CELLULAR
IMMUNOLOGY
130,66-78 (1990)
Up-Regulation and Down-Regulation of Cell Surface and mRNA Expression of CD5 Antigen by Various Humoral Factors on Murine 7OZ/3 pre-B Cell Leukemia Cell Line: IL-4 Down-Regulates CD5 Antigen Expression’ HARUMI
JYONOUCHI,
ROSE M. Voss, AND ROBERT
A. GOOD
University of South Florida, AN Children’s Hospital, 801 Sixth Street South, St. Petersberg, Florida 33701 Received February 9, 1990; accepted April 25, 1990 CD5, a pan-T cell antigen, is expressed on a minor subset of normal B lymphocytes and on cells of most B lineage tumors or transformed B cells in both man and animal models. In the present study, the effects of various humoral factors on CD5 expression by cells of a subcloned 702/3 murine pre-B leukemia cell line were investigated. Among the humoral factors studied, only LPS up-regulated CD5 expression on 7OZ/3 cells (three- to fourfold) in a dose-dependent manner. However, this upregulatory effect of LPS was not observed when cells were cultured in serum-free medium. NZB-serum factor (NZB-SF), a cytokine we have identified and shown to enhance the maturation and proliferation of immature B cells, synergistically enhanced CD5 expression in the presence of suboptimal doses of LPS. IL-4 down-regulated CD5 expression by 7OZ/3 cells induced by LPS or LPS plus NZB-SF in a dose-dependent manner. IL-4 also suppressed spontaneous CD5 expression by 7OZ/3 cells. No other cytokine tested showed an inhibitory effect. LPS, EN-y, NZB-SF, and IL-1 enhanced sIg expression on 7OZ/3 cells and their action on sIg expression was not inhibited by IL-4. Thus, the down-regulatory action of IL-4 on CD5 expression appeared specific for this antigen. IFN-7, which inhibits IL-4 induced CD23 and DR expression on B cells, does not abolish the down-regulatory action of IL-4 on CD5 expression by 7OZ/3 cells. Changes in mRNA levels on coding CD5 were also examined following the incubation of 7OZ/3 cells (24 hr) in the presence of humoral factors which can influence CD5 Ag expression. The levels of mRNA for CD5 Ag were moderately increased in the presence of LPS and NZB-SF. IL-4 appeared to suppress the actions of NZB-SF and LPS at least in part by reducing the levels of mRNA encoding CD5. o 1990 Academic press, hc.
INTRODUCTION During the various stages of differentiation and activation, B lineage cells express a number of cell surface antigens, including CD19, CD20, CD2 1, CD23 (FcnR II), CD10 (CALLA), and MHC class II antigens. The expression of some of these cell surface antigens is thought to be related to the function or stage of B cell differentiation/activation and controlled to some degree by the complex interactions of soluble humoral factors (1). For example, IL-4 up-regulates CD23 and MHC class II antigen expression on B cells (2-4). This action of IL-4 can be down-regulated in turn by IFN’ This study is supported by Grants AI22360 and A125062. 66 OOO8-8749/90 $3.00 Copyright Q 1990 by Academic Press, Inc. All rights of repreduction in any form reserved.
REGULATION
OF CD5 ANTIGEN
EXPRESSION
67
y* (5,6). The antigen CD5, which was originally described as a pan-T cell marker, is also expressed on cells of a unique B cell subpopulation (7). In humans and experimental animals, the CD5+ B cell subset seems to include B cells which spontaneously produce IgM autoantibodies, the putative natural autoantibodies. These cells also appear to be independent from the main line of B lymphopoiesis in bone marrow (S10). In mice, CD5+ B cells are abundant in the peritoneal cavity and are thought to have an unusual self-renewal capacity (S- 10). CDS+ B cells spontaneously secrete Ig without stimulation, and have been shown to be relatively unresponsive to specific antigenic stimuli (9, 10). Whether secondary to their unusual self-renewing capacity or for reasons yet to be determined, clonal expansion appears to occur more frequently in CD5+ B cells than in CD5- B cells. This view is supported by the fact that although the CDS+ B cell subset is a very minor one (9- 12) human B-CLL cells and chemically or virally transformed B lineage cell lines express CD5 Ag in high frequency. The CD5+ B cell subpopulation also seems to expand as a consequence of immune dysregulation. For example, bone marrow transplant patients briefly express increased numbers of CD5+ B cells prior to full immune reconstitution (13). In some of the autoimmuneprone strains of mice, most notably NZB mice, CD5+ B cells are significantly increased in number in the spleen and peritoneal cavities (9, 10, 14, 15). However, the function of CD5 Ag on B lineage cells has not been well understood, nor have regulatory mechanisms which may control CD5 expression by such cells been elucidated. In the present study we examined regulatory mechanisms that might control CD5 Ag expression by cells of a 7OZ/3 pre-B leukemia cell line. We found that LPS or suboptimal doses of LPS plus NZB-SF can up-regulate CD5 expression by 7OZ/3 cells. Moreover, we show herein that IL-4 can specifically down-regulate the enhanced or spontaneous expression of CD5. Studies of mRNA expression for CD5 Ag showed that CD5 expression may be regulated by humoral factors at various levels including transcriptional and translational events. MATERIALS
AND METHODS
Cell cultures and cell surface marker analysis. 70213 murine B lineage leukemic cell line (ATCC, Rockville, MD) was subcloned and maintained in RPM1 1640 supplemented with 10% FCS (Hyclone, Logan, UT), 5 X lop5 M 2-ME and other additives. Cells were incubated for 24 hr (2.5 X lo5 cells/ml) in the presence of various humoral factors. Then 7OZ/3 cells were harvested, stained, and subjected for cell surface marker analysis by a flow cytometer (EPICS C, Coulter, Miami, FL). CD5 antigen was stained with fluorescein-conjugated mouse anti-Ly- 1 monoclonal antibody (Becton-Dickinson, Mountain View, CA). sIg was stained with fluorescein conjugated goat anti-mouse IgM antibody ((Fab), fragment). Anti-thy 1.2 Ab-FITC conjugate (Becton-Dickinson) was used for detecting nonspecific binding. CD5 Ag was detected with single-color analysis (usually axis low:40, GFL: 1100) (Fig. 2a). Reagerzts. Con A, PHA, PMA (Sigma, St. Louis, MO), PWM (GIBCO, Grand Island, NY), LPS (Difco, Baxter Scientific, Ocala, FL), rmIL-2, rmIL-3, rmIL-4, rmIL* Abbreviations used: colony stimulating factor-GM (CSF-GM), interleukin (IL), interferon-y (IFN-r), KB-specific DNA binding protein (NF-02 I KB), lipopolysaccharide (LPS), phenylmethyl sulfonyl fluoride (PMSF), phorbol myristate acetate (PMA), surface Ig(sIg), transforming growth factor a (TGF,), tumor necrosis factor (TNF), NZB serum factor (NZB-SF).
