency gene; XLR, X-linked gene family. 379 0090-1229/86 $1.50 Copyright All rights
Q 1986 by Academic Press. Inc. of reproduction in any form reserved.
380
SMITH, -LITTELL,
AND STEINBERG
thermore, it appears that xid mice do not have a total arrest in maturation of B cells: in selected lymphoid sites (e.g., Peyer’s patches), young xid mice have mature cell surface antigens (12) and are responsive to TI-2 antigens (13). The experiments presented herein were performed in an attempt to better understand whether xid B cells migrate differently from mature and immature nonxid B cells and whether the failure of mature B cells to be found in xid spleen is secondary to migratory differences. Because the migratory characteristics of cells can be dramatically affected by alterations of surface membranes (14-18), the technique, as developed by Weissman, of internally fluorescein labeling donor cells was employed (19). This technique, which does not alter lymphocyte migration (19), permits analysis of lymphocyte surface membrane phenotypes after migration has occurred. Our results indicate that (a) the migration of xid splenic B cells to spleen differs from that of non-xid splenic B cells, (b) the migration of adult xid splenic B cells to spleen resembles that of neonatal (xid and non-xid) B cells, and (c) equal masses of xid and non-xid spleens have an equal capacity to receive either xid or non-xid splenic B cells. These results suggest that migratory differences exist in xid B cells but they do not account for the failure of mature B cells to be present in xid spleen. MATERIAL
AND METHODS
Mice. Both adult (3 months) and neonatal (7 to 10 days) male and female DBA/2N, DBA/2N.xid (D2.xid), (CBA/N x DBA/2N)F,, and (DBA/2N x CBA/N)F, mice were obtained from our own facilities. All mice were individually ear tagged, housed in our animal facility, and allowed free access to food and water. Antigens. TNP,,-AECM-Ficoll was purchased from Biosearch (San Rafael, Calif.). TNP-BA was kindly provided by Dr. J. Mond, Uniformed Services University of the Health Sciences (Bethesda, Md.). TNP-LPS (butenol extracted) was purchased from Sigma Chemical Company (St. Louis, MO.). Cell cultures. Spleen cells from mice killed by cervical dislocation were prepared by gentle teasing and the cells were washed in balanced salt solution. The standard tissue culture medium consisted of RPM1 1640 supplemented with 10% bovine calf serum, 2 mM L-glutamine, 10 mM Hepes, nonessential amino acids, sodium pyruvate, 5 x lo-’ M 2-mercaptoethanol, and penicillin-streptomycin. Washed cells (250 p,l of 4 x lo-‘j/ml) were cultured in 96-well Costar 3596 microtiter plates (Cooke Co., Alexandria, Va.) in a humidified 5% CO,-95% air incubator at 37°C for 4 days. Appropriate wells received optimal doses of TNP-Ficoll (50 &ml), TNP-BA (1: IO5 of stock), or TNP-LPS (10 pg/ml) and were cultured for 4 days. Plaque-forming cell assay. Sheep red blood cells were conjugated with TNP by the method of Rittenberg and Pratt (20). Direct anti-TNP PFC were enumerated by the Cunningham slide modification of the Jerne hemolytic plaque assay (21). Background PFC to TNP were subtracted. Standard errors were less than 20% of the means. Cell labeling. Cells were labeled as described by Butcher and Weissman (19). A stock solution of fluorescein isothiocyanate (FITC; Sigma) was prepared and fil-
MIGRATION
OF xid AND
non-d
SPLENIC
B CELLS
381
tered. The concentration was determined spectrophotometrically. Spleen cells pooled from at least four mice were prepared as described above, washed in RPM1 1640, and resuspended at 2.5-5.0 x 10’ cells/ml in medium (adjusted to pH 7.0) containing equal parts of Medium 199 and phosphate-buffered saline (Biofluids, Rockville, Md.), and 5% bovine calf serum (Hyclone, Logan, Utah). Cells were labeled with FITC (40 pg/ml) at 37°C for 30 min. Cells were washed with cold RPM1 1640, centrifuged, resuspended in RPM1 1640, layered over 5 ml (Falcon tube 2095, Oxnard, Calif.) of bovine calf serum, and centrifuged. Cell migration. Fluorescein labeled cells were resuspended at 100 x lo6 cells/ ml and 0.5 ml was injected intravenously into at least four recipient mice. After 2 hr, unless otherwise indicated, recipient mice were sacrificed. Single cell suspensions of spleen, liver, or lung were prepared in RPM1 1640 medium by gentle teasing with forceps or by placing the organ between two glass slides and applying gentle pressure, followed by passage through a 25-gauge needle. Cells were allowed to settle for 5 min to remove clumps. Fluorescence-positive cells thus represented cells of donor origin. The percentage of migrating donor cells could be expressed either as a percentage of injected donor cells which were recovered in a particular organ (e.g., spleen, liver, lung) or as percentage of all of the cells in an individual organ which was from the donor. Significant differences between sample groups were determined using Student’s t test. To further analyze how many donor B cells or T cells had migrated to recipient spleens, single cell suspensions of recipient organs were stained with a second antibody (red labeled) to IgM or Thy 1. For example, if the second stain was anti-IgM, four populations of cells were possible, namely (a) those which had no color and were recipient non-B cells, (b) those which were red only and were recipient B cells, (c) those which were green only and were donor non-B cells, and (d) those which were green and red and were donor B cells. Cell staining andflow cytometry. Single cell suspensions at lo6 cells/ml of recipient organs were prepared as described above, washed, and resuspended in sorter buffer (phosphate-buffered saline containing 0.5% bovine serum albumen and 0.05% sodium azide). Cells were incubated with previously titrated biotinylated monoclonal antibodies for 20 min at 4”C, washed, and resuspended in buffer. Previously titered Texas red avidin or FITC-avidin was incubated with the cells for 20 min at 4°C. A biotinylated monoclonal antibody to Thy 1.2 was purchased from Becton-Dickinson (Sunnyvale, Calif.). Biotinylated F(ab’), goat anti-mouse IgM was purchased from Zymed Laboratories (San Francisco, Calif.). Cells were washed and resuspended in 0.5 ml sorter buffer. Flow cytometry was performed on a Becton-Dickinson FACS-II equipped with an argon laser emitting 200 mW at 488 nm and a krypton ion laser emitting 150 mW at 568 nm. Fluorescence data were collected with logarithmic amplification. For each sample, data from 50,000 to 100,000 light scatter-gated viable cells were collected. Data were analyzed with use of computer programs as previously described (22). RESULTS
Differences in migration
of xid and non-xid spleen cells in syngeneic recipients.
SMITH,
382
LITTELL,
AND STEINBERG
Initial studies suggested that differences existed in the migratory ability of xid and non-xid spleen cells. In order to better assess these differences, the short-term kinetics of migration were studied. Single cell suspensions of FITC-labeled spleen cells from xid (CBA/N x DBA/2)F, male and non-xid (DBA/2 x CBA/N)F, male mice were injected into syngeneic mice. At various times after injection, recipient spleens were obtained and the percentage of FITC-positive donor cells which had migrated to the recipient spleen was determined by flow microfluorometry. Multiple experiments demonstrated that the percentage of syngeneic FITC-labeled spleen cells migrating to the spleen tended to slowly decrease with time and was consistently greater in non-xid (DBAI2 x CBA/N)F, mice than xid (CBA/N x DBA/2)F, mice (Fig. 1). Similar results were obtained using congenic DBA/2 and D2.xid mice (results not shown). It was necessary to ensure that the migratory cells were actually located within the spleen as opposed to merely residing within the blood vessels of the spleen. At the various times as indicated in Fig. 1, the percentages of internally FITC-labeled donor spleen cells in peripheral blood were consistently ~0.2%. In addition, histologic examination of frozen sections of the spleen with examination for the placement of fluorescein-labeled cells revealed that the migrating cells were within the lymphoid areas of the spleen. Sections of recipient liver and lung were studied at the 2-hr postinjection time point for nonspecific trapping of donor cells. There was no difference in trapping between donor xid and donor non-xid cells. Fluorescein-labeled xid (CBA/N x DBA/2)F, male, or non-xid (DBA/2 x CBA/N)F, male, spleen cells were intravenously injected into both xid and non-xid recipients. There were no significant differences in the percentage of xid and non-xid spleen cells which lodged in the liver or lung in either xid or non-xid recipients. Specifically, the percentages of fluorescein-positive xid donor cells that were recovered in the livers of xid and non-xid recipients were 27.4 and 28.8%, respectively. The percentages of fluores-
10,
I
@BAD
X CBAlNlF,b
ICBAiN X DBAOIF,
I 2
6 HOURS
I 12 AFTER
6
I 18
I 24
INJECTION
FIG. 1. Kinetics of migration to spleen of syngeneic FITC-labeled spleen cells. spleens from (CBA/ N x DBA/2)F,, xid/Y, and (DBA/2 x CBA/N)F,, +A’, were made into single cell suspensions, FITC labeled, and 5 x 10’ injected into each of a large number of syngeneic mice. At various times after injection, several recipient spleens were obtained and the percentage of FITC-labeled cells which had migrated to each recipient spleen was determined by flow microfluorometry. The percentage of syngeneic FITC-labeled spleen cells migrating to the spleen was consistently greater in non-xid (DBA/2 x CBA/N)F, mice than in xid (CBA/N x DBA/2)FI mice.
