INTERLEUKIN 2 PRODUCED BY ACTIVATED LYMPHOCYTES ACTS AS AN AUTOCRINE PROLIFERATION-INDUCING LYMPHOKINE Anand
Lagoo,
C. Kent Tseng,
Stewart
B
Sell*
Normal peripheral blood B cells produce a soluble factor after activation that is functionally indistinguishable from interleukin 2 (IL 2) and can support B cell proliferation in vitro. Purified rabbit peripheral blood B cells, when stimulated with a combination of ionomycin (0.5 pg/mL) and phorbol myristate acetate (PMA) (1 ng/mL), secreted a soluble factor in the culture medium that supported the IL 2-dependent cell line CTLL-2. The ability of these supernatants to support CTLL-2 growth was almost completely blocked by rabbit antibodies against human recombinant IL 2 and by the anti-IL 2 receptor monoclonal antibody 7D4. These data strongly suggest that the growth factor secreted by rabbit B cells is IL 2. To examine the possibility that the IL 2 activity detected in the B-cell cultures may be derived from residual T cells, B ceils were further purified by successive panning with a pan-T-cell monoclonal antibody, Lll-135, and goat anti-rabbit IgG. These highly purified B cells produced levels of IL 2 activity comparable to those produced by the initial B cell populations. Comparison of IL 2 production by decreasing numbers of purified T cells and purified B cells also indicated that the B cells were the source of IL 2 activity. Supernatants of activated B cells could support proliferation of B-cell blasts, and this activity could be completely absorbed by CTLL-2 cells, indicating that IL 2 is a major growth factor for B cells. Measurement of DNA synthesis by activated B cells in a limiting dilution assay suggested that T cells are not required for the induction of DNA synthesis and also that IL 2 produced by B cells can act as an autocrine growth factor. o 1990 by W.B. Saunders Company.
B cells from several species undergo proliferation follofiing activation with high doses of anti-immunoglobulin (anti-Ig) antibodies, apparently without the need for help from T cells.lm3On the other hand, it is also known that a significant increase in proliferation occurs upon addition of T-cell-derived lymphokines to anti-Igactivated B-cell cultures,4 and it is postulated by some investigators that T-cell-derived factors are essential for B-cell proliferation.5 We have recently demonstrated that additional activation signals provided by interleukin 2 (IL 2) in late G, are necessary for B cells to enter S phase.6 Thus it appears that even though B cells can proliferate in the absence of T cells, they cannot do so in the absence of lymphokines, particularly IL 2, raising the possibility of autocrine IL 2 production by activated
Department of Pathology and Laboratory Medicine, University of Texas Medical School, Houston, TX 77030. *To whom correspondence should be addressed at: Department of Pathology and Laboratory Medicine, University of Texas Medical School, P.O. Box 20708, Houston, TX 77225. o 1990 by W.B. Saunders Company. 1043-4666/90/0204-00$05.00/O KEY
272
WORDS:
Autocrine/B
Lymphocyte/IL
2
B cells. Earlier studies using Epstein Barr virustransformed B-cell lines have demonstrated the presence of an autocrine growth factor in culture supernatants of such cells after activation.7 However, it is not known whether normal peripheral blood B cells can produce autocrine growth factors that support their proliferation in the absence of T cells. IL 2 was initially described as a T-cell growth factor* produced by helper T cells.g It appears to play a major role in controlling B-cell proliferation as well. B cells have been reported to express both the ~55 (Taclike) and the ~70-75 components of the IL 2 receptor”,” and activated, but not resting, B cells respond to T-cell-derived IL 2 as well as to recombinant IL 2.12,13 Recent evidence indicates that certain B-cell lines and splenic B cells can be induced to secrete IL 2 activity after appropriate stimulation.‘4M16 On the basis of these observations it has been postulated, but never demonstrated experimentally, that B-cell-derived IL 2 may act as an autocrine growth factor.15 It is not known whether normal resting (Go) peripheral blood B cells can produce IL 2 or whether such autocrine IL 2 plays any role in controlling the proliferation of these cells. Although T-cell-derived IL 2 is known to act as a B-cell growth CYTOKINE,
Vol. 2, No. 4 (July),
1990: pp 272-279
IL 2 production
TABLE 1. Surface Ig staining and DNA synthesis by PBL and purified B cells. DNA Cell population
PBL B Cells
sIg + Cells”
Medium
47% 95%
144 + 14.2 104 zk 19.5
Synthesisb after activation
1,195 f 40.5 1,275 f 35.9
Con Ad
6,786 + 484.6 84 + 7.8
“Measured by flow cytometry following direct staining with FITC-labeled goat anti-rabbit IgG. bExpressed by cpm of ‘Z51UdR incorporation, mean k SD of triplicate cultures. ‘Affinity purified goat anti-rabbit Ig at 50 pg/mL. %oncanavalin A at 10 pg/mL.
Functional IL 2 Activity Supernatants
/ 273
in Activated
B-Cell
Supernatants from B cells and PBLs (1 x lo6 cells/ml in each case) activated with a combination of ionomycin (0.5 kg/mL) and phorbol myristate acetate (PMN (1 ng/mL) were harvested after 48 hr and tested in a standard IL 2 assay using the IL 2-dependent cell line CTLL-2. The response of these indicator cells to the lymphocyte culture supernatants was compared to their response to recombinant human IL 2 (Fig. 1). Rabbit PBLs and B cells produced a growth factor that could support the growth of CTLL-2, which was therefore presumed to represent IL 2 activity. By defining one unit of IL 2 activity as the amount of supernatant that induces half-maximal DNA synthesis in CTLL-2, it was observed that one million PBLs produced about 125 U/mL and one million B cells produced about 15 U/mL.
factor,” structural heterogeneity appears to exist between the T-cell-derived and B-cell-derived IL 2 due to differential glycosylation. l7 Therefore, it is necessary to investigate independently the ability of B-cell-derived IL 2 to function as a B-cell growth factor. Herein we provide evidence that highly purified rabbit peripheral blood B cells produce substantial quantities of IL 2 upon activation. The production of IL 2 appears to be limited to the first cell cycle after exiting Go and this IL 2 produced by B cells is capable of inducing DNA synthesis and supporting proliferation of activated B cells without the requirement of T cells in T-cell-depleted cultures. We discuss the possible role of IL 2 in B-cell proliferation and differentiation.
RESULTS Characterization
B cells
ing and DNA synthesis in response to anti-Ig and Con A of a representative cell preparation are shown in Table 1. In various experiments sIg+ cells in B cell preparations ranged from 95% to 98%. The B cells responded vigorously by DNA synthesis to anti-Ig but not to the T-cell mitogen, Con A. Thus these B-cell preparations were of high purity and depleted of T cells by functional and phenotypic criteria.
with
Anti-Ig’
by rabbit
Characterization of the Lymphokine B-Cell Supernatants
of B Cells
B cells purified from peripheral blood lymphocytes (PBLs) by treatment with anti-thymocyte serum and complement were tested by surface phenotype and mitogen response to determine the purity of the preparation. The results of surface immunoglobulin (sIg) stain-
Activity
in
To further characterize the growth-promoting activity produced by B cells, we tested the ability of a rabbit anti-human recombinant IL 2 antiserum to block the action of B cell supernatants on CTLL-2 cells. According to the supplier of the reagent, 1 ~1 of antibody can
Recombinant
10
IL 2 added
2.5
(U/mL)
0.6
0.15
0.04
Figure 1. Titration of IL 2 activity in supernatants of activated PBL and B cells using the CTLG2 cell line. Supernatants were harvested from 1 x 106/mL PBL or B cells activated for 48 hr with ionomycin (0.5 Fg/mL) and PMA (1 ng/mL). CTLL-2 cells (5,000 cells/well) were cultured in a total volume of 200 PI/well in the presence of the indicated final dilution of the supernatant or human rIL-2 for 48 hr and DNA synthesis by the indicator cells was measured as lZSIdUR incorporation during the last 4 hr of culture.
E Y:zj,
4
,
y>,yy.*z
8
16
---“-,
32 Reciprocal
64 dilution
128
256
of supernatant
512 added
10242048
1
214
/ Lagoo,
Tseng,
Sell
CYTOKINE,
neutralize 20 U of human recombinant IL 2. However, the antibody is species specific and was tested in greater quantities to neutralize the putative rabbit IL 2. As shown in Table 2, the anti-IL 2 antibody caused a dose-dependent inhibition of DNA synthesis by CTLL-2 cultured with rabbit B-cell supernatants. At a final dilution of 1:8, it caused more than 95% inhibition. In contrast, an irrelevant rabbit anti-human antibody (for example, anti-albumin) did not cause more than 30% inhibition of CTLL-2 even when added at a final dilution of 1:4 (not shown). In a parallel experiment, the effect of anti-IL 2 receptor monoclonal antibody on B-cell supernatantinduced CTLL-2 proliferation was studied (Table 3). The monoclonal antibody 7D4 was preincubated with the indicator cells for 30 min at 37°C before the addition of IL 2. The antibody caused 90% or more inhibition of proliferation driven by recombinant IL 2, PBL supernatants, or B-cell supernatants. In contrast, an irrelevant monoclonal antibody, LAB-2, had no inhibitory effect on CTLL-2 cells. Similarly, 7D4 antibody did not inhibit the proliferation of an IL 2-independent rabbit T-cell line RL-5 (data not shown). Taken together, these data indicate that the soluble factor detected in the activated rabbit B-cell culture supernatants is functionally and serologically identical to IL 2.