68
JYONOUCHI,
VOSS, AND GOOD
5, rhIL-6, and rmCSF-GM (Genzyme, Boston, MA) were obtained from commercial sources as indicated. Murine IL- 1 was kindly provided by Hoffmann-LaRoche (Nutley, NJ). Murine TNF and IFN-y were kindly provided by Genentech, Inc. (S. San Francisco, CA). NZB-serum factor (NZB-SF) was prepared in our laboratory as described earlier (16). Briefly, young NZB mice (4-6 weeks old) were injected with 1 mg C. parvum ip and sacrificed 14 days later. Prepared spleen cells were cultured for 24 hr in RPM1 1640, 2-ME, P + S, Hepes, Na pyruvate and 1% Nutridoma sp (Boehringer-Mannheim, Indianpolis, IN). Supematants were harvested and NZB-SF was purified by membrane affinity chromatography conjugated with mAb specific for NZB-SF (Mac 50, MEMTEK, Billerica, MA). After vigorous washing, NZB-SF was eluted by 3M NaSCN and dialyzed immediately. This NZB-SF preparation showed the major band of 60 kDa in SDS-PAGE analysis and was stored at 4°C in the presence ofproteinase inhibitor (0.2 tiphenylmethyl sulfonyl fluoride (PMSF), 1 mM EDTA, and 1 pg/ml apopriten (Sigma)). Slot blot analysis of mRNA specific for CD5. 70213 cells were incubated in the presence of various humoral factors for 6-24 hr (2.5 X lo5 cells/ml). Cells were washed once with RPM1 1640 and lysed in ice-cold lysis buffer (50 n&f Tris-HCl, pH 8.0, 100 mM NaCl, 5 mM MgCl*, 0.5% NP-40) in the presence of RNasin (10 U/ml) (Promega, Madison, WI). After proteinase K treatment and removal of DNA/ cell debris by microcentrifuger (3 min at 1S,OOOg), supematants were transferred to Eppendorfmicrocentrifuge tubes containing 4 ~1 of 20% SDS and the remaining protein was removed by phenol/chloroform extraction ( 17). RNA was precipitated with 2 vol of ethanol and 0.1 volume of 3M Na acetate, pH 5.2. After washing with 70% ethanol, RNA was air-dried, dissolved in distilled water and kept in a -70°C freezer. Denatured RNA with 1 Mglyoxal was slot blotted (Bio-Dot SF microfiltration units, Bio-Rad, Richmond, CA) on nylon membrane (Z probe, Bio-Rad). The cDNA probe for mouse CD5 was radiolabeled with [32P]d(CTP) by the oligolabeling method (Oligolabeling kit, Pharmacia-LKB, Uppsala, Sweden). The plasmid vectors inserted with a cDNA probe that codes for mouse CD5 were kindly provided by Drs. H. J. S. Huang and L. A. Herzenberg of Stanford University ( 18). Plasmid DNA was amplified with Escherichia coli strain, AG, cell (Stratogene, La Jolla, CA), purified (Qiagen plasmid DNA preparation kit, Qiagen, Studio City, CA) and digested with EcoRI and EcoRV (BRL, Gaithensburg, MD). cDNA fragment coding for CD5 Ag (400 bp) was electroeluted from the agarose gel in TAE buffer. A synthetic oligonucleotide (27 mer) for mouse a actin (Chlonotech, Palo Alto, CA) or cDNA fragment (400 bp) of human B actin (ATCC, Rockville, MD) was used as control probes. The x-ray films were exposed at -70°C and screened using a scanning densitometer (Pharmacia-LKB, Uppsala, Sweden). The average peak absorbance of 3-5 reading points detected by CD5 probe was standardized with the average peak absorbance by actin probe. RESULTS Efects of humoral factors on the CD5 expression of 7OZ/3 pre-B cell leukemia cell line. To screen the effects of humoral factors on CD5 expression by 7OZ/3 cells, subcloned 7OZ/3 cells were incubated for 24 hr in the presence of IL-l, IL-2, IL-3, IL-4, IL-5, IL-6, LPS, PWM, PHA, Con A, CSF-GM, TNF, IFN-y, PMA or NZBSF. As summarized in Table 1, only LPS significantly up-regulated CD5 expression by 7OZ/3 cells when cells were cultured in FCS-containing media. Although IL- 1 and IFN-y have been reported to enhance sIg expression on 7OZ/3 cells by inducing KB-
REGULATION
OF CD5 ANTIGEN
69
EXPRESSION
TABLE 1 CD5 Expression by 7OZ/3 Cells in the Presence of Various Soluble Factors Additives to the cells culture media Experiment 1 Medium only LPS IL- I IL-2 IL-3 IL-4 IL-5 rhIL-6
1m-Y CSFGM
NZB-SF TNF PMA Experiment 2 Medium only LPS PHA-P Con A PWM
9’0of CD5 + 7OZ/3
100 U/ml IOU lOOU IOU IOOU IOU IOOU IOU 1oou 1ou 1oou IOU 1oou IOU 1oou 1ou IOU IOOU 1ou IO rig/ml
6.8” 30.4 7.5 7.0 7.1 7.9 12.6 6.7 1.3 4.2 7.7 8.4 6.2 7.6 9.7 6.2 5.6 7.8 10.0 8.0 7.4 5.7
1 rdml I 1 5
5.8 41.4 4.8 5.1 3.3
1&ml