MIGRATION
383
OF xid AND non-xid SPLENIC B CELLS
cein positive non-xid donor cells recovered in the livers of xid and non-x-id recipients were 29.7 and 2.5.2%, respectively. An equal percentage (10.0%) of labeled donor xid spleen cells was recovered in the lungs of either xid or non-xid recipients. Similarly, 11 .O and 12.8% of labeled non-xid spleen cells were recovered in xid and non-xid lungs, respectively. Thus, the nonspecific trapping of xid and non-xid spleen cells in nonlymphoid organs was similar. More importantly, donor cells which migrated to the spleen were found to localize within the lymphoid areas of the spleen. Ability of FITC-labeled spleen cells to respond to antigen. Previous studies have revealed that the inability of spleen cells from xid mice to respond to type 2 antigens (e.g., TNP-Ficoll) can be corrected after injection of histocompatible non-xid spleen cells (23) and that this response to type 2 antigens is solely attributable to the donor non-xid cells (23). The inability of xid cells to respond to TNP-Ficoll was utilized in studying spleen cell function after FITC labeling. Either non-xid or xid donor spleen cells were FITC labeled and intravenously injected into xid recipients. After 2 hr, the recipient spleens were removed, made into single cell suspensions, and cultured with various antigens. The response to antigen was determined at Day 4 by enumeration of plaque-forming cells. Spleen cell cultures from xid mice which had been injected with FITC-labeled non-xid spleen cells responded to TNP-Ficoll, whereas those receiving FITC-labeled xid spleen cells failed to respond (Table 1). Thus, a function solely attributable to the non-xid donor spleen cells, namely the response to TNP-Ficoll, remained intact. The responses to the type 1 antigens TNP-BA and TNP-LPS were present in both groups and are attributable to both recipient (xid) and donor (either xid or non-xid) spleen cells. These data indicate that donor spleen cells retain their functional ability to respond to antigen after FITC labeling, in vivo injection, and migration to spleen. Migration of FITC-labeled xid and non-xid spleen cells to xid and non-xid spleen. The percentage of fluorescence-positive spleen cells recovered in spleens
after injection of syngeneic FITC-labeled spleen cells is greater in non-xid mice than in xid mice (Fig. 1). In an attempt to determine whether these observed differences were due to differences in (a) inherent migratory abilities of spleen cells and/or (b) the capability of spleens to receive such cells, both non-xid and TABLE ABILITY
OF INTERNALLY FITC-LABELED
Donor”
+ IY’ xidlY
Recipient xidlY xidiY
1
NON-xid
DONOR SPLEEN CELLS TO IN RECIPIENT xid MICE
TNP-Ficollb 135 5
TNP-BAb 195 88
RESFQNDTO TNP-FXOLL TNP-LPSb 95 44
LIDonor cells were FITC labeled and injected into recipients. After 2 hr, recipient spleens were removed, made into single cell suspensions, and cultured with or without the indicated antigen. b Numbers of plaque-forming cells/106 spleen cells, to TNP-SRBC on Day 4. Standard errors were ~20%. No antigen (“background”) gave ~5 PFC/106 spleen cells. ‘xid/Y = (CBA/N x DBA/2)F,; +iY = (DBAIZ x CBA/N)F,.
SMITH,
384
LITTELL,
AND
STEINBERG
xid mice were injected with FITC-labeled spleen cells from either non-xid or xid mice. Experiments representative of three similar ones are shown. Donor spleen cells were determined by flow microfluorometry. The percentage of donor xid spleen cells migrating to non-xid (+/Y) spleens was 17.0 + 1.6% and to xid spleens was 10.2 + 2.0% (Table 2). The percentage of donor non-xid spleen cells migrating to non-xid (+N) spleens was 26.5 r 1.0% and to xid spleens was 16.5 & 1.O% (Table 2). This decreased absolute capacity of xid spleens as compared to non-xid spleens to receive transferred cells does not account for the smaller size of the xid spleen. To correct for differences in the spleen size, these data can be expressed as a ratio of donor xid cells to donor non-xid cells migrating to recipient spleen and then the ratios compared. For non-xid recipients, the ratio of the percentage of migrating donor xid to donor non-xid was 0.64 (17.0/26.5%) (Table 2). Similarly, for xid recipients, this ratio of the percentage of migrating donor xid to donor non-xid cells was 0.62 (10.2/16.5%) (Table 2). Furthermore, not calculated in Table 2, the ratio of the percentage of donor xid spleen cell migrating to recipient xid and non-xid spleens was 0.60 (10.2/17.0%). The ratio of the percentage of donor non-xid spleen cell migration to recipient xid and non-xid spleens was 0.66 (16.5/26.5%). Thus, as a function of either donor migration, or recipient capacity, these ratios remain remarkably similar. These data indicate that (a) there is a decreased absolute capacity of xid spleens to receive transferred cells, (b) when one accounts for the differences in spleen size, equal masses of xid and non-xid spleens have similar capacities to receive spleen cells, and (c) the migration to spleen of non-xid spleen cells exceeds that of xid spleen cells. TABLE
2
MIGRATION OF FITC-LABELED xid AND NON-xid SPLEEN CELLS TO SPLEENS OF xid AND NON-xid RECIPIENTS: EQUAL RELATIVE CAPACITY OF + N AND xidN SPLEENSTO RECEIVE DONOR CELLS
DonoF xidNc
Recipient l kN’
Percentage of donor spleen cells migrating to recipient spleen (absolute capacity)
Ratiob (relative capacity)
17.0 + 1.6d 0.64’
f l.Of 10.2 * 2.0
+N xidN
+N xidlY
26.5
+N
xidlY
16.5 f 1.0
0.62 g
a Donor spleen cells were internally FITC labeled and injected into four recipients. The percentage of injected donor spleen cells migrating to the spleen was determined at 2 hr by flow microfluorometry. Recipient spleen ceil autofluorescence (with or without injected non-FITC-labeled spleen cells) was 0.05, Student’s t test. c The ratio of (xid/Y into + N) over (+/Y into + N) is shown. f Statistically different from (+/Y into xidN), P > 0.05, Student’s t test. c The ratio of (xid/Y into xid/Y) over (+ N into xid/Y) is shown.
MIGRATION
OF xid AND non-d
SPLENIC
B CELLS
385
We further analyzed the percentage of cells in an individual recipient organ which is of donor cell origin. The percentage of spleen cells that was fluorescence positive was determined by flow microfluorometry. After injection with FITC-labeled donor xid spleen cells, the percentages of non-xid and xid recipient spleen cells that were fluorescence positive and thus of donor cell origin, were 5.7 and 5.9, respectively (Table 3). In contrast, following injection of FITC-labeled donor non-xid spleen cells, the percentages of non-xid and xid spleen cells that were fluorescence positive were greater, 9.2 and 11.8, respectively (Table 3). Thus, the xid spleens were able to receive +/Y cells as well as -t/Y spleens could receive + /Y cells. Migration of neonatal spleen cells to adult spleen. Many similarities exist between the B cells of xid mice and those of neonatal non-xid mice (4, 5) and may indicate that the xid defect reflects a failure in B-cell maturation. Recently, however, significant differences have been found between xid and neonatal non-xid B cells (6- 11) suggesting that xid B cells may be intrinsically abnormal and have no normal counterpart (6, 8, 11). Because donor xid cells differ from non-xid cells in their migration, it would be of interest to compare xid and neonatal non-xid spleen cell migration. Neonatal donor spleen cells were FITC labeled and injected into adult recipients. After 2 hr, the percentage of fluorescence-positive cells in recipient spleens was determined. It was observed that migration of neonatal xid spleen cells to spleen did not significantly differ from that of neonatal non-xid spleen cells (Table 4). However, the migration to spleen of both neonatal xid and neonatal non-xid spleen cells was significantly less than that of adult nonxid spleen cells (Table 4). Adult xid spleen cell migration did not differ from neonatal xid spleen cell migration (Table 4). Thus, the migration to spleen of xid spleen cells closely resembles that of neonatal but not adult non-xid spleen cells.
TABLE 3 MIGRATION OF FITC-LABELED xid AND NON-xid SPLEEN CELLS TO SPLEENSOF xid AND NON-xid RECIPIENTS: EQUAL ABILITY OF + IY CELLS TO MIGRATE TO xidlY AND + IY RECIPIENT SPLEENS
Donor”
Recipient
xidlYb xidlY
+lYb xidlY +lY lid/Y
i-/Y +/y
Percentage fluorescencepositive spleen cells in recipient spleen 5.7 k 0.8’ 5.9 2 1.0 9.2 IT 0.5d
11.8 2 0.4
n Donor spleen cells were internally FITC labeled and injected into four recipients. The percentage of cells in the recipient spleen which were of donor cell origin and were fluorescence positive was determined at 2 hr by flow microfluorometry. Recipient spleen cell autofluorescence (with or without injected non-FITC-labeled spleen cells) was < 1.O%. bxidh’ = (CBA/N x DBAlZ)F,; +lY = (DBAR x CBAlN)F,. c Not statistically different from (xidlY into xidly), P > 0.05; statistically different from (+ IY into +/Y), P < 0.01, Student’s f test. d Not statistically different from (+lY into xidlY). p > 0.05; statistically different from (xid/Y into xidlY), P < 0.05, Student’s t test.
386
SMITH,
LITTELL,
MIGRATION OF xid AND NON-xid
AND STEINBERG
TABLE 4 NEONATAL SPLEEN CELLS TO ADULT SPLEEN Percentage FITC-positive spleen cells in recipient”
Donor xidNb
+/y xidlY +N
Age
Adult recipient
Neonate Neonate Adult Adult
xidlY xidlY xidN xidN
Expt 1 5.6 4.2 5.3 10.3
f f * f
Expt 2
0.7’ 0.5 0.4’ 0.7e
4.8 ?I 0.8” 6.0 -r- 0.3 NDd NDd
u Donor spleen cells were (internally) FITC labeled and injected into four adult recipients. The percentage of fluorescence-positive donor spleen cells migrating to recipient spleen was determined at 2 hr. Recipient spleen cells autofluorescence was < 1.O%. b xidN = (CBA/N x DBA/Z)F, and +/Y = (DBA/2 x CBA/N)F,. c Not significantly different from neonatal + N donor, P > 0.05, Student’s t test. d Not determined. e Significantly different from all other groups, P < 0.01, Student’s 1 test.
