G-5-2 antibody recognizes alkaline phosphatase in lymphocytes
Eur. J. Immunol. 1990.20: 947-950
947
Short paper Carlos lMrquezov, Antonio De la Hers., Esther LeonardooA, Luis Pezziov, Andreas Strasseroo and Carlos Martinez-A.O Centro de Biologia Molecularo, Universidad Autbnoma, Madrid and Basel Institute for Immunologyo, Basel
Identity of PB76 differentiation antigen and lymphocyte alkaline phosphatase" Alkaline phosphatases (APases, EC 3.1.3.1) are ecto-enzymes bound to cell membranes by a phosphatidyl-inositol anchor. We have previously shown that APase is present on activated murine B cells and its expression correlates with the process of B cell differentiation into immunoglobulin secretion. Recently, a monoclonal antibody (mAb), G-5-2, that recognizes a 76-kDa molecule preferentially expressed on the surface of pre-B and plasma cells (PB76) was described. Some features shared by APase and PB76 differentiation antigen suggest that the G-5-2 mAb might be specific for lymphocyte APase. Here,we have analyzed this possibility and found an absolute correlation between PB76 expression in cells and their APase activity. Although PB76 has been described as a B cell-restricted marker, PB76 is also expressed on some T cells, such as theYAC-1 Tcell lymphoma, that are known to bear APase.Treatment of YAC-1 cells with phosphatidylinositol-specificphospholipase C resulted in a quantitatively correlated removal of both APase and PB76 antigens. Moreover, we demonstrate that PB76 antigen has APase activity using an enzyme-antigen immunoassay with the G-5-2 mAb. We conclude that PB76 and lymphocyte APase are one and the same antigen.
1 Introduction
We have previously shown that APase behaves as a differentiation antigen in the B cell lineage since: (a) Alkaline phosphatase (APase, orthophosphoricmonoester activated, but not resting, mouse B cells express high levels phosphohydrolase,EC 3.1.3.1) is the genetic denomination of ecto-APase activity [6] and (b) APase expression of a group of isozymes that catalyzes the hydrolysis of a correlates with mouse B cell maturation into Ig secretion, wide variety of phosphomonoesters at an alkaline pH [l]. rather than cellular proliferation [7]. The APase expressed They belong to a category of proteins that are bound to the on activated B cells belongs to the liverhonekidney type cellular membrane via a phosphatidyl-inositol anchor [2]. [7]. Along these lines, some properties of the PB76 B cell The physiological role of APase isozymes in vivo is differentiation antigen described by Strasser using the unknown. In mouse developmentAPase activity is special- G-5-2 mAb [8] suggested that this mAb could be directed ly prominent in primordial germ cells, early embryo cells against lymphocyte ecto-APase: (a) the similarity in the and placenta [3, 41. In the adult mouse, it is detectable in molecular weight reported for both molecules (- 76 kDa, most organs but high activities are found only in a few; (b) the expression of G-5-2 on pre-B cells and antibodysmall intestine, kidney and bone have specific activities secretingcells but not resting mature B or Tcells and (c) the about two orders of magnitude above those in other organs pattern of expression in several cell lines [9]. In the present [3,4].The gene encoding the isozyme expressed in placenta study we have analyzed this possibility and found that PB76 as well as liverhonekidney maps to the Akp-2 locus in and lymphocyte APase are the same differentiation antichromosome 4, whereas the isozyme occurring in the gen. intestine is structurally unrelated and is encoded by a separate locus [4,51. 