Source of IL 2 Activity We next examined the possibility that the IL 2 activity detected in the purified B-cell cultures was in fact derived from residual contaminating T cells. The amount of IL 2 produced by purified T-cell cultures containing decreasing numbers of T cells was compared to the IL 2 production by purified B cells (Fig. 2). B cells (1 x 106) produced more IL 2 than was produced by 1 x lo5 T cells. If it is assumed that the B cells do not produce any IL 2 and the IL 2 detected in the B cell cultures was derived entirely from contaminating T cells, these data would require that the B cell preparaTABLE 2. proliferation Stimulus None B Cell B Cell B Cell B Cell B Cell B Cell B Cell
Sup’ Sup” Sup’ Sup’ Sup’ Sup’ Sup’
Effect of anti-IL 2 antibody on CTLL-2 induced by B-cell supernatant. Anti-IL
2 antibody”
11256 1:128 1~64 1:32 1:16 1:8
DNA synthesisb 18 13,467 13,905 12,051 11,454 9,021 4,822 92
“Polyvalent rabbit anti-human recombinant IL 2 antibody added at the beginning of the culture at final dilution shown. %pm of ‘*‘IdUR incorporated by 5 x 10’ CTLL-2 cells during a 4-hr pulse after 2 days in culture. ‘Dilution (1:4 final) of supernatant of purified B cells stimulated for 48 hr with ionomycin (0.5 ag/mL) and PMA (lng/mL).
Vol. 2, No. 4 (July
1990: 272-279)
TABLE 3. Effect of anti-IL 2 receptor monoclonal antibody on CTLL-2 proliferation induced by B-cell supernatant. DNA Synthesis”
after addition of
Monoclonal
Antibodyb
(~g/culture)
7D4 Stimulus None rIL-2’ lOu/mL rIL-2 2.5u/mL PBL Supd 1:16 PBL Sup 1:64 B Cell Sup” 1:4 B Cell Sup 1:16
Medium 35 11,091 4,233 13,847 8,272 12,592 6,659
LAB-2
1
0.5
1
0.5
69 1,694 693 1,732 762 1,279 251
100 3,664 1,374 3,687 1,171 3,205 399
117 10,360 3,533 17,703 7,846 11,989 5,929
77 10,983 3,722 16,722 7,906 12,515 7,087
*cpm of ‘*‘IdUR incorporated by 5 x lo3 CTLL-2 cells, cultured in a total volume of 200 ~1, during a 4-hr pulse following 44 hr in culture. %dicated monoclonal antibody, in the form of a hybidoma supernatant, was added to CTLL-2 cells, at the final dilutions shown, 30 min prior to the addition of the stimulus. ‘Human recombinant IL 2 added at the indicated concentrations. %pernatant of PBL that were stimulated for 48 hr with ionomycin (0.5 pg/mL) and PMA (lng/mL). The supernatant was added at final dilutions shown. ‘Supernatant of purified B cells that were stimulated for 48 hr with ionomycin (0.5 pg/mL) and PMA (1 ng/mL). The supernatant was added at final dilutions shown.
tions tested must contain more than 10% T cells. However, the B cell preparations tested in these experiments contained less than 2% sIg- cells (possibly T cells). Therefore, it appears that the B cells themselves produced the IL 2 activity. To investigate further the ability of B cells to produce IL 2, the B-cell population was purified in two additional steps of panning as described in Materials and Methods. The surface phenotype of these different populations was determined by fluorescent staining for sIg (Fig. 3a). In a typical experiment, B’ cells were more than 99% sIg+, while B” cells were 92% sIg’. Since the B” population was prepared from B’ cells, the reduced proportion of sIg+ cells in the former is probably due to endocytosis and/or shedding of sIg during panning with anti-IgG and during scraping. The IL 2 production by the three B-cell populations is compared in Fig. 3b. The ability of the highly purified B” population to produce IL 2 provides further evidence that this lymphokine is produced by the B cells themselves. Furthermore, the amount of IL 2 produced by the three B-cell populations is similar. The absence of significantly reduced IL 2 production by B” cells in comparison with B cells suggests that the initial B-cell population itself was substantially free of contaminating T cells.
Autocvine Growth Factor Activity Produced by B Cells
of IL 2
The ability of B-cell-derived IL 2 to support B-cell proliferation was tested by employing B-cell blasts. The effects of B-cell and PBL supernatants on these B-cell
IL 2 production
by rabbit
B cells
/ 275
PBL +-
1x10"
T Cells
Figure 2. Titration of IL 2 activity cells cultured at various densities.
produced
1 x 106
-A-
5x
---
1x104
by T
Purified T cells at the indicated cell numbers/ml were cultured with ionomycin (5 pg/mL) and PMA (1 ng/mL) for 48 hr. Supernatants of these cultures as well as those from PBL and B cells, cultured as described in Fig. 1, were tested for IL 2 activity in the CTLL-2 assay.
blasts and on CTLL-2 are compared in Fig. 4. B-cell supernatant could induce DNA synthesis in B-cell blasts in a dose-dependent manner, analogous to its effect on CTLL-2. To determine if IL 2 in B-cell supernatants was responsible for inducing proliferation of B-cell blasts, CTLL-2 cells were overgrown the supernatants to absorb IL 2. Such IL 2-depleted supernatant could not induce DNA synthesis in B-cell blasts or in CTLL-2 cells (not shown), indicating that IL 2 is the major B-cell growth promoting lymphokine present in the B-cell supernatants. PBL supernatants also induced DNA synthesis in B-cell blasts, and this response was also abrogated by absorption of IL 2 by culturing CTLL-2 cells in the supernatant. Since IL 2 produced by B cells was seen to have significant B-cell growth factor activity, we next tested the ability of B-cell-derived IL 2 to act as an autocrine growth factor. DNA synthesis by PBLs and B cells cultured in decreasing numbers was measured after activation with ionomycin+ PMA. As seen from Table 4, DNA synthesis in B cells decreases proportionately to the decreasing cell number. There is no sudden cutoff as would be the case if DNA synthesis was dependent on the presence of T cells. Apparently, IL 2 produced by B cells is sufficient to promote DNA synthesis in activated B cells. Taken together these results suggest that IL 2 produced by B cells may act as an autocrine growth factor for B cells.
DISCUSSION We have reported here B cells produce functionally with ionomycin and PMA. factor secreted by B cells criteria. First, B-cell culture the proliferation of the IL
-*-
that rabbit peripheral blood active IL 2 after stimulation The identity of the growth was established by several supernatants could support 2-dependent murine cyto-
4
8
16
Reciprocal
dilution
32
64
128
lo4
of supernatant
toxic cell line, CTLL-2 (Fig. 1); second, this activity was neutralized by a rabbit anti-human rIL-2 antibody (Table 2); and finally, an anti-IL 2 receptor antibody could block CTLL-2 proliferation driven by B-cell supernatants (Table 3). The source of the lymphokine was established as B cells by the fact that the IL 2 produced by B-cell cultures was more than could be derived from the maximum possible number of T cells contaminating such cultures (Fig. 2) and by the observation that highly purified B cells could produce this lymphokine (Fig. 3). The present report thus shows for the first time that normal peripheral blood B lymphocytes can be induced to produce IL 2. Maino and Pace’* demonstrated that a murine B cell lymphoma cell line, 2PK-13, could, when activated with certain immune sera, produce a lymphokine capable of supporting the growth of two IL 2-dependent cell lines, CTLL-2 and HT-2. A more recent study of the same B-cell lymphoma cell line has shown in addition that the putative IL 2 activity can be abrogated by a monoclonal antibody to human IL 2 or to mouse IL 2 receptor.15 This study also demonstrated the presence of IL 2-specific mRNA in activated, but not in nonactivated, cells by the use of a nuclease protection assay. In the report of Taira et al., highly purified murine splenic B cells were shown to produce IL 2 after activation with A23187 (a calcium ionophore) and PMA, as well as with anti-Ig.l4 Parodi et al. have demonstrated the production of IL 2 from cells bearing B-cell phenotypes that have grown in a longterm culture of bone marrow from a patient with unclassified lymphoblastic leukemia.16 Is B-cell-derived IL 2 identical to T-cell-derived IL 2 in every respect? Our results and those of others mentioned above provide evidence that the two are functionally identical and share many structural properties as well. It is known that the human genome contains only one IL 2 gene.ig It seems likely that IL 2 derived
2%
/ Lagoo,
Tseng,
CYTOKINE,
Sell
Vol. 2, No. 4 (July
1990: 272-279)
a
b 13
I I 1BY-----
.88 ,
= 2 6 t;;
'*
9 8
-&- BCells i
p.. \ \ .. .. \
4
8 Reciprocal
Figure 3. (a) Surface B-cell populations.