The xid migratory defect maps to the B cell. Analysis of the migratory abilities of populations of B cells and T cells from unfractionated spleens was performed by means of two-color flow microfluorometry. Donor spleen cells were internally labeled with FITC and injected into syngeneic recipients. After 2 hr, recipient spleens were removed, the cell surface membranes stained with biotinylated antiIgM or biotinylated anti-Thy 1.2, and Texas-Red avidin was added. The percentage of two-color-positive (green and red) spleen cells was determined. The percentage of total donor spleen cells migrating to spleen was greater in (DBA/2 X CBA/N)Ft +/Y mice (non-xi4 (10.3%) than in (CBA/N x DBA/2)F, xid/Y mice (7.4%) (Table 5). This difference in total donor cells migrating to spleen was TABLE 5 MIGRATION OF SPLENIC B CELLS, T CELLS, AND NON-B, NON-T CELLS FROM xid AND NON-xid SYNGENEICSPLEEN
MICE TO
Double-labeled cells” as percentage of recipient spleen cells (DBA/2 x CBA/N)F, (-t/Y) B cells T cells Non-B/non-T cells (by subtraction) Total cells
5.2 f 0.7b 3.1 2 0.56 2.0 10.3 f 1.4
(CBA/N x DBA/2)F, (xid/Y)
2.1 * 0.2 3.6 k 0.6 1.7 7.4 f 1.1
D Donor spleen cells were (internally) FITC labeled (green) and injected into four syngenic recipients. Recipient spleen cells were removed after 2 hr, the surface membranes stained with either biotinylated anti-IgM or biotinylated anti-Thy 1.2, and Texas red avidin was added (red). Two-color positive (green and red) spleen cells were determined as a percentage of recipient spleen cells. b Statistically different from xid/Y recipients, P < 0.05, Student’s t test. c Not statistically different from xidN recipients, P > 0.05, Student’s f test.
MIGRATION
OF xid AND non-xid
SPLENIC
387
B CELLS
largely due to the increased percentage of migrating non-xid B cells (5.2%) compared to migrating xid B cells (2.1%) (Table 5). Recipients of non-xid and xid spleen cells had a similar percentage of two-color-positive donor T cells (3.1 and 3.6%, respectively) and donor non-B, non-T cells (2.0 and 1.7%, respectively) (Table 5). This result was confirmed in repeated experiments. Furthermore, virtually no differences were discernable in the staining profile of donor T-cell fluorescence intensity between non-xid and xid mice (Fig. 2). Thus, all differences in migration between xid and non-xid spleen cells can be attributed to differences in B-cell populations in the respective donors. The xid B-ceil migratory defect does not result from xid recipient abnormalities. To further evaluate if the observed differences in B-cell migration was secondary to (a) an inability or decreased capacity of xid spleens to receive certain
populations
of B cells (e.g., dull IgM+)
; ‘~~~
03A/ZxCBAIN)F,
-
ICBA/NxDBA/21F,
or (b) a deficit in the migratory
K33A/NxDBA/21F, XidiY
1Ol I ldo FLUORESCENCE INTENSITY FITC
-
ability of
IDBAIZxCBAINIF, +/y
10 (green)
FIG. 2. T-cell migration to spleen of FITC-labeled spleen cells. Donor spleen cells from either (DBA/Z x CBA/N)F,, +/Y, (non-xi&, or (CBA/N x DBA/2)F,, xid/Y, (xid) mice were intkmally labeled with FITC (green). These cells were injected into either (DBA12 x CBA/N)F,, +/Y, or (CBA/N x DBA/2)F,, xid/Y, mice. Recipient spleens were removed after 2 hr, made into single cell suspensions, and the cell surface stained with anti-Thy 1.2 (Texas red). Each axis represents increasing intensity of fluorescence on a log scale (X axis = FITC; Y axis = anti-Thy 1.2). The percentages and profiles of double-labeled (migrating) donor splenic T cells were similar in all groups.
388
SMITH,
LITTELL,
AND
STEINBERG
donor xid spleen cells, advantage was taken of the characteristic IgM profiles of non-xid (both dull and bright IgM+) and xid (predominantly bright IgM+) spleen cells (5). Donor spleen cells were FITC labeled and injected into recipients. After 2 hr, recipient spleens were removed, made into single cell suspensions, and the cell surface stained with anti-IgM (Texas red). Donor IgM+ spleen cells appear as two-color positive (green and red) and are therefore distinguishable from recipient spleen cells (not green) or donor non-IgM+ cells (green only). By computer analysis, the independent IgM (red) fluorescence profiles of recipient (green negative) and donor (green positive) spleen cells ware obtainable (Fig. 3). Non-xid mice receiving either non-xid or xid spleen cells demonstrated the capacity to receive both bright and dull IgM+ cells as evidenced by the characteristic donor IgM profiles (Fig. 3C). Similarly, xid recipients had the capacity to receive both bright and dull IgM+ spleen cells (Fig. 3D). Thus, both non-xid and xid recipients have the capacity to receive non-xid and xid splenic B cells. Therefore, the reduced migration to spleen of splenic xid B cells does not result from an inability of the xid spleen to receive such cells, but from an inherent migratory defect of xid B cells. Mixing experiments. In order to correct for the presence in non-xid spleens of both more B cells and more “normal” B cells (3, 5), mixing experiments were performed. Either xid or non-xid FITC-labeled spleen cells were mixed in various percentages with unlabeled spleen cells of the opposite kind. After injection into xid recipients, cells were harvested and labeled with anti-surface IgM (red). The percentage of FITC-labeled (green) donor splenic B cells migrating to recipient spleen was determined correcting for differences in donor B-cell numbers. Comparison of the slope of the lines of non-xid and xid B-cell migration reveals that non-xid B cells have a greater ability to migrate to spleen than do xid B cells (Fig. 4). Moreover, the straight lines indicate that there is minimal, if any, interference by one population of B cells with the migration of the other. Thus, inherent differences exist in the migratory capacity of non-xid and xid splenic B cells. DISCUSSION
In the present study, we have presented evidence for differences in the migratory ability of B cells, but not T cells, from xid and non-xid mice. The migration to spleen of adult xid B cells resembles that of neonatal B cells from both xid and non-xid mice. Furthermore, these observed differences in B-cell migration are not due to differences in the capacity of equal masses of xid and non-xid spleens to receive such cells, but rather, appear to be due to intrinsic differences in the B cells. Using a different approach, Sprent and colleagues have recently made similar observations (24). Many factors are known to be important in the migration of lymphocytes, (reviewed in (14)) however, little is known about the precise mechanisms regulating lymphocyte migration to the spleen. Spleen does not have the postcapillary high endothelial venules (HEV) through which lymphocytes migrate in order to enter and gain access to lymph nodes and Peyer’s patches (25-27). Indeed, the mechanism of migration to spleen differs from the HEV-mediated migration of lymph node and Peyer’s patches as demonstrated by the dramatic decrease in lymphocyte migration to lymph node and Peyer’s patches but not to spleen after trypsin
MIGRATION
OF rid AND non-xid SPLENIC
1.0,
B CELLS
389
A
0.8 1
Donor IDEA/2 x CBA/NIF, +/y
0.61
0.8
;*.
Donor ICBNN
x DBAl21F, ad/Y
:‘: I :
0.6 d
0.4
$I a 3
0.2
5 0
l.Or
:.
/
',, '.. -._ .
.
_ ,,.,,
'.,"'
C (DBAIP x CBAlNlF,-IDBAIP x CBA’NIF, +lY +/y ICBAIN x DBA/2lF, - IDBAl2 x CWN)F, id/Y +/y
i (DBAl2
x CBAfN)F,
FLUORESCENCE
-
(CBA/N x DB&‘PIF, xtd/Y
INTENSITY ANTI-IgM
,-I , ,
,-,
(RED)
FIG. 3. B-cell migration of donor spleen cells; Anti-IgM fluorescence profile. Donor spleen cells were FITC labeled (internally) and injected into recipients. Recipient spleens were removed after 2 hr, made into single cell suspensions, and the cell surface stained with anti-IgM (Texas red). The characteristic splenic IgM profiles of non-xid (A) (both dull and bright IgM+) and xid (B) (relative reduction in dull IgM+) spleen cells were observed. Cells that had migrated to the spleen were selected on the basis of green fluorescence. These were then further analyzed with regard to IgM profiles (red). The IgM (red) profiles of two-color positive (green and red) donor splenic B cells migrating to recipient spleens are shown (C and D). The data indicate that (i) both non-xid and xid recipients have the capacity to receive both non-xid (-) and xid ( . . . . ) spleen cells. and (ii) the donor cells retain their characteristic IgM profiles despite migration to recipient spleens.
treatment of lymphocytes (16, 17). Although the structures which regulate the uptake into spleen of migratory lymphocytes are unknown, B cells do migrate better to spleen (and Peyer’s patches) than do T cells (28). The present studies were performed to determine whether or not the spleen of xid mice might fail to
390
SMITH,
PERCENT
LITTELL,
AND STEINBERG
OF INDICATED
SPLEEN CELLS INJECTED
FIG. 4. Percentage of donor splenic B cells migrating to recipient spleens. Either xid or non-xid (+/Y) internally FITC-labeled spleen cells (green) were mixed in the percentages indicated with unlabeled non-xid or xid spleen cells, respectively. Thus, on the abscissa the 25% point on the xid/Y line represents a cell mixture in which 25% of the cells were FITC-labeled xid cells and 75% were unlabeled +/Y cells. The 25% point on the +/Y line represents a mixture in which 25% of the cells were FITC-labeled + N cells and 75% were unlabeled xid/Y cells. After injection into xid recipients, cells were harvested and labeled with anti-surface IgM with a second color (red) as in Fig. 3. The percentage of two-color-labeled donor splenic IgM+ B cells which had migrated to the recipient spleens is shown. A greater percentage of non-xid splenic B cells migrates to xid spleen relative to xid splenic B cells. There is no interference of xid B-cell migration by non-xid B cells or vice ver~a as indicated by the straight line.