n 7OZ/3 cells were incubated in the presence of additives described above for 24 hr and examined for CD5 expression. Percentage of nonspecific staining detected by fluorescein conjugated anti-Thy 1.2 mAb was subtracted from the percentage of CD5+ 7OZ/3 cells. Data given are representative examples of repeated experiments.
specific DNA binding protein (NF-KB), (19-20), they showed no effect on CD5 expression on 7OZ/3 cells in these investigations. The up-regulatory effect of LPS on CD5 Ag expression was dose-dependent; the minimum amount of LPS for the full enhancement was 0.1 pg/ml (Fig. 1). LPS enhanced CD5 expression as long as FCS was present in the culture media (l-10%). However, when the culture medium was replaced with serum-free medium by supplementing 1% Nutridoma sp (BoehringerMannheim, Indianapolis, IN), LPS lost its enhancing effect. No other factor tested showed a significant effect on CD5 expression. Screening of the synergistic efects of humoral factors in regard to CD5 expression in the presence of suboptimal doses of LPS. Various cytokines are known synergistically to influence cells as costimulators. For example, IL-2 stimulates thymocytes in the presence of suboptimal dose of PHA (2 1). Therefore, we screened the effects of cytokines available to us on CD5 expression by 7OZ/3 cells in the presence of subopti-
70
JYONOUCHI, Up-regulation
VOSS, AND GOOD
of cm
-3
e~~ression
-2 Dose
Of LPS
0” 70213
-1 Ipdml)
cells
0
by LPS
1x10
FIG. 1. Percentage of CD5+ 7OZ/3 cells examined after incubating cells for 24 hr in the presence of various doses of LPS (0.001-2.5 rg/ml).
ma1 doses of LPS (0.05 pg/ml). Among cytokines tested, only NZB-SF synergistically enhanced CD5 expression by 7OZ/3 cells under these conditions. This synergistic action was dose-dependent (Fig. 2) and was abrogated when 7OZ/3 cells were cultured in serum-free medium. IL-4 down-regulates CD5 expression induced by LPS or LPS plus NZB-SF. We also studied whether some cytokines could down-regulate CD5 expression that was enhanced by LPS or LPS plus NZB-SF. Among the cytokines tested, IL-4 significantly inhibited the up-regulatory actions of LPS (Table 2 and Fig. 3). Enhanced CD5 expression by LPS plus NZB-SF was also inhibited by IL-4 (LPS 0.05 pg/ml + NZBSF 25 U/ml) (Fig. 4). Spontaneous expression of CD5 by 7OZ/3 cells was partly but not completely suppressed by IL-4. The inhibitory effect of IL-4 on CD5 expression seemed to be dose-dependent (Figs. 3 and 4). LPS, IFN-7, IL-l, and NZB-SF have been reported to induce sIg expression on 7OZ/3 cells, and their action on sIg expression was not inhibited when IL-4 (50 U/ml) was added to the culture media (Table 3). The addition of IL-4 did not seem to inhibit proliferation of 7OZ/3 cells significantly, since the cell concentrations detected by trypan-blue dye exclusion test were not significantly altered by the addition of IL-4 (10-1000 U/ml) (data not shown). IFN-7 inhibited IL4-induced CD23 and Ia expression (9, lo), but did not abolish the down-regulatory effect of IL-4 on CD5 expression by 7OZ/3 cells (Table 3). Thus, the down-regulatory effect of IL-4 on CD5 expression by 7OZ/3 cells appears to be specific for this antigen. Efects of IL-4, LPS, and NZB-SF on mRNA levels of CD5 antigen on 7OZ/3 cells. Total cytoplasmic RNA prepared from 7OZ/3 cells incubated with additives (LPS, IL-4, NZB-SF, LPS plus NZB-SF, IL-4 plus LPS, IL-4 plus NZB-SF, and IL-4 plus LPS plus NZB-SF). In initial experiments we found that mRNA levels of CD5 Ag increased more significantly following 24 hr of incubation than it did at 6 hr of incubation. Consequently, in all other experiments the cells were incubated for 24 hr. RNA samples prepared were slot blotted and hybridized with 32P-labeled cDNA fragments encoding CD5 Ag. The developed x-ray films were scanned by densitometer. The average peak absorbance of 3-5 reading points for each slot blot was standardized with average peak absorbance with actin probes. As shown in Fig. 5, the level of
REGULATION
OF CD5 ANTIGEN
71
EXPRESSION
a rx1 C D 5 + 7 0 : 3 c e I I s
Fluorescence
Intensity
[log]
Fluorescence
Intensity
[log]
FIG. 2. Percentage of CDS 7OZ/3 cells after incubating cells for 24 hr in the presence of LPS (0.05 pg/ ml) plus various doses of NZB-SF (0.0 I-100 U/ml). Typical changes of fluorescence intensity of CD5 Ag were also shown (b): A (unstained), B (control), C (LPS 0.05 &ml), and D (LPS + NZB-SF (100 U/ml).