phenotype
of purified
B-cell
populations
and (b) titration
16 dilution
-.a-
B’ Cells
-@
- B” Cells
32
of supernatant
of IL 2 activity
produced
by various
Fifteen thousand cells were examined for surface fluorescence on a flow cytometer (a), analyzing unstained PBL (A) or PBL and successively purified B cell populations stained with FITC-GARIg (B-E). Stained cells were (B) PBL; (C) B cells; (D) B’ cells; and (E) B” cells. Abscissa, cell number; ordinate, fluorescence intensity (channel number). (b) PBL and the various B cell populations were activated and their supernatants were tested for IL 2 content in the CTLL-2 proliferation assay as described in Fig. 1.
from T and B cells are products of the same gene. mRNA analysis carried out by others’4*‘5 provides evidence that B-cell-derived IL 2 is genetically identical to T-cell-derived IL 2. However, it is likely that posttranslational modifications occur differently in B and T cells, giving rise to similar but nonidentical IL 2 molecules. Such structural heterogeneity has been observed in IL 2 derived from unseparated tonsil cells and was attributed to differences in glycosylation.‘7 It is interesting that in the same study, the IL 2 derived from the T-cell leukemia cell line JURKAT was structurally
homogeneous. The possibility of structural microheterogeneity in B-cell-derived and T-cell-derived IL 2 necessitates independent testing of the functional properties of the former. Several groups have reported that IL 2 can act as a growth factor for B cells. Purified T-cellderived or recombinant IL 2 has been shown to induce DNA synthesis in murine12320and human’3321 activated B cells. However, the quantity of IL 2 required to induce proliferation in activated B cells is relatively high. It has been postulated I5 that this is due to the occupation of
IL 2 production
B-cell sup
by rabbit
B cells
/ 277
PBL sup
500.
Figure 4. B-cell growth-promoting produced by B cells and PBL.
activity of IL 2
B-cell and PBL supernatants were prepared and tested on CTLL-2 cells as described in Fig. 1. The B-cell growth-promoting activity of these supernatants was tested by culturing 5 x lo4 B cell blasts/well in the presence of the indicated dilution of the supernatant for 48 hr with DNA synthesis measured as the “‘IdUR incorporation during the last 4 hr of culture. Results are expressed as stimulation index, which equals DNA synthesis in the presence of the supernatant divided by DNA synthesis in the presence of medium alone..
L .g ,{ 300 50 .g .k 6%
200
100
100 0
DNA synthesis by PBL and B cells”: effect of cell Cells cultured
C&/Well 2 1 5 2.5 1 5 2.5 1
x x x x X x x x
lo5 lo5 lo4 lo4 lo4 10’ lo3 lo3
PBL
B Cells
25,709 10,016 3,342 1,026 366 182 95 41
11,651 3,658 1,257 496 188 95 37 20
synthesis by the indicated number of cells was measured after the cells were activated with ionomycin (0.5 Gg/mL) and PMA (1 ng/mL) for 48 hr and cultured in 96-well, flat-bottomed microtiter plates. “DNA
16
64
256 1024 Reciprocal
most of the IL 2 receptors by autocrine IL 2 and not because IL 2 is not a potent growth factor for B cells. In contrast to this view, some investigators have reported that IL 2 does not cause DNA synthesis in B cells.** Moreover, B-cell growth factors (BCGFs) distinct from IL 2 have been reported in mouse4 and man. 5 These apparently contradictory reports are partially reconciled by the fact that B cells can be divided into different populations according to their responsiveness to IL 2 or BCGFs. Thus, Alderson and othersz3 reported that in single-cell cultures of murine B lymphocytes, only a small fraction of cells proliferate in response to B-cell stimulatory factor-l. Vazquez et a1.24have found that human B cells can be divided into CD25+ and CD25populations after activation with anti-F antibody. Only the CD25+ population responds to IL 2, whereas only the CD25- population responds to a high molecular weight BCGF. A similar functional diversity may exist in rabbit B cells. Alternatively, IL 2 and BCGF may act synergistically, as was shown to occur in activated
TABLE 4. number.
4
0 dilution
:.a 4
16
64
256
1024
of supernatant
human B cells.25 The redundancy in lymphokines is becoming obvious (reviewed in reference 26) and emphasizes the need for establishing the physiological role of each lymphokine more carefully. The biological significance of B-cellderived IL 2 is not known. Our results indicate that it has B-cell growth factor activity, and the absorption experiment suggests that IL 2 is probably the only significant B-cell growth factor produced by rabbit B cells under the conditions of activation used here (Fig. 4). The limiting dilution experiment further suggests that the B-cell-derived IL 2 may act as an autocrine growth factor. Giovarelli et al. have shown that a B-cell-derived IL 2-like factor can enhance DNA synthesis in B-cell chronic lymphocytic leukemia cells activated with Staphylococcus aureus Cowan I (SAC).27 In the present study, B cells were activated with ionomycin and PMA, which give a stronger proliferation signal than does anti-Ig. However, B cells activated in vivo also receive additional signals from accessory cells besides those signals transduced through sIg. Therefore, a stronger activation than that caused by anti-Ig may not necessarily be non-physiological. Moreover, the observation by Walker et al.” that SAC can cause IL 2 secretion from B cells, demonstrates that activation through the usual surface receptor, that is, sIg, on B cells is capable of inducing IL 2 production. Therefore it seems likely that B-cell-derived IL 2 may be involved in B-cell activation in vivo. The autocrine effect of B-cell-derived IL 2 is probably limited to the first cell cycle. This postulation is based on the observation that B-cell blasts, produced by activation of resting B cells with ionomycin and PMA, cannot synthesize DNA when washed 48 hr after initiation of cultures. The preformed IL 2 is removed during washing and the cells presumably do not make more IL 2 during
218
/ Lagoo,
Tseng,
Sell
CYTOKINE,
the subsequent cell cycles, but respond to exogenous IL 2 (Fig. 4). In addition to acting as an autocrine growth factor, B-cell-derived IL 2 may play an important role in B-T cell interactions in vivo. B cells are known to function as antigen-presenting cells.28 It is possible that B cells interacting with the antigen are activated sufficiently to produce IL 2, as is shown to occur after activation with anti-Ig.t4 The production of IL 2 and presentation of antigen by B cells may provide the necessary signals for recruiting helper T cells. IL 2 can cause production of other lymphokines from T cells.2g These lymphokines in turn may be involved in clonal expansion and differentiation of B cells. This cyclic relationship may provide a means of amplifying the humoral immune response.
MATERIALS
AND
METHODS
Chemicals and Reagents ‘251-iododeoxyuridine (IdUR) was obtained from ICN Radiochemicals (Irvine, CA). PMA and pig skin gelatin (175 bloom) were purchased from Sigma Chemicals (St Louis, MO) and Ionomycin from Calbiochem (La Jolla, CA). Recombinant human IL 2 as well as rabbit anti-human recombinant IL 2 antibody were from Genzyme (Boston, MA). Affinitypurified goat anti-rabbit IgG (GARIgG) and fluorescein isothiocynate (FITC)-GARIgG were purchased from Cappel Worthington (Melvern, PA).
Cell Lines and Monoclonal Antibodies Hybridomas Lll-135 and 7D4 were obtained from the American Type Culture Collection (ATCC) (Rockville, MD). Ll l- 135 secretesan IgG, antibody that recognizesa differentiation antigen on all rabbit T cells. The monoclonal antibody from this hybridoma was obtained as ascites from Balb-C mice. Hybridoma 7D4 secretesan IgM antibody that binds to murine IL 2 receptor and inhibits IL 2driven proliferation but does not block IL 2 binding. Monoclonal antibody LAB-2, which recognizesan epitope on rabbit B cells, was generated in our laboratory and has been previously described.30These two antibodies were prepared as culture supernatants of respective hybridomas grown to saturation in DMEM medium supplemented with 10% fetal calf serum. The IL 2-dependent murine cell-line CTLL-2 was obtained from ATCC and maintained in medium containing 10% rat T-cell polyclone (Collaborative Research, Bedford, MA).
B-Cell Purijcation PBLs of adult New Zealand white male rabbits were obtained from defibrinated arterial blood by gravity sedimentation in 3.5% pig skin gelatin as described previously.31 Lymphocyte populations enriched for B cellswere prepared by complement-mediated lysis of T cells after treatment with goat anti-rabbit thymocyte serum as reported.32 The purity of the B-cell population was tested by direct immunofluorescence staining for sIg and by mitogen assaysas described elsewhere.33 These cell were designated B cells. In some experiments,
Vol. 2, No. 4 (July
1990: 272-279)
further purification of B cells was achieved by successive panning with two antibodies. In the first step, ascitesfrom mice carrying the Ll l-135 hybridoma was used in a 1: 10 dilution for panning. The non-adherent cells were harvested and these cells were called B’ cells. These cells were then panned on plates coated with 2 mL of affinity-purified goat anti-rabbit IgG containing 1 mg/mL antibody in phosphate buffered saline (PBS). The adherent cells were harvested by scraping and designated B” cells. Purified T cells were prepared by panning PBL on plates coated with goat anti-rabbit IgG and carefully harvesting the non-adherent cells.