receive mature B cells and thereby explain the failure of mature B cells to be found in their spleen. Such a result could come about by a failure of the xid spleen to have a receptor to which mature B cells might bind. We found, herein, that the xid spleen has no defect in receiving mature B cells. Thus, the finding of mature B cells in Peyer’s patches, but not spleen, of xid mice cannot be explained by an inability of xid spleen to receive migrating mature B cells. More likely, the local immune stimulation in Peyer’s patches accounts for the finding of mature B cells in xid Peyer’s patches. This formulation is further supported by the ability of polyclonal stimulation of xid mice to induce mature B-cell functions (29). The other possibility is that the xid B cells lack a receptor by which those cells could grow and mature in the spleen. Evidence was found that xid cells did not migrate as well as non-xid B cells to the non-xid spleen; however, there was suffrcient migration of xid B cells to non-xid spleen to render unlikely the possibility
MIGRATION
OF xid AND
non-d
SPLENIC
B CELLS
391
that a migration-specific cell surface receptor was sufficient to explain the extreme lack of mature B cells in the spleens of xid mice. Whereas evidence has been obtained for a defect in a cell surface glycoprotein in Wiskott-Aldrich syndrome (30) and could lead to such a migration defect, we do not believe that the evidence from our migration studies of xid cells points to such a defect. In the present study, spleen cell suspensions were analyzed for the presence of fluorescein-tagged cells. The utility of the method depends upon the stability of the marker which has been well demonstrated and which we have confirmed. This method avoids the problems of such markers as SICr which may leach out of one cell and be picked up by another. The study also requires identification in recipient spleens of subpopulations of migrating donor cells. One approach could have been isolation of subpopulations of cells prior to injection, however, such isolation procedures might have not been complete, and, more importantly, might have damaged cell membrances so as to alter their migratory ability. Instead, unmanipulated cells were internally labeled and analyzed by surface label only after they had migrated to recipient spleen. This method allowed an easy identification of B cells and T cells and even preserved the characteristic subtle surface IgM profiles of donor B-cell subpopulations. Thus, we believe that the method was quite satisfactory for the present study. A potential problem in the present experiments concerns the location of lymphocytes within the spleen versus lymphocytes in the vascular spaces of the spleen. Our results depend upon the measurement of cells which have migrated to the spleen as opposed to merely residing in the blood vessels of the spleen. As explained in the results, the contribution of the blood to the fluorescein-tagged cells in the spleen was negligible. However, the splenic cells could be in the red pulp and merely be percolating through the spleen, or actually be part of the lymphoid structures of the spleen. Histologic examination of frozen sections of the spleens with examination for placement of fluorescein-tagged cells clearly demonstrated that the migrating cells were within the lymphoid areas of the spleen. Thus, the data demonstrate the migration of lymphocytes to the spleen rather than their mere presence within the vascular spaces or the red pulp. Even modest damage would leave lymphocytes in the red pulp; thus the latter finding is a further confirmation of the viability of the migrating cells. Furthermore, the possibility of an unrelated interaction of xid cells with other organs in viva was examined. Differential trapping of xid cells as compared to non-xid cells in nonlymphoid tissue, such as liver or lung, could account for the observed differences. Such was not the case: both xid and non-xid spleen cells were found equally in both liver and lung. Specific homing markers have been found for certain lymphocytes. The best studied are those which allow lymphocytes to specifically enter either peripheral lymph nodes, Peyer’s patches, or both (25-28). The receptor which allows lymphocytes to enter lymph nodes can be recognized by the MEL-14 reagent and binds to a specific complementary structure on high endothelial venules (31). Whether or not analogous specific receptor-ligand interactions are critical for migration to the spleen is unknown. In this regard, the xid and non-xid B cells differ by a number of determinants, at least quantitatively, including IgD, IgM,
392
SMITH,
LITTELL,
AND STEINBERG
Lyb 3, Lyb 5, and Lyb 7 (3-6); one or more of these markers may serve as a structure which facilitates migration to the spleen. Further studies of specificity of migration to the spleen may, in the future, shed light on this question. The present studies are consistent with the hypothesis that the major defect in xid mice is one of B-cell maturation. This possibility would explain the similar migration capabilities of young xid and young non-xid cells as well as young nonxid and older xid cells. It would also be consistent with the idea that the B cells of older xid mice do not have a normal counterpart (6, 8, 11): normal cells are able to complete the maturation process whereas xid B cells have a marked arrest in maturation. In addition, such a formulation is consistent with recent molecular genetic studies of xid mice which indicate that an X-linked gene family (XLR) is expressed only in the most mature B cells (32, 33), B cells which are markedly under-represented, especially in spleen, in xid mice (34). ACKNOWLEDGMENTS We express our thanks to Ms. Cindy Mizgala for her excellent technical assistance, Dr. Thomas Chused and Ms. Linette Edison for help with flow microfluorometry, and Ms. Betty Irene Roupe for preparation of the manuscript.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
Kincade, P. W., Moore, M. A. S., Lee, G., and Paige, C. J., Cell. Immunol. 40, 294, 1978. DeFranco, A. L., Kung, J. T., and Paul, W. E., Zmmunol. Rev. 64, 161, 1982. Scher, I., Adv. Immunol. 33, 1, 1982. Ahmed, A., Scher, I., Sharrow, S. O., Smith, A. H., Paul, W. E., Sachs, D. H., and Sell, K. W., J. Exp. Med. 145, 101, 1977. Scher, I., Beming, A. K., Kessler, S., and Finkelman, F. D., J. Zmmunol. 125, 1686, 1980. Hardy, R. R., Hayakawa, K., Parks, D. R., and Herzenberg, L. A., Nature (London) 306, 270, 1983. Ono, S., Yaffe, L. J., Ryan, J. L., and Singer, A., J. Immunol. 130, 2014, 1983. Webb, S. R., Mosier, D. E., Wilson, D. B., and Sprent, J., J. Exp. Med. 160, 108, 1984. Wortis, H. H., Burkly, L., Hughes, D., Rochelle, S., and Waneck, G., J. Exp. Med. 155, 903, 1982. Mond, J. J., Scher, I., Cossman, J., Kessler, S., Mongini, P. K. A., Hansen, C., Finkelman, E D., and Paul, W. E., J. Exp. Med. 155, 924. 1982. Sprent, J., and Bruce, J., J. Exp. Med. 160, 711, 1984. Eldridge, J. H., Kiyono, H., Michalek, S. M., and McGhee, J. R., J. Exp. Med. 157, 789, 1983. Eldridge, J. H., Yaffe, L. J., Ryan, J. J., Kiyono, H., Scher, I., and, McGhee, J. R., J. Zmmunol. 133, 2308, 1984. de Sousa, M., “Lymphocyte Circulation: Experimental and Clinical Aspects,” Wiley, New York, 1981. Sprent, J., Cell. Immunol. 7, 10, 1973. Woodruff, J. J., and Gesner, B. M., J. Exp. Med. 129, 551, 1969. Woodruff, J. J., and Woodruff, J. F., Cell. Immunol. 10, 78, 1974. Freitas, A. A., and de Sousa, M., Cell. Immunol. 22, 345, 1876. Butcher, E. C., and Weissman, I. L., Immunol. Methods 37, 97, 1980. Rittenberg, M. B., and Pratt, C., Proc. Sot. Exp. Biol. Med. 132, 575, 1969. Cunningham, A. J., and Szenberg, A., Immunology 14, 599, 1968. Smith, H. R., Chused, T. M., and Steinberg, A. D., J. Zmmunol. 131, 1257, 1983. Scher, I., Steinberg, A. D., Beming, A. K., and Paul, W. E., J. Exp. Med. 142, 637, 1975. Sprent, J., Bruce, J., Ron, Y., and Webb, S. R., J. Immunol. 134, 1442, 1985. Stamper, H. B., Jr., and Woodruff, J. J., J. Exp. Med. 144, 828, 1976. Chin, Y-.H., Carey, G. D., and Woodruff, J. J., J. Zmmunol. 129, 1911, 1982.
MIGRATION
OF xid AND non-xid SPLENIC
B CELLS
393
27. Butcher, E. C., Scollay, R. G., and Weissman, I. L.. J. Immunol. 123, 1996, 1979. 28. Stevens, S. K., Weissman, I. L., and Butcher, E. C., J. Immunol. 128, 844, 1982. 29. Smathers, P A., Steinberg, B. J., Reeves, J. P., and Steinberg, A. D., J. Immunol. 128, 1414, 1982. 30. Remold-O’Donnell. E., Kenney. D. M., Parkman, R., Cairns, L., Savage. B., and Rosen, F. S., J. hp. Med. 159, 1705, 1984. 31. Gallatin, W. M., Weissman, I. L.. and Butcher, E. C., Nature (London) 304, 30, 1983. 32. Cohen, D. I., Steinberg, A. D., Paul, W. E., and Davis, M. M., Nature (lo&on) 314, 369, 1985. 33. Cohen, D. I., Hedrick, S. M., Nielsen, E. A., D’Eustachio, P.. Ruddle, F., Steinberg, A. D., Paul, W. E., and Davis, M. M. Nature (London) 314, 372, 1985. 34. Smith, H. R., Yaffe, L. J., Kastner, D. L., and Steinberg, A. D., J. Immunol. 136, 1194, 1986. Received August 16. 1985; accepted with revision January 9. 1986
IMMUNOLOGY
Splenic
AND
IMMUNOPATHOLOGY
Migration
of xid
39, 379-393 (1986)
and Non-xid
Splenic
HOWARD R. SMITH,NANCYLITTELL,ANDALFRED Cellular
Immunology Diabetes,
Section, Arthritis and Rheumatism and Digestir,e and Kidney Diseases, Bethesda, Maryland
Branch, National 20205
B Cells
D. STEINBERG National Institates
Institute of ArthritiJ, of Health,
The X-linked immune deficiency gene (xid) has been viewed as inducing either a deficiency in a B-cell subset or in B-cell maturation. The present experiments were performed in an attempt to better understand whether (a) xid B cells migrate differently from mature and immature non-xid B cells, and (b) whether the failure of mature B cells to be found in xid spleen, but not Peyer’s patches, is secondary to migratory differences, especially the possible inability of xid spleens to receive mature B cells. We employed the technique of internally fluorescein labeling donor cells and subsequent injection into recipients. Double labeling permitted analysis of B or T cells by two-color flow microfluorometry. Functional studies revealed that labeled cells appropriately responded to TI-1 and TI-2 antigens after migration. We found that (a) adult xid splenic B cells do not migrate to spleen as well as adult non-xid splenic B cells, (b) the migration of adult xid splenic B cells to spleen resembles that of neonatal (xid and non-lid) B cells, (c) equal masses of xid and non-xid spleens have an equal capacity to receive either xid or non-xid splenic B cells, and (d) the migration of xid and non-xid T cells is similar. These results suggest that xid B cells do not migrate normally. but that migratory differences cannot account for the failure of mature B cells to be found in xid spleens. 0 1986 Academic Press. Inc
INTRODUCTION
The X-linked immune deficiency gene (xid)’ has provided a unique opportunity to examine B lymphocytes. Comparisons of B lymphocytes from normal mice to those bearing the xid gene have furthered our understanding of B-cell development, maturation, heterogeneity, and activation requirements. Although the functional consequences of xid have been studied in detail, the underlying reasons for the observed differences continue to be an unsettled issue. It is generally thought that the presence of xid results in either (a) the absence of a B-cell lineage, or (b) the relative failure of B-cell maturation (l-3). Thus, xid can be viewed as a genetic marker for either B-cell subsets or maturation (l-3). The precise nature of the xid defect has yet to be fully elucidated. Although xid B cells have many similarities to both Lyb 5- B cells and neonatal B cells (4, 5), controversy exists as to whether xid B cells are equivalent to the Lyb 5- B cells found in non-xid mice. Recent investigations have shown xid B cells to have both phenotypic (6) and functional (7- 11) properties which have no known normal counterparts. Fur-
’ Abbreviations used: BA, Bruce/la abortus; FITC, fluorescein isothiocyanate; Hepes, N-2 hydroxyethylpiperazine-N’-2-ethanesulfonic acid; HEV, high endothelial venules; LPS, lipopolysaccharide; PFC, plaque-forming cells; SRBC, sheep red blood cells; TNP, trinitrophenyl; xid, X-linked immune deficiency gene; XLR, X-linked gene family. 379 0090-1229/86 $1.50 Copyright All rights
Q 1986 by Academic Press. Inc. of reproduction in any form reserved.