2 Materials and methods [I 82081 2.1 Cultures and antibodies * This work was supported by grants from CICYT and EEC. The Basel Institute for Immunology was founded and is supported by Cells were grown in RPMT 1640 medium supplemented with 2 m L-glutamine, ~ 5 0 m ~2 - m , 1 0 m Hepes, F. Hoffmann-La Roche and Co.,Ltd., Basel. antibiotics and 5% of a selected nonstimulatory batch of Recipient of an MEC fellowship. A Recipient of an FISS fellowship. FCS. Mouse G-5-2 mAb (yzb,x) anti-mouse PB76 has been Present address: The Walter and Eliza Hall Institute of Medical described elsewhere [8]. F23.1 mAb (anti-mouseTcRVgs.2) Research,Victoria 3050, Australia was used as control mAb. Rat anti-mouse IL.2Ra (Tac) Correspondence: Carlos Martinez-A., Centro de Biologia Molecu- PC61.5 mAb [lo] was provided by Dr. M. Nabholz, and lar, Universidad Aut6noma de Madrid, Campus de Cantoblanco, used in biotin-conjugated form. FITC-labeled anti-Thy-1.2 28049 Madrid, Spain mAb and FITC (F(ab')Z goat anti-mouse IgG (GaMIg) Abbreviations: APase: Alkaline phosphatase GaMJg: Goat were purchased from Becton Dickinson (Mountain View, anti-mouse Ig PI-PLC: Phosphatidylinositol-specificphospho- CA) and Southern Biotechnology (Birmingham, AL), lipase C respectively. 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
0014-2980/90/0404-0$02.50/0
948
Eur. J. Immunol. 1990.20: 947-950
C. MBrquez, A. De la Hera, E. Leonard0 et al. APase- SP210
2.2 Treatment with phosphatidyliositol-specific phospholipase C (PI-PLC) Cells were washed twice with medium without FCS and resuspended at 5 x lo5 cells/50 pl in releasing buffer for PI-PLC (25 mM Tiis-HCL, 0 . 2 7 ~sucrose and 10 mM glucose, pH 7.4). After adding the appropiate amount of PI-PLC (Funakoshi Pharmaceutical Co. Ltd., Tokyo), tubes were incubated at 37°C for 1h with gentle shaking. Molecules released from the cells were recovered in the SN after centrifugation.
APase+ SP210
c
n
2.3 APase assays Microwells in 96-well plates contained either whole cells or the surface molecules from PI-PLC-treated cells that specifically bound to the antibodies used (G-5-2 or irrelevant mAb). In the first case, cells were attached to the plastic using 0.25 M glutaraldehyde (Merck, Darmstadt, FRG) in PBS. In the second situation, wells were sequentially: (a) coated with 100 p1 GaMIg at 100 pg/ml in buffer (50 m~ K2HP04/KH2P04, pH 8.0) for 2 h at 37°C and washed four times with PBS; (b) incubated twice with 0.5% powdered gelatin (Merck) in 200 p1 of PBS for 1 h at room temperature and extensively washed; (c) incubated for 2 h at room temperature with the indicated amounts of mAb in 100 p1 of PBS-gelatine and washed, and (d) incubated for 2 h at room temperature with 10 p1SN from PI-PLC-treated cells in 100 pl of PBS-gelatin. Finally the APase activity was evaluated measuring the absorbance of p-nitrophenol released from the p-nitrophenylphosphate substrate (Fluka, Buchs, Switzerland) as detailed elsewhere [7]. The results are expressed as the released p-nitrophenol in p~ concentration. 2.4 Quantitative FCM The procedure for direct and indirect immunofluorescence staining of the cells has been described elsewhere [ll]. PE-conjugated streptavidin was purchased from Southern Biotechnology. Quantitation of the surface fluorescence was performed analysing 104 viable cells using an EPICS Profile (Coulter Electronics, Hialeah, FL) equipped with a four-decade logarithmic amplifier. Cells that stained above the second-step reagent background were considered as positive.