IL 2 Assay IL 2 concentration was measured by the standard CTLL-2 proliferation assayas described previously.34
Assayfor B-Cell Growth Factor Activity Purified rabbit B cells (1 x 106/mL) were activated with ionomycin (0.5 pg/mL) and PMA (1 ng/mL) for 48 hr. Cells were washed twice with medium containing 2% fetal calf serum and resuspended to give a density of 2.5 x lo5 live cells/ml. These cells are called B-cell blasts and were used as indicator cells to examine B-cell growth factor activity of IL 2sontaining supernatants. It was confirmed in initial experiments that these cellsdid not synthesizesignificant amounts of DNA in the absence of further stimulation. It was also confirmed that these blasts did not respond to the quantities of ionomycin and PMA that was present in the IL 2-containing supernatants. Acknowledgments We thank M.-A. Hughes and S. Chakraborty for skillful technical support. P.L. Hsu, S. Norris, R. Rossen, W. Geoghegan, and L.S. Rodkey gave valuable suggestions during the course of this work and the preparation of the manuscript. This work was supported in part by NIH Project Grant number A121290-01.
REFERENCES 1. DeFranco AL, Raveche ES, Paul WE (1985) Separate control of B lymphocyte early activation and proliferation in response to anti-IgM antibodies. J Immunol 135:87-94. 2. Klaus GGB, Hawrylowicz CM, Carter CJ (1985) Activation and proliferation signals in mouse B cells. VI. Anti-Ig antibodies induce dose-dependent cell cycle progression in B cells. Immunol 55:41 l-418. 3. Sell S, Gel1 PGH (1965) Studies on rabbit vitro. I. Stimulation of blast transformation with serum. J Exp Med 122:423-441.
lymphocytes in an antiallotype
4. Howard M, Farrar J, Hilfiker M, Johnson B, Takatsu K, Hamaoka T, Paul WE (1982) Identification of a T cell-derived B cell growth factor distinct from interleukin 2. J Exp Med 155:914-923. 5. Muraguchi A, Kehrl JH, Butler JL, Fauci AS (1984) Sequential requirements for cell cycle progression of resting human B cells after activation by anti-Ig. J Immunol 132:176-180. 6. Lagoo A, Tseng CK, Sell S (1990) Molecular activation. II. IL-2-mediated signals are required
signals in B cell in late Gl for
IL 2 production
transition Immunol
to S phase 127:497-505.
after
ionomycin
7. Gordon J, Ley SC, Melmed MD, NC (1984) Immortalized B lymphocytes factor. Nature 310:145-147.
and
PMA
treatment.
English LS, Hughes-Jones produce B-cell growth
8. Morgan DA, Ruscetti FW, Gallo RC (1976) Selective growth of T-lymphocytes from normal human bone marrows. 193:1007-1008.
in vitro Science
9. Farrar JJ, Mizel SB, Fuller-Farrar J, Farrar WL, Hilfiker ML (1980) Macrophage-independent activation of helper T cells. I. Production of interleukin 2. J Immunol 125:793-798. 10. Waldmann TA, Goldman CK, Robb RJ, Depper JM, Leonard WJ, Sharrow SO, Bongiovanni KF, Korsmeyer SJ, Greene WC (1984) Expression of interleukin 2 receptors on activated human B cells. J Exp Med 160:1450-1466. Il. Saiki 0, Tanaka T, Doi S, Kishimoto and the functional role of a p70/75 interleukin human B cell. J Immunol 140:853-858.
S (1988) Expression 2-binding molecule on
12. Zubler RH, Lowenthal JW, Erard F, Hashimoto N, Devos R, MacDonald HR (1984) Activated B cells express receptor for and proliferate in response to pure interleukin 2. J Exp Med 160:11701183. 13. Mingari MC, Gerosa F, Moretta A, Zubler RH, Moretta L (1985) B cell growth factor activity of immunoaffinity-purified and recombinant human interleukin 2. Eur J Immunol 15:193-196. 14. Taira S, Matsui M, Hayakawa K, Yokoyama T, Nariuchi H (1987) Interleukin 2 secretion by B cell lines and splenic B cells stimulated with calcium ionophore and phorbol ester. J Immunol 13912957-2964. 15. Walker E, Leemhuis T, Roeder W (1988) Murine B lymphoma cell lines release functionally active interleukin 2 after stimulation with Staphylococcus aureus. J Immunol 140:859-865. 16. Parodi MT, Baffico M, Dini G, Ferrea G, Massino L, Cornaglia-Ferronis P (1986) EBNA-negative polyclonal B cells derived from a long term culture of unclassified acute lymphoblastic leukemia: IL-2 like activity of culture supernatants. Int J Cell Cloning 41263-273. growth 1094.
17. Robb factor(s)
21.
Cell
RJ, Smith KA (1981) Heterogeneity of human T-cell due to variable glycosylation. Mol Immunol 18:1087-
(1988) human
Nakagawa T, Nakagawa N Differential effects of interleukin B cells. J Immunol 140:465-469.
by rabbit
B cells
/ 279
Jr, Ambrus JL, Fauci AS 2 vs B cell growth factor on
22. Julius MH, Paige CJ, Leanderson T, Cambier JC (1987) Neither interleukin 2 nor gamma interferon directly promote growth or differentiation of mouse B cells. Stand J Immunol25:195-202. 23. Alderson MR, Pike BL, Nossal GJ (1987) Single cell studies on the role of B-cell stimulatory factor 1 in B-cell activation. Proc Nat1 AcadSci USA 84:1389-1393. 24. Vazquez A, Auffredou MT, Gerard JP, Delfraissy JF, Galanaud P (1987) Sequential effect of a high molecular weight B cell growth factor and of interleukin 2 on activated human B cells. J Immunol 139:2344-2348. 25. Romagnani S, Giudizi GM, Maggi E, Almerigogna F, Biagiotti R, Del-Prete G, Mazzetti M, Alessi A, Vercelli D, Ricci M (1985) Synergy of B cell growth factor and interleukin 2 in the proliferation of activated human B cells. Eur J Immunol 15:11581164. 26. Paul WE (1989) Pleiotropy lymphokines in the immune response.
and redundancy: Cell 57:521-524.
T cell-derived
27. Giovarelli M, Foa R, Lusso P, Fiero MT, Forni G (1988) Release of interleukin-2-like material by B-chronic lymphocytic leukemia cells. An autocrine or paracrine model of production and utilization, Leuk Res 12:201-209. 28. Chesnut RW, Grey HM cells to serve as antigen-presenting
(1981) Studies on the capacity of B cells. J Immunol 126:1075-1079.
29. Howard M, Matis L, Malek D, Nakanishi K, Paul WE (1983) reactive T cell line to secrete BCGF-I.
TR, Shevach E, Kell W, Kohen Interleukin 2 induces antigenJ Exp Med 158:2024-2038.
30. Loar LD, Dennison D, Sell characterization of monoclonal antibodies populations I. Tissue immunofluorescence sis. J Immunol 137:2784-2790.
S (1986) Production to rabbit lymphocyte and flow cytometric
and subanaly-
31. Endoh M, Sawada T, Lagoo A, Kimball PM, Gamez N, Loar L, Sell S (1987) DNA and immunoglobulin synthesis by rabbit peripheral blood lymphocytes in vitro: complete and incomplete stimulation. Cell Immunol 107:455-464.
18. Maino VC, Pace DR (1985) Induction of interleukin 2 from a murine B cell tumor by a factor found in immune serum. J Immunol 13412419-2425.
32. Redelman D, Scott CB, Sheppard HW, Sell S (1976) In vitro studies of the rabbit immune system. II. Functional characterization of rabbit T and B populations separated by adherence to nylon wool or lysis with anti-thymocyte serum and complement. Cell Immunol24: 1 l-23.
19. Fujita T, Takaoka C, Structure of the human interleukin 80:7437-7441.
33. Lagoo A, Sell S (1989) Early of resting B cells after Fab-Anti-Ig crosslinking. Cell Immunol 118:53-67.
Matsui H, Taniguchi T (1983) 2 gene. Proc Nat1 Acad Sci USA
20. Mond JJ, Thompson C, Finkelman FD, Farrar J, Schaefer M, Robb RJ (1985) Affinity-purified interleukin 2 induces proliferation of large but not small B cells. Proc Nat1 Acad Sci USA 82:1518-1521.
activation treatment:
and cell cycle entry role of receptor
34. Lagoo A, Tseng CK, Sell S (1990) Molecular signals in B cell activation. I. Differential refractory effects of incomplete signaling by ionomycin or PMA relate to autocrine IL-2 production and IL-2R expression. Cell Immunol 127:483-496.