mRNA coding for CD5 Ag was increased moderately in the presence of LPS (0.05- 1 pg/ml) or NZB-SF (lo-50 U/ml). However, the combination of NZB-SF and LPS did not increase the level of mRNA for CD5 Ag in an additive way, when compared to incubation with LPS or NZB-SF alone. When 7OZ/3 cells were incubated with IL4, the level of mRNA for CD5 Ag was decreased moderately (Fig. 5). Moreover, IL-4 repeatedly inhibited the up-regulatory effect of LPS or NZB-SF on the levels of mRNA coding for CD5 (Fig. 5). The decrease of mRNA levels by IL-4 were 34.7 f 10.7% in unstimulated 7OZ/3 cells, 53.4 f 10.0% in cells stimulated with LPS, and 6 1.2 f 10.4% in cells stimulated with LPS and NZB-SF. The results of three experiments were shown in Fig. 5. The down-regulatory action of IL-4 on CD5 Ag expression can be explained in part by the decrease of mRNA coding for CD5. DISCUSSION The 7OZ/3 pre-B cell leukemia cell line was originally developed from a thymectomized, methyl-nitrosurea injected BDFl mouse (22). The original cell line expressed
72
JYONOUCHI,
VOSS, AND GOOD TABLE 2
IL-4 but Not Other Humoral Factors Tested Inhibited the Actions of LPS on CD5 Expression by 7OZ/3 Cells % ofCDS+ 70213 cells”
Additives to the culture media Experiment 1 LPS (1 &ml) LPS + IL- 1 LPS + IL-2 LPS + IL-3 LPS + IL-4 LPS + IFNr LPS + TFN LPS + CSFoM Experiment 2 LPS (1 &ml) LPS + IL- 1 LPS + IL-2 LPS + IL-3 LPS + IL-4 LPS + IL-5 LPS + IL-6 LPS + TNF LPS + CSFor.,, LPS + NZB-SF
100 U/ml 1OOu 1oou 1oou 1oou 1oou 1oou
11.3 56.1 58.7 54.7 61.0 2.3 57.3 57.8 57.5
100 U/ml 1oou 1oou 1oou 1oou 1oou 1oou 1oou 100 U
11.6 34.8 32.8 32.3 37.1 9.0 33.0 38.5 33.0 35.6 42.5
n 7OZ/3 cells were incubated for 24 hr in the presence of additives as described above.
small amounts of sIg at the cell surfaces but had prominent cytoplasmic s heavy chains (22). sIg can be induced on these cells by several humoral factors including IL- 1, LPS, and NZB-SF through activation of transcription of already rearranged K Effects
C D 5 + 7 0
of
IL-4
on CD5 ex~rcssion
dose
-1 of
enhanced
by LPS
ULI 40.
3D-
F 3 20. c e I I s
10 -
01
1
-2
1
’
0
’ IL-4
t. 0
”
1
X
W/ml I
FIG. 3. Percentage of CDS+ 7OZ/3 cells after incubating cells for 24 hr in the presence of LPS (1 &ml), plus various doses of IL-4 (0.0 l-50 U/ml).