380
SMITH, -LITTELL,
AND STEINBERG
thermore, it appears that xid mice do not have a total arrest in maturation of B cells: in selected lymphoid sites (e.g., Peyer’s patches), young xid mice have mature cell surface antigens (12) and are responsive to TI-2 antigens (13). The experiments presented herein were performed in an attempt to better understand whether xid B cells migrate differently from mature and immature nonxid B cells and whether the failure of mature B cells to be found in xid spleen is secondary to migratory differences. Because the migratory characteristics of cells can be dramatically affected by alterations of surface membranes (14-18), the technique, as developed by Weissman, of internally fluorescein labeling donor cells was employed (19). This technique, which does not alter lymphocyte migration (19), permits analysis of lymphocyte surface membrane phenotypes after migration has occurred. Our results indicate that (a) the migration of xid splenic B cells to spleen differs from that of non-xid splenic B cells, (b) the migration of adult xid splenic B cells to spleen resembles that of neonatal (xid and non-xid) B cells, and (c) equal masses of xid and non-xid spleens have an equal capacity to receive either xid or non-xid splenic B cells. These results suggest that migratory differences exist in xid B cells but they do not account for the failure of mature B cells to be present in xid spleen. MATERIAL
AND METHODS
Mice. Both adult (3 months) and neonatal (7 to 10 days) male and female DBA/2N, DBA/2N.xid (D2.xid), (CBA/N x DBA/2N)F,, and (DBA/2N x CBA/N)F, mice were obtained from our own facilities. All mice were individually ear tagged, housed in our animal facility, and allowed free access to food and water. Antigens. TNP,,-AECM-Ficoll was purchased from Biosearch (San Rafael, Calif.). TNP-BA was kindly provided by Dr. J. Mond, Uniformed Services University of the Health Sciences (Bethesda, Md.). TNP-LPS (butenol extracted) was purchased from Sigma Chemical Company (St. Louis, MO.). Cell cultures. Spleen cells from mice killed by cervical dislocation were prepared by gentle teasing and the cells were washed in balanced salt solution. The standard tissue culture medium consisted of RPM1 1640 supplemented with 10% bovine calf serum, 2 mM L-glutamine, 10 mM Hepes, nonessential amino acids, sodium pyruvate, 5 x lo-’ M 2-mercaptoethanol, and penicillin-streptomycin. Washed cells (250 p,l of 4 x lo-‘j/ml) were cultured in 96-well Costar 3596 microtiter plates (Cooke Co., Alexandria, Va.) in a humidified 5% CO,-95% air incubator at 37°C for 4 days. Appropriate wells received optimal doses of TNP-Ficoll (50 &ml), TNP-BA (1: IO5 of stock), or TNP-LPS (10 pg/ml) and were cultured for 4 days. Plaque-forming cell assay. Sheep red blood cells were conjugated with TNP by the method of Rittenberg and Pratt (20). Direct anti-TNP PFC were enumerated by the Cunningham slide modification of the Jerne hemolytic plaque assay (21). Background PFC to TNP were subtracted. Standard errors were less than 20% of the means. Cell labeling. Cells were labeled as described by Butcher and Weissman (19). A stock solution of fluorescein isothiocyanate (FITC; Sigma) was prepared and fil-
MIGRATION
OF xid AND
non-d
SPLENIC
B CELLS
381
tered. The concentration was determined spectrophotometrically. Spleen cells pooled from at least four mice were prepared as described above, washed in RPM1 1640, and resuspended at 2.5-5.0 x 10’ cells/ml in medium (adjusted to pH 7.0) containing equal parts of Medium 199 and phosphate-buffered saline (Biofluids, Rockville, Md.), and 5% bovine calf serum (Hyclone, Logan, Utah). Cells were labeled with FITC (40 pg/ml) at 37°C for 30 min. Cells were washed with cold RPM1 1640, centrifuged, resuspended in RPM1 1640, layered over 5 ml (Falcon tube 2095, Oxnard, Calif.) of bovine calf serum, and centrifuged. Cell migration. Fluorescein labeled cells were resuspended at 100 x lo6 cells/ ml and 0.5 ml was injected intravenously into at least four recipient mice. After 2 hr, unless otherwise indicated, recipient mice were sacrificed. Single cell suspensions of spleen, liver, or lung were prepared in RPM1 1640 medium by gentle teasing with forceps or by placing the organ between two glass slides and applying gentle pressure, followed by passage through a 25-gauge needle. Cells were allowed to settle for 5 min to remove clumps. Fluorescence-positive cells thus represented cells of donor origin. The percentage of migrating donor cells could be expressed either as a percentage of injected donor cells which were recovered in a particular organ (e.g., spleen, liver, lung) or as percentage of all of the cells in an individual organ which was from the donor. Significant differences between sample groups were determined using Student’s t test. To further analyze how many donor B cells or T cells had migrated to recipient spleens, single cell suspensions of recipient organs were stained with a second antibody (red labeled) to IgM or Thy 1. For example, if the second stain was anti-IgM, four populations of cells were possible, namely (a) those which had no color and were recipient non-B cells, (b) those which were red only and were recipient B cells, (c) those which were green only and were donor non-B cells, and (d) those which were green and red and were donor B cells. Cell staining andflow cytometry. Single cell suspensions at lo6 cells/ml of recipient organs were prepared as described above, washed, and resuspended in sorter buffer (phosphate-buffered saline containing 0.5% bovine serum albumen and 0.05% sodium azide). Cells were incubated with previously titrated biotinylated monoclonal antibodies for 20 min at 4”C, washed, and resuspended in buffer. Previously titered Texas red avidin or FITC-avidin was incubated with the cells for 20 min at 4°C. A biotinylated monoclonal antibody to Thy 1.2 was purchased from Becton-Dickinson (Sunnyvale, Calif.). Biotinylated F(ab’), goat anti-mouse IgM was purchased from Zymed Laboratories (San Francisco, Calif.). Cells were washed and resuspended in 0.5 ml sorter buffer. Flow cytometry was performed on a Becton-Dickinson FACS-II equipped with an argon laser emitting 200 mW at 488 nm and a krypton ion laser emitting 150 mW at 568 nm. Fluorescence data were collected with logarithmic amplification. For each sample, data from 50,000 to 100,000 light scatter-gated viable cells were collected. Data were analyzed with use of computer programs as previously described (22). RESULTS
Differences in migration
of xid and non-xid spleen cells in syngeneic recipients.
SMITH,
382
LITTELL,
AND STEINBERG
Initial studies suggested that differences existed in the migratory ability of xid and non-xid spleen cells. In order to better assess these differences, the short-term kinetics of migration were studied. Single cell suspensions of FITC-labeled spleen cells from xid (CBA/N x DBA/2)F, male and non-xid (DBA/2 x CBA/N)F, male mice were injected into syngeneic mice. At various times after injection, recipient spleens were obtained and the percentage of FITC-positive donor cells which had migrated to the recipient spleen was determined by flow microfluorometry. Multiple experiments demonstrated that the percentage of syngeneic FITC-labeled spleen cells migrating to the spleen tended to slowly decrease with time and was consistently greater in non-xid (DBAI2 x CBA/N)F, mice than xid (CBA/N x DBA/2)F, mice (Fig. 1). Similar results were obtained using congenic DBA/2 and D2.xid mice (results not shown). It was necessary to ensure that the migratory cells were actually located within the spleen as opposed to merely residing within the blood vessels of the spleen. At the various times as indicated in Fig. 1, the percentages of internally FITC-labeled donor spleen cells in peripheral blood were consistently ~0.2%. In addition, histologic examination of frozen sections of the spleen with examination for the placement of fluorescein-labeled cells revealed that the migrating cells were within the lymphoid areas of the spleen. Sections of recipient liver and lung were studied at the 2-hr postinjection time point for nonspecific trapping of donor cells. There was no difference in trapping between donor xid and donor non-xid cells. Fluorescein-labeled xid (CBA/N x DBA/2)F, male, or non-xid (DBA/2 x CBA/N)F, male, spleen cells were intravenously injected into both xid and non-xid recipients. There were no significant differences in the percentage of xid and non-xid spleen cells which lodged in the liver or lung in either xid or non-xid recipients. Specifically, the percentages of fluorescein-positive xid donor cells that were recovered in the livers of xid and non-xid recipients were 27.4 and 28.8%, respectively. The percentages of fluores-
10,
I
@BAD
X CBAlNlF,b
ICBAiN X DBAOIF,
I 2
6 HOURS
I 12 AFTER
6
I 18
I 24
INJECTION
FIG. 1. Kinetics of migration to spleen of syngeneic FITC-labeled spleen cells. spleens from (CBA/ N x DBA/2)F,, xid/Y, and (DBA/2 x CBA/N)F,, +A’, were made into single cell suspensions, FITC labeled, and 5 x 10’ injected into each of a large number of syngeneic mice. At various times after injection, several recipient spleens were obtained and the percentage of FITC-labeled cells which had migrated to each recipient spleen was determined by flow microfluorometry. The percentage of syngeneic FITC-labeled spleen cells migrating to the spleen was consistently greater in non-xid (DBA/2 x CBA/N)F, mice than in xid (CBA/N x DBA/2)FI mice.