Lagoo,
C. Kent Tseng,
Stewart
B
Sell*
Normal peripheral blood B cells produce a soluble factor after activation that is functionally indistinguishable from interleukin 2 (IL 2) and can support B cell proliferation in vitro. Purified rabbit peripheral blood B cells, when stimulated with a combination of ionomycin (0.5 pg/mL) and phorbol myristate acetate (PMA) (1 ng/mL), secreted a soluble factor in the culture medium that supported the IL 2-dependent cell line CTLL-2. The ability of these supernatants to support CTLL-2 growth was almost completely blocked by rabbit antibodies against human recombinant IL 2 and by the anti-IL 2 receptor monoclonal antibody 7D4. These data strongly suggest that the growth factor secreted by rabbit B cells is IL 2. To examine the possibility that the IL 2 activity detected in the B-cell cultures may be derived from residual T cells, B ceils were further purified by successive panning with a pan-T-cell monoclonal antibody, Lll-135, and goat anti-rabbit IgG. These highly purified B cells produced levels of IL 2 activity comparable to those produced by the initial B cell populations. Comparison of IL 2 production by decreasing numbers of purified T cells and purified B cells also indicated that the B cells were the source of IL 2 activity. Supernatants of activated B cells could support proliferation of B-cell blasts, and this activity could be completely absorbed by CTLL-2 cells, indicating that IL 2 is a major growth factor for B cells. Measurement of DNA synthesis by activated B cells in a limiting dilution assay suggested that T cells are not required for the induction of DNA synthesis and also that IL 2 produced by B cells can act as an autocrine growth factor. o 1990 by W.B. Saunders Company.
B cells from several species undergo proliferation follofiing activation with high doses of anti-immunoglobulin (anti-Ig) antibodies, apparently without the need for help from T cells.lm3On the other hand, it is also known that a significant increase in proliferation occurs upon addition of T-cell-derived lymphokines to anti-Igactivated B-cell cultures,4 and it is postulated by some investigators that T-cell-derived factors are essential for B-cell proliferation.5 We have recently demonstrated that additional activation signals provided by interleukin 2 (IL 2) in late G, are necessary for B cells to enter S phase.6 Thus it appears that even though B cells can proliferate in the absence of T cells, they cannot do so in the absence of lymphokines, particularly IL 2, raising the possibility of autocrine IL 2 production by activated
Department of Pathology and Laboratory Medicine, University of Texas Medical School, Houston, TX 77030. *To whom correspondence should be addressed at: Department of Pathology and Laboratory Medicine, University of Texas Medical School, P.O. Box 20708, Houston, TX 77225. o 1990 by W.B. Saunders Company. 1043-4666/90/0204-00$05.00/O KEY
272
WORDS:
Autocrine/B
Lymphocyte/IL
2
B cells. Earlier studies using Epstein Barr virustransformed B-cell lines have demonstrated the presence of an autocrine growth factor in culture supernatants of such cells after activation.7 However, it is not known whether normal peripheral blood B cells can produce autocrine growth factors that support their proliferation in the absence of T cells. IL 2 was initially described as a T-cell growth factor* produced by helper T cells.g It appears to play a major role in controlling B-cell proliferation as well. B cells have been reported to express both the ~55 (Taclike) and the ~70-75 components of the IL 2 receptor”,” and activated, but not resting, B cells respond to T-cell-derived IL 2 as well as to recombinant IL 2.12,13 Recent evidence indicates that certain B-cell lines and splenic B cells can be induced to secrete IL 2 activity after appropriate stimulation.‘4M16 On the basis of these observations it has been postulated, but never demonstrated experimentally, that B-cell-derived IL 2 may act as an autocrine growth factor.15 It is not known whether normal resting (Go) peripheral blood B cells can produce IL 2 or whether such autocrine IL 2 plays any role in controlling the proliferation of these cells. Although T-cell-derived IL 2 is known to act as a B-cell growth CYTOKINE,
Vol. 2, No. 4 (July),
1990: pp 272-279
IL 2 production
TABLE 1. Surface Ig staining and DNA synthesis by PBL and purified B cells. DNA Cell population
PBL B Cells
sIg + Cells”
Medium
47% 95%
144 + 14.2 104 zk 19.5
Synthesisb after activation
1,195 f 40.5 1,275 f 35.9
Con Ad
6,786 + 484.6 84 + 7.8
“Measured by flow cytometry following direct staining with FITC-labeled goat anti-rabbit IgG. bExpressed by cpm of ‘Z51UdR incorporation, mean k SD of triplicate cultures. ‘Affinity purified goat anti-rabbit Ig at 50 pg/mL. %oncanavalin A at 10 pg/mL.
Functional IL 2 Activity Supernatants
/ 273
in Activated
B-Cell
Supernatants from B cells and PBLs (1 x lo6 cells/ml in each case) activated with a combination of ionomycin (0.5 kg/mL) and phorbol myristate acetate (PMN (1 ng/mL) were harvested after 48 hr and tested in a standard IL 2 assay using the IL 2-dependent cell line CTLL-2. The response of these indicator cells to the lymphocyte culture supernatants was compared to their response to recombinant human IL 2 (Fig. 1). Rabbit PBLs and B cells produced a growth factor that could support the growth of CTLL-2, which was therefore presumed to represent IL 2 activity. By defining one unit of IL 2 activity as the amount of supernatant that induces half-maximal DNA synthesis in CTLL-2, it was observed that one million PBLs produced about 125 U/mL and one million B cells produced about 15 U/mL.
factor,” structural heterogeneity appears to exist between the T-cell-derived and B-cell-derived IL 2 due to differential glycosylation. l7 Therefore, it is necessary to investigate independently the ability of B-cell-derived IL 2 to function as a B-cell growth factor. Herein we provide evidence that highly purified rabbit peripheral blood B cells produce substantial quantities of IL 2 upon activation. The production of IL 2 appears to be limited to the first cell cycle after exiting Go and this IL 2 produced by B cells is capable of inducing DNA synthesis and supporting proliferation of activated B cells without the requirement of T cells in T-cell-depleted cultures. We discuss the possible role of IL 2 in B-cell proliferation and differentiation.
RESULTS Characterization
B cells
ing and DNA synthesis in response to anti-Ig and Con A of a representative cell preparation are shown in Table 1. In various experiments sIg+ cells in B cell preparations ranged from 95% to 98%. The B cells responded vigorously by DNA synthesis to anti-Ig but not to the T-cell mitogen, Con A. Thus these B-cell preparations were of high purity and depleted of T cells by functional and phenotypic criteria.
with
Anti-Ig’
by rabbit
Characterization of the Lymphokine B-Cell Supernatants
of B Cells
B cells purified from peripheral blood lymphocytes (PBLs) by treatment with anti-thymocyte serum and complement were tested by surface phenotype and mitogen response to determine the purity of the preparation. The results of surface immunoglobulin (sIg) stain-
Activity
in
To further characterize the growth-promoting activity produced by B cells, we tested the ability of a rabbit anti-human recombinant IL 2 antiserum to block the action of B cell supernatants on CTLL-2 cells. According to the supplier of the reagent, 1 ~1 of antibody can
Recombinant
10
IL 2 added
2.5
(U/mL)
0.6
0.15
0.04
Figure 1. Titration of IL 2 activity in supernatants of activated PBL and B cells using the CTLG2 cell line. Supernatants were harvested from 1 x 106/mL PBL or B cells activated for 48 hr with ionomycin (0.5 Fg/mL) and PMA (1 ng/mL). CTLL-2 cells (5,000 cells/well) were cultured in a total volume of 200 PI/well in the presence of the indicated final dilution of the supernatant or human rIL-2 for 48 hr and DNA synthesis by the indicator cells was measured as lZSIdUR incorporation during the last 4 hr of culture.
E Y:zj,
4
,
y>,yy.*z
8
16
---“-,
32 Reciprocal
64 dilution
128
256
of supernatant
512 added
10242048
1
214
/ Lagoo,
Tseng,
Sell
CYTOKINE,
neutralize 20 U of human recombinant IL 2. However, the antibody is species specific and was tested in greater quantities to neutralize the putative rabbit IL 2. As shown in Table 2, the anti-IL 2 antibody caused a dose-dependent inhibition of DNA synthesis by CTLL-2 cultured with rabbit B-cell supernatants. At a final dilution of 1:8, it caused more than 95% inhibition. In contrast, an irrelevant rabbit anti-human antibody (for example, anti-albumin) did not cause more than 30% inhibition of CTLL-2 even when added at a final dilution of 1:4 (not shown). In a parallel experiment, the effect of anti-IL 2 receptor monoclonal antibody on B-cell supernatantinduced CTLL-2 proliferation was studied (Table 3). The monoclonal antibody 7D4 was preincubated with the indicator cells for 30 min at 37°C before the addition of IL 2. The antibody caused 90% or more inhibition of proliferation driven by recombinant IL 2, PBL supernatants, or B-cell supernatants. In contrast, an irrelevant monoclonal antibody, LAB-2, had no inhibitory effect on CTLL-2 cells. Similarly, 7D4 antibody did not inhibit the proliferation of an IL 2-independent rabbit T-cell line RL-5 (data not shown). Taken together, these data indicate that the soluble factor detected in the activated rabbit B-cell culture supernatants is functionally and serologically identical to IL 2.