REGULATION
OF CD5 ANTIGEN
73
EXPRESSION
TABLE 3 IL-4 Does Not Inhibit the Enhancement of slg by IFNq, IL- I, or NZB-SF on 7OZ/3 Cells Additives to the culture media
% of CD5+ of 7OZ/3 cells”
% of slgf 7oz/3 cell@
6.3 5.4 5.0 5.8 25.8 2.0 2.0 0.6 0.9 0.7
75.2 90.9 88.2 91.3 90.8 72.9 82.7 85.3 89.0 90.0
Experiment 1: IFN--, IL-l NZB-SF LPS IL-4 IL-4 + IF?+-, IL-4 + IL- 1 IL-4 + NZB-SF IL-4 + LPS
50 U/ml 50 U/ml 10 U/ml 1 &ml 50 U/ml 50 U/ml 50 U/ml 10 U/ml 1 at/ml
% of CD5+ 7OZ/3 cells Experiment 2: LPS (1 a/ml) LPS + IL-4 50 U/ml LPS + IL-4
IMMUNOLOGY
130,66-78 (1990)
Up-Regulation and Down-Regulation of Cell Surface and mRNA Expression of CD5 Antigen by Various Humoral Factors on Murine 7OZ/3 pre-B Cell Leukemia Cell Line: IL-4 Down-Regulates CD5 Antigen Expression’ HARUMI
JYONOUCHI,
ROSE M. Voss, AND ROBERT
A. GOOD
University of South Florida, AN Children’s Hospital, 801 Sixth Street South, St. Petersberg, Florida 33701 Received February 9, 1990; accepted April 25, 1990 CD5, a pan-T cell antigen, is expressed on a minor subset of normal B lymphocytes and on cells of most B lineage tumors or transformed B cells in both man and animal models. In the present study, the effects of various humoral factors on CD5 expression by cells of a subcloned 702/3 murine pre-B leukemia cell line were investigated. Among the humoral factors studied, only LPS up-regulated CD5 expression on 7OZ/3 cells (three- to fourfold) in a dose-dependent manner. However, this upregulatory effect of LPS was not observed when cells were cultured in serum-free medium. NZB-serum factor (NZB-SF), a cytokine we have identified and shown to enhance the maturation and proliferation of immature B cells, synergistically enhanced CD5 expression in the presence of suboptimal doses of LPS. IL-4 down-regulated CD5 expression by 7OZ/3 cells induced by LPS or LPS plus NZB-SF in a dose-dependent manner. IL-4 also suppressed spontaneous CD5 expression by 7OZ/3 cells. No other cytokine tested showed an inhibitory effect. LPS, EN-y, NZB-SF, and IL-1 enhanced sIg expression on 7OZ/3 cells and their action on sIg expression was not inhibited by IL-4. Thus, the down-regulatory action of IL-4 on CD5 expression appeared specific for this antigen. IFN-7, which inhibits IL-4 induced CD23 and DR expression on B cells, does not abolish the down-regulatory action of IL-4 on CD5 expression by 7OZ/3 cells. Changes in mRNA levels on coding CD5 were also examined following the incubation of 7OZ/3 cells (24 hr) in the presence of humoral factors which can influence CD5 Ag expression. The levels of mRNA for CD5 Ag were moderately increased in the presence of LPS and NZB-SF. IL-4 appeared to suppress the actions of NZB-SF and LPS at least in part by reducing the levels of mRNA encoding CD5. o 1990 Academic press, hc.
INTRODUCTION During the various stages of differentiation and activation, B lineage cells express a number of cell surface antigens, including CD19, CD20, CD2 1, CD23 (FcnR II), CD10 (CALLA), and MHC class II antigens. The expression of some of these cell surface antigens is thought to be related to the function or stage of B cell differentiation/activation and controlled to some degree by the complex interactions of soluble humoral factors (1). For example, IL-4 up-regulates CD23 and MHC class II antigen expression on B cells (2-4). This action of IL-4 can be down-regulated in turn by IFN’ This study is supported by Grants AI22360 and A125062. 66 OOO8-8749/90 $3.00 Copyright Q 1990 by Academic Press, Inc. All rights of repreduction in any form reserved.
REGULATION
OF CD5 ANTIGEN
EXPRESSION
67
y* (5,6). The antigen CD5, which was originally described as a pan-T cell marker, is also expressed on cells of a unique B cell subpopulation (7). In humans and experimental animals, the CD5+ B cell subset seems to include B cells which spontaneously produce IgM autoantibodies, the putative natural autoantibodies. These cells also appear to be independent from the main line of B lymphopoiesis in bone marrow (S10). In mice, CD5+ B cells are abundant in the peritoneal cavity and are thought to have an unusual self-renewal capacity (S- 10). CDS+ B cells spontaneously secrete Ig without stimulation, and have been shown to be relatively unresponsive to specific antigenic stimuli (9, 10). Whether secondary to their unusual self-renewing capacity or for reasons yet to be determined, clonal expansion appears to occur more frequently in CD5+ B cells than in CD5- B cells. This view is supported by the fact that although the CDS+ B cell subset is a very minor one (9- 12) human B-CLL cells and chemically or virally transformed B lineage cell lines express CD5 Ag in high frequency. The CD5+ B cell subpopulation also seems to expand as a consequence of immune dysregulation. For example, bone marrow transplant patients briefly express increased numbers of CD5+ B cells prior to full immune reconstitution (13). In some of the autoimmuneprone strains of mice, most notably NZB mice, CD5+ B cells are significantly increased in number in the spleen and peritoneal cavities (9, 10, 14, 15). However, the function of CD5 Ag on B lineage cells has not been well understood, nor have regulatory mechanisms which may control CD5 expression by such cells been elucidated. In the present study we examined regulatory mechanisms that might control CD5 Ag expression by cells of a 7OZ/3 pre-B leukemia cell line. We found that LPS or suboptimal doses of LPS plus NZB-SF can up-regulate CD5 expression by 7OZ/3 cells. Moreover, we show herein that IL-4 can specifically down-regulate the enhanced or spontaneous expression of CD5. Studies of mRNA expression for CD5 Ag showed that CD5 expression may be regulated by humoral factors at various levels including transcriptional and translational events. MATERIALS
AND METHODS
Cell cultures and cell surface marker analysis. 70213 murine B lineage leukemic cell line (ATCC, Rockville, MD) was subcloned and maintained in RPM1 1640 supplemented with 10% FCS (Hyclone, Logan, UT), 5 X lop5 M 2-ME and other additives. Cells were incubated for 24 hr (2.5 X lo5 cells/ml) in the presence of various humoral factors. Then 7OZ/3 cells were harvested, stained, and subjected for cell surface marker analysis by a flow cytometer (EPICS C, Coulter, Miami, FL). CD5 antigen was stained with fluorescein-conjugated mouse anti-Ly- 1 monoclonal antibody (Becton-Dickinson, Mountain View, CA). sIg was stained with fluorescein conjugated goat anti-mouse IgM antibody ((Fab), fragment). Anti-thy 1.2 Ab-FITC conjugate (Becton-Dickinson) was used for detecting nonspecific binding. CD5 Ag was detected with single-color analysis (usually axis low:40, GFL: 1100) (Fig. 2a). Reagerzts. Con A, PHA, PMA (Sigma, St. Louis, MO), PWM (GIBCO, Grand Island, NY), LPS (Difco, Baxter Scientific, Ocala, FL), rmIL-2, rmIL-3, rmIL-4, rmIL* Abbreviations used: colony stimulating factor-GM (CSF-GM), interleukin (IL), interferon-y (IFN-r), KB-specific DNA binding protein (NF-02 I KB), lipopolysaccharide (LPS), phenylmethyl sulfonyl fluoride (PMSF), phorbol myristate acetate (PMA), surface Ig(sIg), transforming growth factor a (TGF,), tumor necrosis factor (TNF), NZB serum factor (NZB-SF).