MIGRATION
383
OF xid AND non-xid SPLENIC B CELLS
cein positive non-xid donor cells recovered in the livers of xid and non-x-id recipients were 29.7 and 2.5.2%, respectively. An equal percentage (10.0%) of labeled donor xid spleen cells was recovered in the lungs of either xid or non-xid recipients. Similarly, 11 .O and 12.8% of labeled non-xid spleen cells were recovered in xid and non-xid lungs, respectively. Thus, the nonspecific trapping of xid and non-xid spleen cells in nonlymphoid organs was similar. More importantly, donor cells which migrated to the spleen were found to localize within the lymphoid areas of the spleen. Ability of FITC-labeled spleen cells to respond to antigen. Previous studies have revealed that the inability of spleen cells from xid mice to respond to type 2 antigens (e.g., TNP-Ficoll) can be corrected after injection of histocompatible non-xid spleen cells (23) and that this response to type 2 antigens is solely attributable to the donor non-xid cells (23). The inability of xid cells to respond to TNP-Ficoll was utilized in studying spleen cell function after FITC labeling. Either non-xid or xid donor spleen cells were FITC labeled and intravenously injected into xid recipients. After 2 hr, the recipient spleens were removed, made into single cell suspensions, and cultured with various antigens. The response to antigen was determined at Day 4 by enumeration of plaque-forming cells. Spleen cell cultures from xid mice which had been injected with FITC-labeled non-xid spleen cells responded to TNP-Ficoll, whereas those receiving FITC-labeled xid spleen cells failed to respond (Table 1). Thus, a function solely attributable to the non-xid donor spleen cells, namely the response to TNP-Ficoll, remained intact. The responses to the type 1 antigens TNP-BA and TNP-LPS were present in both groups and are attributable to both recipient (xid) and donor (either xid or non-xid) spleen cells. These data indicate that donor spleen cells retain their functional ability to respond to antigen after FITC labeling, in vivo injection, and migration to spleen. Migration of FITC-labeled xid and non-xid spleen cells to xid and non-xid spleen. The percentage of fluorescence-positive spleen cells recovered in spleens
after injection of syngeneic FITC-labeled spleen cells is greater in non-xid mice than in xid mice (Fig. 1). In an attempt to determine whether these observed differences were due to differences in (a) inherent migratory abilities of spleen cells and/or (b) the capability of spleens to receive such cells, both non-xid and TABLE ABILITY
OF INTERNALLY FITC-LABELED
Donor”
+ IY’ xidlY
Recipient xidlY xidiY
1
NON-xid
DONOR SPLEEN CELLS TO IN RECIPIENT xid MICE
TNP-Ficollb 135 5
TNP-BAb 195 88
RESFQNDTO TNP-FXOLL TNP-LPSb 95 44
LIDonor cells were FITC labeled and injected into recipients. After 2 hr, recipient spleens were removed, made into single cell suspensions, and cultured with or without the indicated antigen. b Numbers of plaque-forming cells/106 spleen cells, to TNP-SRBC on Day 4. Standard errors were ~20%. No antigen (“background”) gave ~5 PFC/106 spleen cells. ‘xid/Y = (CBA/N x DBA/2)F,; +iY = (DBAIZ x CBA/N)F,.
SMITH,
384
LITTELL,
AND
STEINBERG
xid mice were injected with FITC-labeled spleen cells from either non-xid or xid mice. Experiments representative of three similar ones are shown. Donor spleen cells were determined by flow microfluorometry. The percentage of donor xid spleen cells migrating to non-xid (+/Y) spleens was 17.0 + 1.6% and to xid spleens was 10.2 + 2.0% (Table 2). The percentage of donor non-xid spleen cells migrating to non-xid (+N) spleens was 26.5 r 1.0% and to xid spleens was 16.5 & 1.O% (Table 2). This decreased absolute capacity of xid spleens as compared to non-xid spleens to receive transferred cells does not account for the smaller size of the xid spleen. To correct for differences in the spleen size, these data can be expressed as a ratio of donor xid cells to donor non-xid cells migrating to recipient spleen and then the ratios compared. For non-xid recipients, the ratio of the percentage of migrating donor xid to donor non-xid was 0.64 (17.0/26.5%) (Table 2). Similarly, for xid recipients, this ratio of the percentage of migrating donor xid to donor non-xid cells was 0.62 (10.2/16.5%) (Table 2). Furthermore, not calculated in Table 2, the ratio of the percentage of donor xid spleen cell migrating to recipient xid and non-xid spleens was 0.60 (10.2/17.0%). The ratio of the percentage of donor non-xid spleen cell migration to recipient xid and non-xid spleens was 0.66 (16.5/26.5%). Thus, as a function of either donor migration, or recipient capacity, these ratios remain remarkably similar. These data indicate that (a) there is a decreased absolute capacity of xid spleens to receive transferred cells, (b) when one accounts for the differences in spleen size, equal masses of xid and non-xid spleens have similar capacities to receive spleen cells, and (c) the migration to spleen of non-xid spleen cells exceeds that of xid spleen cells. TABLE
2
MIGRATION OF FITC-LABELED xid AND NON-xid SPLEEN CELLS TO SPLEENS OF xid AND NON-xid RECIPIENTS: EQUAL RELATIVE CAPACITY OF + N AND xidN SPLEENSTO RECEIVE DONOR CELLS
DonoF xidNc
Recipient l kN’
Percentage of donor spleen cells migrating to recipient spleen (absolute capacity)
Ratiob (relative capacity)
17.0 + 1.6d 0.64’
f l.Of 10.2 * 2.0
+N xidN
+N xidlY
26.5
+N
xidlY
16.5 f 1.0
0.62 g
a Donor spleen cells were internally FITC labeled and injected into four recipients. The percentage of injected donor spleen cells migrating to the spleen was determined at 2 hr by flow microfluorometry. Recipient spleen ceil autofluorescence (with or without injected non-FITC-labeled spleen cells) was
MIGRATION
OF xid AND non-d
SPLENIC
B CELLS
385
We further analyzed the percentage of cells in an individual recipient organ which is of donor cell origin. The percentage of spleen cells that was fluorescence positive was determined by flow microfluorometry. After injection with FITC-labeled donor xid spleen cells, the percentages of non-xid and xid recipient spleen cells that were fluorescence positive and thus of donor cell origin, were 5.7 and 5.9, respectively (Table 3). In contrast, following injection of FITC-labeled donor non-xid spleen cells, the percentages of non-xid and xid spleen cells that were fluorescence positive were greater, 9.2 and 11.8, respectively (Table 3). Thus, the xid spleens were able to receive +/Y cells as well as -t/Y spleens could receive + /Y cells. Migration of neonatal spleen cells to adult spleen. Many similarities exist between the B cells of xid mice and those of neonatal non-xid mice (4, 5) and may indicate that the xid defect reflects a failure in B-cell maturation. Recently, however, significant differences have been found between xid and neonatal non-xid B cells (6- 11) suggesting that xid B cells may be intrinsically abnormal and have no normal counterpart (6, 8, 11). Because donor xid cells differ from non-xid cells in their migration, it would be of interest to compare xid and neonatal non-xid spleen cell migration. Neonatal donor spleen cells were FITC labeled and injected into adult recipients. After 2 hr, the percentage of fluorescence-positive cells in recipient spleens was determined. It was observed that migration of neonatal xid spleen cells to spleen did not significantly differ from that of neonatal non-xid spleen cells (Table 4). However, the migration to spleen of both neonatal xid and neonatal non-xid spleen cells was significantly less than that of adult nonxid spleen cells (Table 4). Adult xid spleen cell migration did not differ from neonatal xid spleen cell migration (Table 4). Thus, the migration to spleen of xid spleen cells closely resembles that of neonatal but not adult non-xid spleen cells.
TABLE 3 MIGRATION OF FITC-LABELED xid AND NON-xid SPLEEN CELLS TO SPLEENSOF xid AND NON-xid RECIPIENTS: EQUAL ABILITY OF + IY CELLS TO MIGRATE TO xidlY AND + IY RECIPIENT SPLEENS
Donor”
Recipient
xidlYb xidlY
+lYb xidlY +lY lid/Y
i-/Y +/y
Percentage fluorescencepositive spleen cells in recipient spleen 5.7 k 0.8’ 5.9 2 1.0 9.2 IT 0.5d
11.8 2 0.4
n Donor spleen cells were internally FITC labeled and injected into four recipients. The percentage of cells in the recipient spleen which were of donor cell origin and were fluorescence positive was determined at 2 hr by flow microfluorometry. Recipient spleen cell autofluorescence (with or without injected non-FITC-labeled spleen cells) was < 1.O%. bxidh’ = (CBA/N x DBAlZ)F,; +lY = (DBAR x CBAlN)F,. c Not statistically different from (xidlY into xidly), P > 0.05; statistically different from (+ IY into +/Y), P < 0.01, Student’s f test. d Not statistically different from (+lY into xidlY). p > 0.05; statistically different from (xid/Y into xidlY), P < 0.05, Student’s t test.