Source of IL 2 Activity We next examined the possibility that the IL 2 activity detected in the purified B-cell cultures was in fact derived from residual contaminating T cells. The amount of IL 2 produced by purified T-cell cultures containing decreasing numbers of T cells was compared to the IL 2 production by purified B cells (Fig. 2). B cells (1 x 106) produced more IL 2 than was produced by 1 x lo5 T cells. If it is assumed that the B cells do not produce any IL 2 and the IL 2 detected in the B cell cultures was derived entirely from contaminating T cells, these data would require that the B cell preparaTABLE 2. proliferation Stimulus None B Cell B Cell B Cell B Cell B Cell B Cell B Cell
Sup’ Sup” Sup’ Sup’ Sup’ Sup’ Sup’
Effect of anti-IL 2 antibody on CTLL-2 induced by B-cell supernatant. Anti-IL
2 antibody”
11256 1:128 1~64 1:32 1:16 1:8
DNA synthesisb 18 13,467 13,905 12,051 11,454 9,021 4,822 92
“Polyvalent rabbit anti-human recombinant IL 2 antibody added at the beginning of the culture at final dilution shown. %pm of ‘*‘IdUR incorporated by 5 x 10’ CTLL-2 cells during a 4-hr pulse after 2 days in culture. ‘Dilution (1:4 final) of supernatant of purified B cells stimulated for 48 hr with ionomycin (0.5 ag/mL) and PMA (lng/mL).
Vol. 2, No. 4 (July
1990: 272-279)
TABLE 3. Effect of anti-IL 2 receptor monoclonal antibody on CTLL-2 proliferation induced by B-cell supernatant. DNA Synthesis”
after addition of
Monoclonal
Antibodyb
(~g/culture)
7D4 Stimulus None rIL-2’ lOu/mL rIL-2 2.5u/mL PBL Supd 1:16 PBL Sup 1:64 B Cell Sup” 1:4 B Cell Sup 1:16
Medium 35 11,091 4,233 13,847 8,272 12,592 6,659
LAB-2
1
0.5
1
0.5
69 1,694 693 1,732 762 1,279 251
100 3,664 1,374 3,687 1,171 3,205 399
117 10,360 3,533 17,703 7,846 11,989 5,929
77 10,983 3,722 16,722 7,906 12,515 7,087
*cpm of ‘*‘IdUR incorporated by 5 x lo3 CTLL-2 cells, cultured in a total volume of 200 ~1, during a 4-hr pulse following 44 hr in culture. %dicated monoclonal antibody, in the form of a hybidoma supernatant, was added to CTLL-2 cells, at the final dilutions shown, 30 min prior to the addition of the stimulus. ‘Human recombinant IL 2 added at the indicated concentrations. %pernatant of PBL that were stimulated for 48 hr with ionomycin (0.5 pg/mL) and PMA (lng/mL). The supernatant was added at final dilutions shown. ‘Supernatant of purified B cells that were stimulated for 48 hr with ionomycin (0.5 pg/mL) and PMA (1 ng/mL). The supernatant was added at final dilutions shown.
tions tested must contain more than 10% T cells. However, the B cell preparations tested in these experiments contained less than 2% sIg- cells (possibly T cells). Therefore, it appears that the B cells themselves produced the IL 2 activity. To investigate further the ability of B cells to produce IL 2, the B-cell population was purified in two additional steps of panning as described in Materials and Methods. The surface phenotype of these different populations was determined by fluorescent staining for sIg (Fig. 3a). In a typical experiment, B’ cells were more than 99% sIg+, while B” cells were 92% sIg’. Since the B” population was prepared from B’ cells, the reduced proportion of sIg+ cells in the former is probably due to endocytosis and/or shedding of sIg during panning with anti-IgG and during scraping. The IL 2 production by the three B-cell populations is compared in Fig. 3b. The ability of the highly purified B” population to produce IL 2 provides further evidence that this lymphokine is produced by the B cells themselves. Furthermore, the amount of IL 2 produced by the three B-cell populations is similar. The absence of significantly reduced IL 2 production by B” cells in comparison with B cells suggests that the initial B-cell population itself was substantially free of contaminating T cells.
Autocvine Growth Factor Activity Produced by B Cells
of IL 2
The ability of B-cell-derived IL 2 to support B-cell proliferation was tested by employing B-cell blasts. The effects of B-cell and PBL supernatants on these B-cell
IL 2 production
by rabbit
B cells
/ 275
PBL +-
1x10"
T Cells
Figure 2. Titration of IL 2 activity cells cultured at various densities.
produced
1 x 106
-A-
5x
---
1x104
by T
Purified T cells at the indicated cell numbers/ml were cultured with ionomycin (5 pg/mL) and PMA (1 ng/mL) for 48 hr. Supernatants of these cultures as well as those from PBL and B cells, cultured as described in Fig. 1, were tested for IL 2 activity in the CTLL-2 assay.
blasts and on CTLL-2 are compared in Fig. 4. B-cell supernatant could induce DNA synthesis in B-cell blasts in a dose-dependent manner, analogous to its effect on CTLL-2. To determine if IL 2 in B-cell supernatants was responsible for inducing proliferation of B-cell blasts, CTLL-2 cells were overgrown the supernatants to absorb IL 2. Such IL 2-depleted supernatant could not induce DNA synthesis in B-cell blasts or in CTLL-2 cells (not shown), indicating that IL 2 is the major B-cell growth promoting lymphokine present in the B-cell supernatants. PBL supernatants also induced DNA synthesis in B-cell blasts, and this response was also abrogated by absorption of IL 2 by culturing CTLL-2 cells in the supernatant. Since IL 2 produced by B cells was seen to have significant B-cell growth factor activity, we next tested the ability of B-cell-derived IL 2 to act as an autocrine growth factor. DNA synthesis by PBLs and B cells cultured in decreasing numbers was measured after activation with ionomycin+ PMA. As seen from Table 4, DNA synthesis in B cells decreases proportionately to the decreasing cell number. There is no sudden cutoff as would be the case if DNA synthesis was dependent on the presence of T cells. Apparently, IL 2 produced by B cells is sufficient to promote DNA synthesis in activated B cells. Taken together these results suggest that IL 2 produced by B cells may act as an autocrine growth factor for B cells.
DISCUSSION We have reported here B cells produce functionally with ionomycin and PMA. factor secreted by B cells criteria. First, B-cell culture the proliferation of the IL
-*-
that rabbit peripheral blood active IL 2 after stimulation The identity of the growth was established by several supernatants could support 2-dependent murine cyto-
4
8
16
Reciprocal
dilution
32
64
128
lo4
of supernatant
toxic cell line, CTLL-2 (Fig. 1); second, this activity was neutralized by a rabbit anti-human rIL-2 antibody (Table 2); and finally, an anti-IL 2 receptor antibody could block CTLL-2 proliferation driven by B-cell supernatants (Table 3). The source of the lymphokine was established as B cells by the fact that the IL 2 produced by B-cell cultures was more than could be derived from the maximum possible number of T cells contaminating such cultures (Fig. 2) and by the observation that highly purified B cells could produce this lymphokine (Fig. 3). The present report thus shows for the first time that normal peripheral blood B lymphocytes can be induced to produce IL 2. Maino and Pace’* demonstrated that a murine B cell lymphoma cell line, 2PK-13, could, when activated with certain immune sera, produce a lymphokine capable of supporting the growth of two IL 2-dependent cell lines, CTLL-2 and HT-2. A more recent study of the same B-cell lymphoma cell line has shown in addition that the putative IL 2 activity can be abrogated by a monoclonal antibody to human IL 2 or to mouse IL 2 receptor.15 This study also demonstrated the presence of IL 2-specific mRNA in activated, but not in nonactivated, cells by the use of a nuclease protection assay. In the report of Taira et al., highly purified murine splenic B cells were shown to produce IL 2 after activation with A23187 (a calcium ionophore) and PMA, as well as with anti-Ig.l4 Parodi et al. have demonstrated the production of IL 2 from cells bearing B-cell phenotypes that have grown in a longterm culture of bone marrow from a patient with unclassified lymphoblastic leukemia.16 Is B-cell-derived IL 2 identical to T-cell-derived IL 2 in every respect? Our results and those of others mentioned above provide evidence that the two are functionally identical and share many structural properties as well. It is known that the human genome contains only one IL 2 gene.ig It seems likely that IL 2 derived
2%
/ Lagoo,
Tseng,
CYTOKINE,
Sell
Vol. 2, No. 4 (July
1990: 272-279)
a
b 13
I I 1BY-----
.88 ,
= 2 6 t;;
'*
9 8
-&- BCells i
p.. \ \ .. .. \
4
8 Reciprocal
Figure 3. (a) Surface B-cell populations.