Log Fluorescence Intensity
Figure 1. Immunofluorescence profiles of APase+ and APaseSp2/0 cells stained with the PB76-specific G-5-2 mAb. Parental APase+ Sp2/0 cells and an APase- mutant subclone were analyzed for surface expression of the PB76 antigen and for APase activity. PB76 expression was quantitated using indirect immunofluorescence with mAb G-5-2 and FITC-GaMIg and an EPICS-Profile flow cytometer. Background (FITC-GaMIg) fluorescence limit is indicated in the upper panel. APase activity was assayed by measuring the release of p-nitrophenol (pM) by 3 X 10" cells in 20 min. Values of APase activity in six independent experiments were 780 f 160and < 5 p~ for the APase+ and APase- Sp2/0cells, respectively.
Interestingly, cells in a representative APase- Sp2/0 mutant subclone do not express either PB76 surface antigen or APase activity. 3.2 PB76 is expressed in T cell lines and, like APase, is bound to the cell membrane by a PI anchor PB76 was originally defined as a B cell lineage-specific surface antigen which is preferentially expressed by pre-B and plasma cells, but not T and myeloid lineages. High levels of APase are also present in pre-B and plasma cell but not on myeloid cell lines [12]. APase is, however, expressed in some T cells [13-151. We have studied this apparent discrepancy between the distribution of PB76 and APase in T cells by analyzing whether YAC-1 cells, a mouse T cell lymphoma known t o express APase [13, 141, bears PB76. Control
PB76
3 Results 3.1 The lack of APase activity correlates with the loss of PB76 surface expression in a APase- mutant B cell hybridoma Previously we have characterized a number of subclones derived from the Sp2/0 B cell hybridoma for the lack of APase expression [9].We have now analyzed in parallel the parental APase+ Sp2/0 cells and the APase- Sp2/0 mutants for the correlated surface expression of PB76 differentiation antigen, defined by the G-5-2 mAb, and APase activity. As shown in Fig. 1, cells in the parental Sp2/0 line are recognized by the G-5-2 mAb and bear APase activity as revealed by FCM and enzymatic analyses, respectively.
Log Fluorescence Intensity Figure2. The T cell lymphoma YAC-1 expresses both PB76 differentiation antigen and APase activity. Cells were assayed for PB76 expression and for APase activity as indicated in Fig. 1. APase activity in six independent experiments was 160 k 35 pM. Background (FITC-GaMIg) fluorescence limit is indicated by a dashed line.
m l
Eur. J. Immunol. 1990.20: 947-950 CONTROL-FITC
CONTROL-FITC
G-5-2 antibody recognizes alkaline phosphatase in lymphocytes
(A)
CONTROL-PE
i
I L2-RcK 98.3x
90 %
f+p91.4 %
no
y--t--t"i" 35.9 %
0.002 u
98.5%
* 10.5 x
1 2 3
PI-PLC
99.9%
95.1x
1 2 3
0.01
u
2 3 0.05U
Loglo Flu0rescence >
100
-L 90 I-
949
As shown in Fig. 2,YAC-1 cells specifically bind the G-5-2 mAb as quantitated by immunofluorescence and FCM. APase is inserted in the cytoplasmic membrane of some cell types via a PI anchor that substitutes for the conventional transmembrane protein domain [2]. Given the absolute correlation in the distribution of PB76 expression and APase activity in several other cell lines (C. MBrquez, unpublished results), we examined whether the APase activity and the surface PB76 antigen expression were similarly affected by treatment with PI-PLC. For that purpose,YAC-1 lymphoma cells were treated with different doses of PI-PLC and both the surface expression of PB76 and the APase activity remaining in each cell group were measured using immunofluorescence and APase assays, respectively. Thy-1 and I L 2 R a molecules were taken as positive and negative internal controls (i.e., Thy-1 was chosen because it is another PL-linked glycoprotein [16], while LL2Ra is not). Fig. 3 A shows the staining for these markers in treated and untreated YAC-1 cells. There is a marked dose-dependent reduction of both the intensity of fluorescence for G-5-2 and Thy-1 (note the logarithmic scale) as well as in the number of positive cells for G-5-2. The effect is specific because IL 2R a antigen density is unaltered by the PI-PLC treatment. Interestingly, the parallel studies of the APase enzymatic activity, Fig. 3B, showed that YAC-1 cells also lose APase expression as a consequence of PI-PLC treatment. Moreover, the plots of remaining amount of APase activity and of PB76 antigen are tightly correlated.