68
JYONOUCHI,
VOSS, AND GOOD
5, rhIL-6, and rmCSF-GM (Genzyme, Boston, MA) were obtained from commercial sources as indicated. Murine IL- 1 was kindly provided by Hoffmann-LaRoche (Nutley, NJ). Murine TNF and IFN-y were kindly provided by Genentech, Inc. (S. San Francisco, CA). NZB-serum factor (NZB-SF) was prepared in our laboratory as described earlier (16). Briefly, young NZB mice (4-6 weeks old) were injected with 1 mg C. parvum ip and sacrificed 14 days later. Prepared spleen cells were cultured for 24 hr in RPM1 1640, 2-ME, P + S, Hepes, Na pyruvate and 1% Nutridoma sp (Boehringer-Mannheim, Indianpolis, IN). Supematants were harvested and NZB-SF was purified by membrane affinity chromatography conjugated with mAb specific for NZB-SF (Mac 50, MEMTEK, Billerica, MA). After vigorous washing, NZB-SF was eluted by 3M NaSCN and dialyzed immediately. This NZB-SF preparation showed the major band of 60 kDa in SDS-PAGE analysis and was stored at 4°C in the presence ofproteinase inhibitor (0.2 tiphenylmethyl sulfonyl fluoride (PMSF), 1 mM EDTA, and 1 pg/ml apopriten (Sigma)). Slot blot analysis of mRNA specific for CD5. 70213 cells were incubated in the presence of various humoral factors for 6-24 hr (2.5 X lo5 cells/ml). Cells were washed once with RPM1 1640 and lysed in ice-cold lysis buffer (50 n&f Tris-HCl, pH 8.0, 100 mM NaCl, 5 mM MgCl*, 0.5% NP-40) in the presence of RNasin (10 U/ml) (Promega, Madison, WI). After proteinase K treatment and removal of DNA/ cell debris by microcentrifuger (3 min at 1S,OOOg), supematants were transferred to Eppendorfmicrocentrifuge tubes containing 4 ~1 of 20% SDS and the remaining protein was removed by phenol/chloroform extraction ( 17). RNA was precipitated with 2 vol of ethanol and 0.1 volume of 3M Na acetate, pH 5.2. After washing with 70% ethanol, RNA was air-dried, dissolved in distilled water and kept in a -70°C freezer. Denatured RNA with 1 Mglyoxal was slot blotted (Bio-Dot SF microfiltration units, Bio-Rad, Richmond, CA) on nylon membrane (Z probe, Bio-Rad). The cDNA probe for mouse CD5 was radiolabeled with [32P]d(CTP) by the oligolabeling method (Oligolabeling kit, Pharmacia-LKB, Uppsala, Sweden). The plasmid vectors inserted with a cDNA probe that codes for mouse CD5 were kindly provided by Drs. H. J. S. Huang and L. A. Herzenberg of Stanford University ( 18). Plasmid DNA was amplified with Escherichia coli strain, AG, cell (Stratogene, La Jolla, CA), purified (Qiagen plasmid DNA preparation kit, Qiagen, Studio City, CA) and digested with EcoRI and EcoRV (BRL, Gaithensburg, MD). cDNA fragment coding for CD5 Ag (400 bp) was electroeluted from the agarose gel in TAE buffer. A synthetic oligonucleotide (27 mer) for mouse a actin (Chlonotech, Palo Alto, CA) or cDNA fragment (400 bp) of human B actin (ATCC, Rockville, MD) was used as control probes. The x-ray films were exposed at -70°C and screened using a scanning densitometer (Pharmacia-LKB, Uppsala, Sweden). The average peak absorbance of 3-5 reading points detected by CD5 probe was standardized with the average peak absorbance by actin probe. RESULTS Efects of humoral factors on the CD5 expression of 7OZ/3 pre-B cell leukemia cell line. To screen the effects of humoral factors on CD5 expression by 7OZ/3 cells, subcloned 7OZ/3 cells were incubated for 24 hr in the presence of IL-l, IL-2, IL-3, IL-4, IL-5, IL-6, LPS, PWM, PHA, Con A, CSF-GM, TNF, IFN-y, PMA or NZBSF. As summarized in Table 1, only LPS significantly up-regulated CD5 expression by 7OZ/3 cells when cells were cultured in FCS-containing media. Although IL- 1 and IFN-y have been reported to enhance sIg expression on 7OZ/3 cells by inducing KB-
REGULATION
OF CD5 ANTIGEN
69
EXPRESSION
TABLE 1 CD5 Expression by 7OZ/3 Cells in the Presence of Various Soluble Factors Additives to the cells culture media Experiment 1 Medium only LPS IL- I IL-2 IL-3 IL-4 IL-5 rhIL-6
1m-Y CSFGM
NZB-SF TNF PMA Experiment 2 Medium only LPS PHA-P Con A PWM
9’0of CD5 + 7OZ/3
100 U/ml IOU lOOU IOU IOOU IOU IOOU IOU 1oou 1ou 1oou IOU 1oou IOU 1oou 1ou IOU IOOU 1ou IO rig/ml
6.8” 30.4 7.5 7.0 7.1 7.9 12.6 6.7 1.3 4.2 7.7 8.4 6.2 7.6 9.7 6.2 5.6 7.8 10.0 8.0 7.4 5.7
1 rdml I 1 5
5.8 41.4 4.8 5.1 3.3
1&ml