386
SMITH,
LITTELL,
MIGRATION OF xid AND NON-xid
AND STEINBERG
TABLE 4 NEONATAL SPLEEN CELLS TO ADULT SPLEEN Percentage FITC-positive spleen cells in recipient”
Donor xidNb
+/y xidlY +N
Age
Adult recipient
Neonate Neonate Adult Adult
xidlY xidlY xidN xidN
Expt 1 5.6 4.2 5.3 10.3
f f * f
Expt 2
0.7’ 0.5 0.4’ 0.7e
4.8 ?I 0.8” 6.0 -r- 0.3 NDd NDd
u Donor spleen cells were (internally) FITC labeled and injected into four adult recipients. The percentage of fluorescence-positive donor spleen cells migrating to recipient spleen was determined at 2 hr. Recipient spleen cells autofluorescence was < 1.O%. b xidN = (CBA/N x DBA/Z)F, and +/Y = (DBA/2 x CBA/N)F,. c Not significantly different from neonatal + N donor, P > 0.05, Student’s t test. d Not determined. e Significantly different from all other groups, P < 0.01, Student’s 1 test.
The xid migratory defect maps to the B cell. Analysis of the migratory abilities of populations of B cells and T cells from unfractionated spleens was performed by means of two-color flow microfluorometry. Donor spleen cells were internally labeled with FITC and injected into syngeneic recipients. After 2 hr, recipient spleens were removed, the cell surface membranes stained with biotinylated antiIgM or biotinylated anti-Thy 1.2, and Texas-Red avidin was added. The percentage of two-color-positive (green and red) spleen cells was determined. The percentage of total donor spleen cells migrating to spleen was greater in (DBA/2 X CBA/N)Ft +/Y mice (non-xi4 (10.3%) than in (CBA/N x DBA/2)F, xid/Y mice (7.4%) (Table 5). This difference in total donor cells migrating to spleen was TABLE 5 MIGRATION OF SPLENIC B CELLS, T CELLS, AND NON-B, NON-T CELLS FROM xid AND NON-xid SYNGENEICSPLEEN
MICE TO
Double-labeled cells” as percentage of recipient spleen cells (DBA/2 x CBA/N)F, (-t/Y) B cells T cells Non-B/non-T cells (by subtraction) Total cells
5.2 f 0.7b 3.1 2 0.56 2.0 10.3 f 1.4
(CBA/N x DBA/2)F, (xid/Y)
2.1 * 0.2 3.6 k 0.6 1.7 7.4 f 1.1
D Donor spleen cells were (internally) FITC labeled (green) and injected into four syngenic recipients. Recipient spleen cells were removed after 2 hr, the surface membranes stained with either biotinylated anti-IgM or biotinylated anti-Thy 1.2, and Texas red avidin was added (red). Two-color positive (green and red) spleen cells were determined as a percentage of recipient spleen cells. b Statistically different from xid/Y recipients, P < 0.05, Student’s t test. c Not statistically different from xidN recipients, P > 0.05, Student’s f test.
MIGRATION
OF xid AND non-xid
SPLENIC
387
B CELLS
largely due to the increased percentage of migrating non-xid B cells (5.2%) compared to migrating xid B cells (2.1%) (Table 5). Recipients of non-xid and xid spleen cells had a similar percentage of two-color-positive donor T cells (3.1 and 3.6%, respectively) and donor non-B, non-T cells (2.0 and 1.7%, respectively) (Table 5). This result was confirmed in repeated experiments. Furthermore, virtually no differences were discernable in the staining profile of donor T-cell fluorescence intensity between non-xid and xid mice (Fig. 2). Thus, all differences in migration between xid and non-xid spleen cells can be attributed to differences in B-cell populations in the respective donors. The xid B-ceil migratory defect does not result from xid recipient abnormalities. To further evaluate if the observed differences in B-cell migration was secondary to (a) an inability or decreased capacity of xid spleens to receive certain
populations
of B cells (e.g., dull IgM+)
; ‘~~~
03A/ZxCBAIN)F,
-
ICBA/NxDBA/21F,
or (b) a deficit in the migratory
K33A/NxDBA/21F, XidiY
1Ol I ldo FLUORESCENCE INTENSITY FITC
-
ability of
IDBAIZxCBAINIF, +/y
10 (green)
FIG. 2. T-cell migration to spleen of FITC-labeled spleen cells. Donor spleen cells from either (DBA/Z x CBA/N)F,, +/Y, (non-xi&, or (CBA/N x DBA/2)F,, xid/Y, (xid) mice were intkmally labeled with FITC (green). These cells were injected into either (DBA12 x CBA/N)F,, +/Y, or (CBA/N x DBA/2)F,, xid/Y, mice. Recipient spleens were removed after 2 hr, made into single cell suspensions, and the cell surface stained with anti-Thy 1.2 (Texas red). Each axis represents increasing intensity of fluorescence on a log scale (X axis = FITC; Y axis = anti-Thy 1.2). The percentages and profiles of double-labeled (migrating) donor splenic T cells were similar in all groups.
388
SMITH,
LITTELL,
AND
STEINBERG
donor xid spleen cells, advantage was taken of the characteristic IgM profiles of non-xid (both dull and bright IgM+) and xid (predominantly bright IgM+) spleen cells (5). Donor spleen cells were FITC labeled and injected into recipients. After 2 hr, recipient spleens were removed, made into single cell suspensions, and the cell surface stained with anti-IgM (Texas red). Donor IgM+ spleen cells appear as two-color positive (green and red) and are therefore distinguishable from recipient spleen cells (not green) or donor non-IgM+ cells (green only). By computer analysis, the independent IgM (red) fluorescence profiles of recipient (green negative) and donor (green positive) spleen cells ware obtainable (Fig. 3). Non-xid mice receiving either non-xid or xid spleen cells demonstrated the capacity to receive both bright and dull IgM+ cells as evidenced by the characteristic donor IgM profiles (Fig. 3C). Similarly, xid recipients had the capacity to receive both bright and dull IgM+ spleen cells (Fig. 3D). Thus, both non-xid and xid recipients have the capacity to receive non-xid and xid splenic B cells. Therefore, the reduced migration to spleen of splenic xid B cells does not result from an inability of the xid spleen to receive such cells, but from an inherent migratory defect of xid B cells. Mixing experiments. In order to correct for the presence in non-xid spleens of both more B cells and more “normal” B cells (3, 5), mixing experiments were performed. Either xid or non-xid FITC-labeled spleen cells were mixed in various percentages with unlabeled spleen cells of the opposite kind. After injection into xid recipients, cells were harvested and labeled with anti-surface IgM (red). The percentage of FITC-labeled (green) donor splenic B cells migrating to recipient spleen was determined correcting for differences in donor B-cell numbers. Comparison of the slope of the lines of non-xid and xid B-cell migration reveals that non-xid B cells have a greater ability to migrate to spleen than do xid B cells (Fig. 4). Moreover, the straight lines indicate that there is minimal, if any, interference by one population of B cells with the migration of the other. Thus, inherent differences exist in the migratory capacity of non-xid and xid splenic B cells. DISCUSSION
In the present study, we have presented evidence for differences in the migratory ability of B cells, but not T cells, from xid and non-xid mice. The migration to spleen of adult xid B cells resembles that of neonatal B cells from both xid and non-xid mice. Furthermore, these observed differences in B-cell migration are not due to differences in the capacity of equal masses of xid and non-xid spleens to receive such cells, but rather, appear to be due to intrinsic differences in the B cells. Using a different approach, Sprent and colleagues have recently made similar observations (24). Many factors are known to be important in the migration of lymphocytes, (reviewed in (14)) however, little is known about the precise mechanisms regulating lymphocyte migration to the spleen. Spleen does not have the postcapillary high endothelial venules (HEV) through which lymphocytes migrate in order to enter and gain access to lymph nodes and Peyer’s patches (25-27). Indeed, the mechanism of migration to spleen differs from the HEV-mediated migration of lymph node and Peyer’s patches as demonstrated by the dramatic decrease in lymphocyte migration to lymph node and Peyer’s patches but not to spleen after trypsin
MIGRATION
OF rid AND non-xid SPLENIC
1.0,
B CELLS
389
A
0.8 1
Donor IDEA/2 x CBA/NIF, +/y
0.61
0.8
;*.
Donor ICBNN
x DBAl21F, ad/Y
:‘: I :
0.6 d
0.4
$I a 3
0.2
5 0
l.Or
:.
/
',, '.. -._ .
.
_ ,,.,,
'.,"'
C (DBAIP x CBAlNlF,-IDBAIP x CBA’NIF, +lY +/y ICBAIN x DBA/2lF, - IDBAl2 x CWN)F, id/Y +/y
i (DBAl2
x CBAfN)F,
FLUORESCENCE
-
(CBA/N x DB&‘PIF, xtd/Y
INTENSITY ANTI-IgM
,-I , ,
,-,
(RED)
FIG. 3. B-cell migration of donor spleen cells; Anti-IgM fluorescence profile. Donor spleen cells were FITC labeled (internally) and injected into recipients. Recipient spleens were removed after 2 hr, made into single cell suspensions, and the cell surface stained with anti-IgM (Texas red). The characteristic splenic IgM profiles of non-xid (A) (both dull and bright IgM+) and xid (B) (relative reduction in dull IgM+) spleen cells were observed. Cells that had migrated to the spleen were selected on the basis of green fluorescence. These were then further analyzed with regard to IgM profiles (red). The IgM (red) profiles of two-color positive (green and red) donor splenic B cells migrating to recipient spleens are shown (C and D). The data indicate that (i) both non-xid and xid recipients have the capacity to receive both non-xid (-) and xid ( . . . . ) spleen cells. and (ii) the donor cells retain their characteristic IgM profiles despite migration to recipient spleens.