80
a Q)
70
v)
8
30 0
60
U
0 50
z
250
z40
U
30
a
f
LL
2o
8
10
0
200
=L
n
a
5
150
0
N O . ' . 0 0 2 .01 .05
N O ".002.01 . 0 5
PI-PLC UNI T S I A S S A Y Figure3. Correlated reduction of the G-5-2 mAb binding and ecto-APase expression after PI-PLC treatment of YAC-1 cells. YAC-1 cells were incubated with or without titrated amounts of PI-PLC in releasing buffer (5 x 105 cells/50 pl). Thereafter cells from each group were either stained with mAb specific for PB76, Thy-1 or IL 2R a differentiation antigens (A) or assayed for APase activity (B). Anti-Thy-1 antibody was directly labeled with FITC. Second-step reagents were FITC-GaMIg (anti-PB76, G-5-2) or PE-conjugated streptavidin (anti-IL 2R a, PC61.5). Accordingly, background controls were FITC-Gah4Ig for G-5-2 and anti-Thy-1 mAb, an irrelevant biotin-conjugated mAb plus PE-conjugated sptreptavidin for IL 2R a (top panels). Results represent either the percentage of cells labeled above the threshold for positive staining indicated by the vertical bars (A) or the percent remaining APase activity and PB76 expression, defined using untreated cells as 100% control (B). Surface fluorescence and APase activity were quantitated after treating the cells with the indicated amounts of PI-PLC.
U
100
50
4016
8
4
2
1
m Ab DILUTIONS-'
Figure 4. In an EAIA assay the G-5-2 mAb recognizes the APase released from lymphocyte surface by PI-PLC treatment. Microwells in 96-well plates were coated with GaMIg and, sequentially with twofold serial dilutions of either no antibody (0),G-5-2 mAb SN [ ( O ) , starting from 50 pUwell] or F23.1 purified mAb [(a), starting from 2 pg/well] and, SN from PI-PLC-treated YAC-1cells. After extensive washing the APase activity retained in the wells was evaluated measuring the amount of p-nitrophenol (p-NP, p ~ ) released in 20 min.
950
Eur. J. Immunol. 1990.20: 947-950
C. Mhquez, A. De la Hera, E. Leonard0 et al.
3.3 PB76 is a lymphocyte APase The previous results suggest that the G-5-2 mAb might recognize the ecto-APase isozyme expressed in mouse lymphocytes. To address this possibility directly, we examined whether the PB76 antigen exhibits APase activity. An enzyme-antigen immunoassay (EAIA) applicable to the specific detection of mAb binding t o APase isozymes has been described. The EAIA offers the advantage that neither purified antibody nor purified enzyme is required [17].We reasoned that this assay could be used to study the putative APase activity of the PB76 antigen released from cells by PI-PLC treatment and specifically recognized by the purified G-5-2 mAb. As shown in Fig. 4, APase activity was detected, in a dose-dependent fashion, only in those wells coated with the G-5-2 mAb and not in those wells coated with an irrelevant mAb (F23.1).
Little is known about the regulation of APase, the physiologic substrates and the role of the distinct isozymes expressed in different tissues. We hope that the present report may help those interested in these problems. The finding that APase expression is tightly regulated during T and B cell differentiation and activation, may provide a suitable model for the analysis of these questions in mammalian cells, and, at the same time, be useful for a better characterization of the phenomena of B and T cell activation and differentiation. We thank Drs. J. Kauffman and G.-K. Simforthe critical reading of the manuscript.
Received December 21, 1989.