n 7OZ/3 cells were incubated in the presence of additives described above for 24 hr and examined for CD5 expression. Percentage of nonspecific staining detected by fluorescein conjugated anti-Thy 1.2 mAb was subtracted from the percentage of CD5+ 7OZ/3 cells. Data given are representative examples of repeated experiments.
specific DNA binding protein (NF-KB), (19-20), they showed no effect on CD5 expression on 7OZ/3 cells in these investigations. The up-regulatory effect of LPS on CD5 Ag expression was dose-dependent; the minimum amount of LPS for the full enhancement was 0.1 pg/ml (Fig. 1). LPS enhanced CD5 expression as long as FCS was present in the culture media (l-10%). However, when the culture medium was replaced with serum-free medium by supplementing 1% Nutridoma sp (BoehringerMannheim, Indianapolis, IN), LPS lost its enhancing effect. No other factor tested showed a significant effect on CD5 expression. Screening of the synergistic efects of humoral factors in regard to CD5 expression in the presence of suboptimal doses of LPS. Various cytokines are known synergistically to influence cells as costimulators. For example, IL-2 stimulates thymocytes in the presence of suboptimal dose of PHA (2 1). Therefore, we screened the effects of cytokines available to us on CD5 expression by 7OZ/3 cells in the presence of subopti-
70
JYONOUCHI, Up-regulation
VOSS, AND GOOD
of cm
-3
e~~ression
-2 Dose
Of LPS
0” 70213
-1 Ipdml)
cells
0
by LPS
1x10
FIG. 1. Percentage of CD5+ 7OZ/3 cells examined after incubating cells for 24 hr in the presence of various doses of LPS (0.001-2.5 rg/ml).
ma1 doses of LPS (0.05 pg/ml). Among cytokines tested, only NZB-SF synergistically enhanced CD5 expression by 7OZ/3 cells under these conditions. This synergistic action was dose-dependent (Fig. 2) and was abrogated when 7OZ/3 cells were cultured in serum-free medium. IL-4 down-regulates CD5 expression induced by LPS or LPS plus NZB-SF. We also studied whether some cytokines could down-regulate CD5 expression that was enhanced by LPS or LPS plus NZB-SF. Among the cytokines tested, IL-4 significantly inhibited the up-regulatory actions of LPS (Table 2 and Fig. 3). Enhanced CD5 expression by LPS plus NZB-SF was also inhibited by IL-4 (LPS 0.05 pg/ml + NZBSF 25 U/ml) (Fig. 4). Spontaneous expression of CD5 by 7OZ/3 cells was partly but not completely suppressed by IL-4. The inhibitory effect of IL-4 on CD5 expression seemed to be dose-dependent (Figs. 3 and 4). LPS, IFN-7, IL-l, and NZB-SF have been reported to induce sIg expression on 7OZ/3 cells, and their action on sIg expression was not inhibited when IL-4 (50 U/ml) was added to the culture media (Table 3). The addition of IL-4 did not seem to inhibit proliferation of 7OZ/3 cells significantly, since the cell concentrations detected by trypan-blue dye exclusion test were not significantly altered by the addition of IL-4 (10-1000 U/ml) (data not shown). IFN-7 inhibited IL4-induced CD23 and Ia expression (9, lo), but did not abolish the down-regulatory effect of IL-4 on CD5 expression by 7OZ/3 cells (Table 3). Thus, the down-regulatory effect of IL-4 on CD5 expression by 7OZ/3 cells appears to be specific for this antigen. Efects of IL-4, LPS, and NZB-SF on mRNA levels of CD5 antigen on 7OZ/3 cells. Total cytoplasmic RNA prepared from 7OZ/3 cells incubated with additives (LPS, IL-4, NZB-SF, LPS plus NZB-SF, IL-4 plus LPS, IL-4 plus NZB-SF, and IL-4 plus LPS plus NZB-SF). In initial experiments we found that mRNA levels of CD5 Ag increased more significantly following 24 hr of incubation than it did at 6 hr of incubation. Consequently, in all other experiments the cells were incubated for 24 hr. RNA samples prepared were slot blotted and hybridized with 32P-labeled cDNA fragments encoding CD5 Ag. The developed x-ray films were scanned by densitometer. The average peak absorbance of 3-5 reading points for each slot blot was standardized with average peak absorbance with actin probes. As shown in Fig. 5, the level of
REGULATION
OF CD5 ANTIGEN
71
EXPRESSION
a rx1 C D 5 + 7 0 : 3 c e I I s
Fluorescence
Intensity
[log]
Fluorescence
Intensity
[log]
FIG. 2. Percentage of CDS 7OZ/3 cells after incubating cells for 24 hr in the presence of LPS (0.05 pg/ ml) plus various doses of NZB-SF (0.0 I-100 U/ml). Typical changes of fluorescence intensity of CD5 Ag were also shown (b): A (unstained), B (control), C (LPS 0.05 &ml), and D (LPS + NZB-SF (100 U/ml).