treatment of lymphocytes (16, 17). Although the structures which regulate the uptake into spleen of migratory lymphocytes are unknown, B cells do migrate better to spleen (and Peyer’s patches) than do T cells (28). The present studies were performed to determine whether or not the spleen of xid mice might fail to
390
SMITH,
PERCENT
LITTELL,
AND STEINBERG
OF INDICATED
SPLEEN CELLS INJECTED
FIG. 4. Percentage of donor splenic B cells migrating to recipient spleens. Either xid or non-xid (+/Y) internally FITC-labeled spleen cells (green) were mixed in the percentages indicated with unlabeled non-xid or xid spleen cells, respectively. Thus, on the abscissa the 25% point on the xid/Y line represents a cell mixture in which 25% of the cells were FITC-labeled xid cells and 75% were unlabeled +/Y cells. The 25% point on the +/Y line represents a mixture in which 25% of the cells were FITC-labeled + N cells and 75% were unlabeled xid/Y cells. After injection into xid recipients, cells were harvested and labeled with anti-surface IgM with a second color (red) as in Fig. 3. The percentage of two-color-labeled donor splenic IgM+ B cells which had migrated to the recipient spleens is shown. A greater percentage of non-xid splenic B cells migrates to xid spleen relative to xid splenic B cells. There is no interference of xid B-cell migration by non-xid B cells or vice ver~a as indicated by the straight line.
receive mature B cells and thereby explain the failure of mature B cells to be found in their spleen. Such a result could come about by a failure of the xid spleen to have a receptor to which mature B cells might bind. We found, herein, that the xid spleen has no defect in receiving mature B cells. Thus, the finding of mature B cells in Peyer’s patches, but not spleen, of xid mice cannot be explained by an inability of xid spleen to receive migrating mature B cells. More likely, the local immune stimulation in Peyer’s patches accounts for the finding of mature B cells in xid Peyer’s patches. This formulation is further supported by the ability of polyclonal stimulation of xid mice to induce mature B-cell functions (29). The other possibility is that the xid B cells lack a receptor by which those cells could grow and mature in the spleen. Evidence was found that xid cells did not migrate as well as non-xid B cells to the non-xid spleen; however, there was suffrcient migration of xid B cells to non-xid spleen to render unlikely the possibility
MIGRATION
OF xid AND
non-d
SPLENIC
B CELLS
391
that a migration-specific cell surface receptor was sufficient to explain the extreme lack of mature B cells in the spleens of xid mice. Whereas evidence has been obtained for a defect in a cell surface glycoprotein in Wiskott-Aldrich syndrome (30) and could lead to such a migration defect, we do not believe that the evidence from our migration studies of xid cells points to such a defect. In the present study, spleen cell suspensions were analyzed for the presence of fluorescein-tagged cells. The utility of the method depends upon the stability of the marker which has been well demonstrated and which we have confirmed. This method avoids the problems of such markers as SICr which may leach out of one cell and be picked up by another. The study also requires identification in recipient spleens of subpopulations of migrating donor cells. One approach could have been isolation of subpopulations of cells prior to injection, however, such isolation procedures might have not been complete, and, more importantly, might have damaged cell membrances so as to alter their migratory ability. Instead, unmanipulated cells were internally labeled and analyzed by surface label only after they had migrated to recipient spleen. This method allowed an easy identification of B cells and T cells and even preserved the characteristic subtle surface IgM profiles of donor B-cell subpopulations. Thus, we believe that the method was quite satisfactory for the present study. A potential problem in the present experiments concerns the location of lymphocytes within the spleen versus lymphocytes in the vascular spaces of the spleen. Our results depend upon the measurement of cells which have migrated to the spleen as opposed to merely residing in the blood vessels of the spleen. As explained in the results, the contribution of the blood to the fluorescein-tagged cells in the spleen was negligible. However, the splenic cells could be in the red pulp and merely be percolating through the spleen, or actually be part of the lymphoid structures of the spleen. Histologic examination of frozen sections of the spleens with examination for placement of fluorescein-tagged cells clearly demonstrated that the migrating cells were within the lymphoid areas of the spleen. Thus, the data demonstrate the migration of lymphocytes to the spleen rather than their mere presence within the vascular spaces or the red pulp. Even modest damage would leave lymphocytes in the red pulp; thus the latter finding is a further confirmation of the viability of the migrating cells. Furthermore, the possibility of an unrelated interaction of xid cells with other organs in viva was examined. Differential trapping of xid cells as compared to non-xid cells in nonlymphoid tissue, such as liver or lung, could account for the observed differences. Such was not the case: both xid and non-xid spleen cells were found equally in both liver and lung. Specific homing markers have been found for certain lymphocytes. The best studied are those which allow lymphocytes to specifically enter either peripheral lymph nodes, Peyer’s patches, or both (25-28). The receptor which allows lymphocytes to enter lymph nodes can be recognized by the MEL-14 reagent and binds to a specific complementary structure on high endothelial venules (31). Whether or not analogous specific receptor-ligand interactions are critical for migration to the spleen is unknown. In this regard, the xid and non-xid B cells differ by a number of determinants, at least quantitatively, including IgD, IgM,
392
SMITH,
LITTELL,
AND STEINBERG
Lyb 3, Lyb 5, and Lyb 7 (3-6); one or more of these markers may serve as a structure which facilitates migration to the spleen. Further studies of specificity of migration to the spleen may, in the future, shed light on this question. The present studies are consistent with the hypothesis that the major defect in xid mice is one of B-cell maturation. This possibility would explain the similar migration capabilities of young xid and young non-xid cells as well as young nonxid and older xid cells. It would also be consistent with the idea that the B cells of older xid mice do not have a normal counterpart (6, 8, 11): normal cells are able to complete the maturation process whereas xid B cells have a marked arrest in maturation. In addition, such a formulation is consistent with recent molecular genetic studies of xid mice which indicate that an X-linked gene family (XLR) is expressed only in the most mature B cells (32, 33), B cells which are markedly under-represented, especially in spleen, in xid mice (34). ACKNOWLEDGMENTS We express our thanks to Ms. Cindy Mizgala for her excellent technical assistance, Dr. Thomas Chused and Ms. Linette Edison for help with flow microfluorometry, and Ms. Betty Irene Roupe for preparation of the manuscript.
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
Kincade, P. W., Moore, M. A. S., Lee, G., and Paige, C. J., Cell. Immunol. 40, 294, 1978. DeFranco, A. L., Kung, J. T., and Paul, W. E., Zmmunol. Rev. 64, 161, 1982. Scher, I., Adv. Immunol. 33, 1, 1982. Ahmed, A., Scher, I., Sharrow, S. O., Smith, A. H., Paul, W. E., Sachs, D. H., and Sell, K. W., J. Exp. Med. 145, 101, 1977. Scher, I., Beming, A. K., Kessler, S., and Finkelman, F. D., J. Zmmunol. 125, 1686, 1980. Hardy, R. R., Hayakawa, K., Parks, D. R., and Herzenberg, L. A., Nature (London) 306, 270, 1983. Ono, S., Yaffe, L. J., Ryan, J. L., and Singer, A., J. Immunol. 130, 2014, 1983. Webb, S. R., Mosier, D. E., Wilson, D. B., and Sprent, J., J. Exp. Med. 160, 108, 1984. Wortis, H. H., Burkly, L., Hughes, D., Rochelle, S., and Waneck, G., J. Exp. Med. 155, 903, 1982. Mond, J. J., Scher, I., Cossman, J., Kessler, S., Mongini, P. K. A., Hansen, C., Finkelman, E D., and Paul, W. E., J. Exp. Med. 155, 924. 1982. Sprent, J., and Bruce, J., J. Exp. Med. 160, 711, 1984. Eldridge, J. H., Kiyono, H., Michalek, S. M., and McGhee, J. R., J. Exp. Med. 157, 789, 1983. Eldridge, J. H., Yaffe, L. J., Ryan, J. J., Kiyono, H., Scher, I., and, McGhee, J. R., J. Zmmunol. 133, 2308, 1984. de Sousa, M., “Lymphocyte Circulation: Experimental and Clinical Aspects,” Wiley, New York, 1981. Sprent, J., Cell. Immunol. 7, 10, 1973. Woodruff, J. J., and Gesner, B. M., J. Exp. Med. 129, 551, 1969. Woodruff, J. J., and Woodruff, J. F., Cell. Immunol. 10, 78, 1974. Freitas, A. A., and de Sousa, M., Cell. Immunol. 22, 345, 1876. Butcher, E. C., and Weissman, I. L., Immunol. Methods 37, 97, 1980. Rittenberg, M. B., and Pratt, C., Proc. Sot. Exp. Biol. Med. 132, 575, 1969. Cunningham, A. J., and Szenberg, A., Immunology 14, 599, 1968. Smith, H. R., Chused, T. M., and Steinberg, A. D., J. Zmmunol. 131, 1257, 1983. Scher, I., Steinberg, A. D., Beming, A. K., and Paul, W. E., J. Exp. Med. 142, 637, 1975. Sprent, J., Bruce, J., Ron, Y., and Webb, S. R., J. Immunol. 134, 1442, 1985. Stamper, H. B., Jr., and Woodruff, J. J., J. Exp. Med. 144, 828, 1976. Chin, Y-.H., Carey, G. D., and Woodruff, J. J., J. Zmmunol. 129, 1911, 1982.
MIGRATION
OF xid AND non-xid SPLENIC
B CELLS
393
27. Butcher, E. C., Scollay, R. G., and Weissman, I. L.. J. Immunol. 123, 1996, 1979. 28. Stevens, S. K., Weissman, I. L., and Butcher, E. C., J. Immunol. 128, 844, 1982. 29. Smathers, P A., Steinberg, B. J., Reeves, J. P., and Steinberg, A. D., J. Immunol. 128, 1414, 1982. 30. Remold-O’Donnell. E., Kenney. D. M., Parkman, R., Cairns, L., Savage. B., and Rosen, F. S., J. hp. Med. 159, 1705, 1984. 31. Gallatin, W. M., Weissman, I. L.. and Butcher, E. C., Nature (London) 304, 30, 1983. 32. Cohen, D. I., Steinberg, A. D., Paul, W. E., and Davis, M. M., Nature (lo&on) 314, 369, 1985. 33. Cohen, D. I., Hedrick, S. M., Nielsen, E. A., D’Eustachio, P.. Ruddle, F., Steinberg, A. D., Paul, W. E., and Davis, M. M. Nature (London) 314, 372, 1985. 34. Smith, H. R., Yaffe, L. J., Kastner, D. L., and Steinberg, A. D., J. Immunol. 136, 1194, 1986. Received August 16. 1985; accepted with revision January 9. 1986