phenotype
of purified
B-cell
populations
and (b) titration
16 dilution
-.a-
B’ Cells
-@
- B” Cells
32
of supernatant
of IL 2 activity
produced
by various
Fifteen thousand cells were examined for surface fluorescence on a flow cytometer (a), analyzing unstained PBL (A) or PBL and successively purified B cell populations stained with FITC-GARIg (B-E). Stained cells were (B) PBL; (C) B cells; (D) B’ cells; and (E) B” cells. Abscissa, cell number; ordinate, fluorescence intensity (channel number). (b) PBL and the various B cell populations were activated and their supernatants were tested for IL 2 content in the CTLL-2 proliferation assay as described in Fig. 1.
from T and B cells are products of the same gene. mRNA analysis carried out by others’4*‘5 provides evidence that B-cell-derived IL 2 is genetically identical to T-cell-derived IL 2. However, it is likely that posttranslational modifications occur differently in B and T cells, giving rise to similar but nonidentical IL 2 molecules. Such structural heterogeneity has been observed in IL 2 derived from unseparated tonsil cells and was attributed to differences in glycosylation.‘7 It is interesting that in the same study, the IL 2 derived from the T-cell leukemia cell line JURKAT was structurally
homogeneous. The possibility of structural microheterogeneity in B-cell-derived and T-cell-derived IL 2 necessitates independent testing of the functional properties of the former. Several groups have reported that IL 2 can act as a growth factor for B cells. Purified T-cellderived or recombinant IL 2 has been shown to induce DNA synthesis in murine12320and human’3321 activated B cells. However, the quantity of IL 2 required to induce proliferation in activated B cells is relatively high. It has been postulated I5 that this is due to the occupation of
IL 2 production
B-cell sup
by rabbit
B cells
/ 277
PBL sup
500.
Figure 4. B-cell growth-promoting produced by B cells and PBL.
activity of IL 2
B-cell and PBL supernatants were prepared and tested on CTLL-2 cells as described in Fig. 1. The B-cell growth-promoting activity of these supernatants was tested by culturing 5 x lo4 B cell blasts/well in the presence of the indicated dilution of the supernatant for 48 hr with DNA synthesis measured as the “‘IdUR incorporation during the last 4 hr of culture. Results are expressed as stimulation index, which equals DNA synthesis in the presence of the supernatant divided by DNA synthesis in the presence of medium alone..
L .g ,{ 300 50 .g .k 6%
200
100
100 0
DNA synthesis by PBL and B cells”: effect of cell Cells cultured
C&/Well 2 1 5 2.5 1 5 2.5 1
x x x x X x x x
lo5 lo5 lo4 lo4 lo4 10’ lo3 lo3
PBL
B Cells
25,709 10,016 3,342 1,026 366 182 95 41
11,651 3,658 1,257 496 188 95 37 20
synthesis by the indicated number of cells was measured after the cells were activated with ionomycin (0.5 Gg/mL) and PMA (1 ng/mL) for 48 hr and cultured in 96-well, flat-bottomed microtiter plates. “DNA
16
64
256 1024 Reciprocal
most of the IL 2 receptors by autocrine IL 2 and not because IL 2 is not a potent growth factor for B cells. In contrast to this view, some investigators have reported that IL 2 does not cause DNA synthesis in B cells.** Moreover, B-cell growth factors (BCGFs) distinct from IL 2 have been reported in mouse4 and man. 5 These apparently contradictory reports are partially reconciled by the fact that B cells can be divided into different populations according to their responsiveness to IL 2 or BCGFs. Thus, Alderson and othersz3 reported that in single-cell cultures of murine B lymphocytes, only a small fraction of cells proliferate in response to B-cell stimulatory factor-l. Vazquez et a1.24have found that human B cells can be divided into CD25+ and CD25populations after activation with anti-F antibody. Only the CD25+ population responds to IL 2, whereas only the CD25- population responds to a high molecular weight BCGF. A similar functional diversity may exist in rabbit B cells. Alternatively, IL 2 and BCGF may act synergistically, as was shown to occur in activated
TABLE 4. number.
4
0 dilution
:.a 4
16
64
256
1024
of supernatant
human B cells.25 The redundancy in lymphokines is becoming obvious (reviewed in reference 26) and emphasizes the need for establishing the physiological role of each lymphokine more carefully. The biological significance of B-cellderived IL 2 is not known. Our results indicate that it has B-cell growth factor activity, and the absorption experiment suggests that IL 2 is probably the only significant B-cell growth factor produced by rabbit B cells under the conditions of activation used here (Fig. 4). The limiting dilution experiment further suggests that the B-cell-derived IL 2 may act as an autocrine growth factor. Giovarelli et al. have shown that a B-cell-derived IL 2-like factor can enhance DNA synthesis in B-cell chronic lymphocytic leukemia cells activated with Staphylococcus aureus Cowan I (SAC).27 In the present study, B cells were activated with ionomycin and PMA, which give a stronger proliferation signal than does anti-Ig. However, B cells activated in vivo also receive additional signals from accessory cells besides those signals transduced through sIg. Therefore, a stronger activation than that caused by anti-Ig may not necessarily be non-physiological. Moreover, the observation by Walker et al.” that SAC can cause IL 2 secretion from B cells, demonstrates that activation through the usual surface receptor, that is, sIg, on B cells is capable of inducing IL 2 production. Therefore it seems likely that B-cell-derived IL 2 may be involved in B-cell activation in vivo. The autocrine effect of B-cell-derived IL 2 is probably limited to the first cell cycle. This postulation is based on the observation that B-cell blasts, produced by activation of resting B cells with ionomycin and PMA, cannot synthesize DNA when washed 48 hr after initiation of cultures. The preformed IL 2 is removed during washing and the cells presumably do not make more IL 2 during
218
/ Lagoo,
Tseng,
Sell
CYTOKINE,
the subsequent cell cycles, but respond to exogenous IL 2 (Fig. 4). In addition to acting as an autocrine growth factor, B-cell-derived IL 2 may play an important role in B-T cell interactions in vivo. B cells are known to function as antigen-presenting cells.28 It is possible that B cells interacting with the antigen are activated sufficiently to produce IL 2, as is shown to occur after activation with anti-Ig.t4 The production of IL 2 and presentation of antigen by B cells may provide the necessary signals for recruiting helper T cells. IL 2 can cause production of other lymphokines from T cells.2g These lymphokines in turn may be involved in clonal expansion and differentiation of B cells. This cyclic relationship may provide a means of amplifying the humoral immune response.
MATERIALS
AND
METHODS
Chemicals and Reagents ‘251-iododeoxyuridine (IdUR) was obtained from ICN Radiochemicals (Irvine, CA). PMA and pig skin gelatin (175 bloom) were purchased from Sigma Chemicals (St Louis, MO) and Ionomycin from Calbiochem (La Jolla, CA). Recombinant human IL 2 as well as rabbit anti-human recombinant IL 2 antibody were from Genzyme (Boston, MA). Affinitypurified goat anti-rabbit IgG (GARIgG) and fluorescein isothiocynate (FITC)-GARIgG were purchased from Cappel Worthington (Melvern, PA).
Cell Lines and Monoclonal Antibodies Hybridomas Lll-135 and 7D4 were obtained from the American Type Culture Collection (ATCC) (Rockville, MD). Ll l- 135 secretesan IgG, antibody that recognizesa differentiation antigen on all rabbit T cells. The monoclonal antibody from this hybridoma was obtained as ascites from Balb-C mice. Hybridoma 7D4 secretesan IgM antibody that binds to murine IL 2 receptor and inhibits IL 2driven proliferation but does not block IL 2 binding. Monoclonal antibody LAB-2, which recognizesan epitope on rabbit B cells, was generated in our laboratory and has been previously described.30These two antibodies were prepared as culture supernatants of respective hybridomas grown to saturation in DMEM medium supplemented with 10% fetal calf serum. The IL 2-dependent murine cell-line CTLL-2 was obtained from ATCC and maintained in medium containing 10% rat T-cell polyclone (Collaborative Research, Bedford, MA).
B-Cell Purijcation PBLs of adult New Zealand white male rabbits were obtained from defibrinated arterial blood by gravity sedimentation in 3.5% pig skin gelatin as described previously.31 Lymphocyte populations enriched for B cellswere prepared by complement-mediated lysis of T cells after treatment with goat anti-rabbit thymocyte serum as reported.32 The purity of the B-cell population was tested by direct immunofluorescence staining for sIg and by mitogen assaysas described elsewhere.33 These cell were designated B cells. In some experiments,
Vol. 2, No. 4 (July
1990: 272-279)
further purification of B cells was achieved by successive panning with two antibodies. In the first step, ascitesfrom mice carrying the Ll l-135 hybridoma was used in a 1: 10 dilution for panning. The non-adherent cells were harvested and these cells were called B’ cells. These cells were then panned on plates coated with 2 mL of affinity-purified goat anti-rabbit IgG containing 1 mg/mL antibody in phosphate buffered saline (PBS). The adherent cells were harvested by scraping and designated B” cells. Purified T cells were prepared by panning PBL on plates coated with goat anti-rabbit IgG and carefully harvesting the non-adherent cells.