mRNA coding for CD5 Ag was increased moderately in the presence of LPS (0.05- 1 pg/ml) or NZB-SF (lo-50 U/ml). However, the combination of NZB-SF and LPS did not increase the level of mRNA for CD5 Ag in an additive way, when compared to incubation with LPS or NZB-SF alone. When 7OZ/3 cells were incubated with IL4, the level of mRNA for CD5 Ag was decreased moderately (Fig. 5). Moreover, IL-4 repeatedly inhibited the up-regulatory effect of LPS or NZB-SF on the levels of mRNA coding for CD5 (Fig. 5). The decrease of mRNA levels by IL-4 were 34.7 f 10.7% in unstimulated 7OZ/3 cells, 53.4 f 10.0% in cells stimulated with LPS, and 6 1.2 f 10.4% in cells stimulated with LPS and NZB-SF. The results of three experiments were shown in Fig. 5. The down-regulatory action of IL-4 on CD5 Ag expression can be explained in part by the decrease of mRNA coding for CD5. DISCUSSION The 7OZ/3 pre-B cell leukemia cell line was originally developed from a thymectomized, methyl-nitrosurea injected BDFl mouse (22). The original cell line expressed
72
JYONOUCHI,
VOSS, AND GOOD TABLE 2
IL-4 but Not Other Humoral Factors Tested Inhibited the Actions of LPS on CD5 Expression by 7OZ/3 Cells % ofCDS+ 70213 cells”
Additives to the culture media Experiment 1 LPS (1 &ml) LPS + IL- 1 LPS + IL-2 LPS + IL-3 LPS + IL-4 LPS + IFNr LPS + TFN LPS + CSFoM Experiment 2 LPS (1 &ml) LPS + IL- 1 LPS + IL-2 LPS + IL-3 LPS + IL-4 LPS + IL-5 LPS + IL-6 LPS + TNF LPS + CSFor.,, LPS + NZB-SF
100 U/ml 1OOu 1oou 1oou 1oou 1oou 1oou
11.3 56.1 58.7 54.7 61.0 2.3 57.3 57.8 57.5
100 U/ml 1oou 1oou 1oou 1oou 1oou 1oou 1oou 100 U
11.6 34.8 32.8 32.3 37.1 9.0 33.0 38.5 33.0 35.6 42.5
n 7OZ/3 cells were incubated for 24 hr in the presence of additives as described above.
small amounts of sIg at the cell surfaces but had prominent cytoplasmic s heavy chains (22). sIg can be induced on these cells by several humoral factors including IL- 1, LPS, and NZB-SF through activation of transcription of already rearranged K Effects
C D 5 + 7 0
of
IL-4
on CD5 ex~rcssion
dose
-1 of
enhanced
by LPS
ULI 40.
3D-
F 3 20. c e I I s
10 -
01
1
-2
1
’
0
’ IL-4
t. 0
”
1
X
W/ml I
FIG. 3. Percentage of CDS+ 7OZ/3 cells after incubating cells for 24 hr in the presence of LPS (1 &ml), plus various doses of IL-4 (0.0 l-50 U/ml).
REGULATION
OF CD5 ANTIGEN
73
EXPRESSION
TABLE 3 IL-4 Does Not Inhibit the Enhancement of slg by IFNq, IL- I, or NZB-SF on 7OZ/3 Cells Additives to the culture media
% of CD5+ of 7OZ/3 cells”
% of slgf 7oz/3 cell@
6.3 5.4 5.0 5.8 25.8 2.0 2.0 0.6 0.9 0.7
75.2 90.9 88.2 91.3 90.8 72.9 82.7 85.3 89.0 90.0
Experiment 1: IFN--, IL-l NZB-SF LPS IL-4 IL-4 + IF?+-, IL-4 + IL- 1 IL-4 + NZB-SF IL-4 + LPS
50 U/ml 50 U/ml 10 U/ml 1 &ml 50 U/ml 50 U/ml 50 U/ml 10 U/ml 1 at/ml
% of CD5+ 7OZ/3 cells Experiment 2: LPS (1 a/ml) LPS + IL-4 50 U/ml LPS + IL-4