IL 2 Assay IL 2 concentration was measured by the standard CTLL-2 proliferation assayas described previously.34
Assayfor B-Cell Growth Factor Activity Purified rabbit B cells (1 x 106/mL) were activated with ionomycin (0.5 pg/mL) and PMA (1 ng/mL) for 48 hr. Cells were washed twice with medium containing 2% fetal calf serum and resuspended to give a density of 2.5 x lo5 live cells/ml. These cells are called B-cell blasts and were used as indicator cells to examine B-cell growth factor activity of IL 2sontaining supernatants. It was confirmed in initial experiments that these cellsdid not synthesizesignificant amounts of DNA in the absence of further stimulation. It was also confirmed that these blasts did not respond to the quantities of ionomycin and PMA that was present in the IL 2-containing supernatants. Acknowledgments We thank M.-A. Hughes and S. Chakraborty for skillful technical support. P.L. Hsu, S. Norris, R. Rossen, W. Geoghegan, and L.S. Rodkey gave valuable suggestions during the course of this work and the preparation of the manuscript. This work was supported in part by NIH Project Grant number A121290-01.
REFERENCES 1. DeFranco AL, Raveche ES, Paul WE (1985) Separate control of B lymphocyte early activation and proliferation in response to anti-IgM antibodies. J Immunol 135:87-94. 2. Klaus GGB, Hawrylowicz CM, Carter CJ (1985) Activation and proliferation signals in mouse B cells. VI. Anti-Ig antibodies induce dose-dependent cell cycle progression in B cells. Immunol 55:41 l-418. 3. Sell S, Gel1 PGH (1965) Studies on rabbit vitro. I. Stimulation of blast transformation with serum. J Exp Med 122:423-441.
lymphocytes in an antiallotype
4. Howard M, Farrar J, Hilfiker M, Johnson B, Takatsu K, Hamaoka T, Paul WE (1982) Identification of a T cell-derived B cell growth factor distinct from interleukin 2. J Exp Med 155:914-923. 5. Muraguchi A, Kehrl JH, Butler JL, Fauci AS (1984) Sequential requirements for cell cycle progression of resting human B cells after activation by anti-Ig. J Immunol 132:176-180. 6. Lagoo A, Tseng CK, Sell S (1990) Molecular activation. II. IL-2-mediated signals are required
signals in B cell in late Gl for
IL 2 production
transition Immunol
to S phase 127:497-505.
after
ionomycin
7. Gordon J, Ley SC, Melmed MD, NC (1984) Immortalized B lymphocytes factor. Nature 310:145-147.
and
PMA
treatment.
English LS, Hughes-Jones produce B-cell growth
8. Morgan DA, Ruscetti FW, Gallo RC (1976) Selective growth of T-lymphocytes from normal human bone marrows. 193:1007-1008.
in vitro Science
9. Farrar JJ, Mizel SB, Fuller-Farrar J, Farrar WL, Hilfiker ML (1980) Macrophage-independent activation of helper T cells. I. Production of interleukin 2. J Immunol 125:793-798. 10. Waldmann TA, Goldman CK, Robb RJ, Depper JM, Leonard WJ, Sharrow SO, Bongiovanni KF, Korsmeyer SJ, Greene WC (1984) Expression of interleukin 2 receptors on activated human B cells. J Exp Med 160:1450-1466. Il. Saiki 0, Tanaka T, Doi S, Kishimoto and the functional role of a p70/75 interleukin human B cell. J Immunol 140:853-858.
S (1988) Expression 2-binding molecule on
12. Zubler RH, Lowenthal JW, Erard F, Hashimoto N, Devos R, MacDonald HR (1984) Activated B cells express receptor for and proliferate in response to pure interleukin 2. J Exp Med 160:11701183. 13. Mingari MC, Gerosa F, Moretta A, Zubler RH, Moretta L (1985) B cell growth factor activity of immunoaffinity-purified and recombinant human interleukin 2. Eur J Immunol 15:193-196. 14. Taira S, Matsui M, Hayakawa K, Yokoyama T, Nariuchi H (1987) Interleukin 2 secretion by B cell lines and splenic B cells stimulated with calcium ionophore and phorbol ester. J Immunol 13912957-2964. 15. Walker E, Leemhuis T, Roeder W (1988) Murine B lymphoma cell lines release functionally active interleukin 2 after stimulation with Staphylococcus aureus. J Immunol 140:859-865. 16. Parodi MT, Baffico M, Dini G, Ferrea G, Massino L, Cornaglia-Ferronis P (1986) EBNA-negative polyclonal B cells derived from a long term culture of unclassified acute lymphoblastic leukemia: IL-2 like activity of culture supernatants. Int J Cell Cloning 41263-273. growth 1094.
17. Robb factor(s)
21.
Cell
RJ, Smith KA (1981) Heterogeneity of human T-cell due to variable glycosylation. Mol Immunol 18:1087-
(1988) human
Nakagawa T, Nakagawa N Differential effects of interleukin B cells. J Immunol 140:465-469.
by rabbit
B cells
/ 279
Jr, Ambrus JL, Fauci AS 2 vs B cell growth factor on
22. Julius MH, Paige CJ, Leanderson T, Cambier JC (1987) Neither interleukin 2 nor gamma interferon directly promote growth or differentiation of mouse B cells. Stand J Immunol25:195-202. 23. Alderson MR, Pike BL, Nossal GJ (1987) Single cell studies on the role of B-cell stimulatory factor 1 in B-cell activation. Proc Nat1 AcadSci USA 84:1389-1393. 24. Vazquez A, Auffredou MT, Gerard JP, Delfraissy JF, Galanaud P (1987) Sequential effect of a high molecular weight B cell growth factor and of interleukin 2 on activated human B cells. J Immunol 139:2344-2348. 25. Romagnani S, Giudizi GM, Maggi E, Almerigogna F, Biagiotti R, Del-Prete G, Mazzetti M, Alessi A, Vercelli D, Ricci M (1985) Synergy of B cell growth factor and interleukin 2 in the proliferation of activated human B cells. Eur J Immunol 15:11581164. 26. Paul WE (1989) Pleiotropy lymphokines in the immune response.
and redundancy: Cell 57:521-524.
T cell-derived
27. Giovarelli M, Foa R, Lusso P, Fiero MT, Forni G (1988) Release of interleukin-2-like material by B-chronic lymphocytic leukemia cells. An autocrine or paracrine model of production and utilization, Leuk Res 12:201-209. 28. Chesnut RW, Grey HM cells to serve as antigen-presenting
(1981) Studies on the capacity of B cells. J Immunol 126:1075-1079.
29. Howard M, Matis L, Malek D, Nakanishi K, Paul WE (1983) reactive T cell line to secrete BCGF-I.
TR, Shevach E, Kell W, Kohen Interleukin 2 induces antigenJ Exp Med 158:2024-2038.
30. Loar LD, Dennison D, Sell characterization of monoclonal antibodies populations I. Tissue immunofluorescence sis. J Immunol 137:2784-2790.
S (1986) Production to rabbit lymphocyte and flow cytometric
and subanaly-
31. Endoh M, Sawada T, Lagoo A, Kimball PM, Gamez N, Loar L, Sell S (1987) DNA and immunoglobulin synthesis by rabbit peripheral blood lymphocytes in vitro: complete and incomplete stimulation. Cell Immunol 107:455-464.
18. Maino VC, Pace DR (1985) Induction of interleukin 2 from a murine B cell tumor by a factor found in immune serum. J Immunol 13412419-2425.
32. Redelman D, Scott CB, Sheppard HW, Sell S (1976) In vitro studies of the rabbit immune system. II. Functional characterization of rabbit T and B populations separated by adherence to nylon wool or lysis with anti-thymocyte serum and complement. Cell Immunol24: 1 l-23.
19. Fujita T, Takaoka C, Structure of the human interleukin 80:7437-7441.
33. Lagoo A, Sell S (1989) Early of resting B cells after Fab-Anti-Ig crosslinking. Cell Immunol 118:53-67.
Matsui H, Taniguchi T (1983) 2 gene. Proc Nat1 Acad Sci USA
20. Mond JJ, Thompson C, Finkelman FD, Farrar J, Schaefer M, Robb RJ (1985) Affinity-purified interleukin 2 induces proliferation of large but not small B cells. Proc Nat1 Acad Sci USA 82:1518-1521.
activation treatment:
and cell cycle entry role of receptor
34. Lagoo A, Tseng CK, Sell S (1990) Molecular signals in B cell activation. I. Differential refractory effects of incomplete signaling by ionomycin or PMA relate to autocrine IL-2 production and IL-2R expression. Cell Immunol 127:483-496.