5 References 4 Discussion We have shown that PB76 is an antigen bound t o the cellular membrane by a PI anchor and that the solubilized molecule has APase activity. Other ecto-enzymes are also expressed on the B cell lineage as differentiation antigens (i.e. human CDlO and CD73/5’-nucleotidase [18]). The surface expression of the CDlO in pre-B cells as well as in some activated B cells was defined using mAb long before its recent identification as neutral endopeptidase [19]. Along these lines, the G-5-2 mAb becomes a useful tool to study the distribution and regulation of APase expression in mouse lymphocytes. We show here that PB76 expression is not restricted to the B cell lineage. The finding that the mouse PB76 antigen recognized by the G-5-2 mAb is the same as, at least, the T a n d B lymphocyte ecto-APase isozyme allows us to integrate data from separate research lines with no apparent relationship. Recently, a mouse lymphocyte alloantigen (Ly-31.1) was described which is an APase not restricted to B cells but also expressed in thymocytes, and in nonlymphoid tissues (preferentially in bone and kidney [20]). These results are consistent with our finding that lymphocyte APase shows the inhibition pattern of the isozyme group termed liverhonekidney [7]. They are also not in contradiction with the distribution of PB76 antigen in normal T cells. We have observed that the low levels of APase expression in activated Tcells [7] are not due t o B cell contamination but correlate with the detection of a subset of PB76+ T cells. A subset of thymocytes also bears PB76 antigen (C. Marquez, unpublished results). Available anti-Ly-31mAb recognize only the Ly-31.1allele while the G-5-2 mAb recognizes lymphocyte APase in both Ly-31.1 and Ly-31.2 strains ([8, 201 and this report). In summary, the lymphocyte APase isozymePB76 antigen can be now visualized as the same differentiation antigen on both B cells, preferentially pre-B cells and plasma cells, and on some subpopulations of thymocytes and Tcells that are still to be defined.
McComb, R. B., Bowers, G. N., Jr. and Posen, S., Alkaline phosphatases, Plenum Press, New York 1979. Low, M. G. and Saltier, A. R., Science 1988. 239: 268. Mog, R., in Celis, J. E. and Bravo, R. (Eds.), Biochemistry of Animal Development, vol. I, Academic Press, New York 1965, p. 307. 4 Terao, M. and Mintz, B., Proc. Natl. Acad. Sci. USA 1987.84: 7051. 5 Terao, M., Pravtcheva, D., Ruddle, F. H. and Mintz, B., Somatic Cell. and Mol. Genet. 1988. 14: 211. 6 Garcia-Rozas, C., Plaza, A., Diaz-Espada, F., Kreisler, M. and Martinez-A., C., J. Zmmunol. 1982.129: 52. 7 MBrquez, C. ,Toribio, M. L., Marcos, M. A. R., De la Hera, A., Barcena, A., Peui, L. and Martinez-A., C., J. Immunol. 1989. 142: 3187. 8 Strasser, A., Eur. J. Immunol. 1988. 18: 1803. 9 MArquez, C., PhD. Thesis, Universidad Autonoma, Madrid, 1989. 10 Ceredig, R., Lowenthal, J., Nabholtz, M. and MacDonald, R., Nature 1985. 314: 98. 11 Toribio, M. L., Martinez-A., C., Marcos, M. A. R., Mhquez, C., Cabrero, E. and De la Hera, A., Proc. Natl. Acad. Sci. USA 1986. 83: 6985. 12 Culvenor, J. G., Hams,W. A., Mande1,T. E.,Whitelaw, A. and Ferber, E., J. Immunol. 1981. 126: 1974. 13 Neumann, H., Klein, E., Hauck-Granoth, R.,Yachnin, S. and Ben-Bassat, H., Proc. Natl. Acad. Sci. USA 1976. 73: 1432. 14 Mosbach-Ozmen, L., Lehuen-Renard, A., Gaveriaux, C. and Loor, F., Ann. Inst. Pasteur 1986. 1370: 109. 15 Goffinet, G., Houben-Defresne, M. P. and Bonivier, J., J. Cancer Res. 1983. 43: 5416. 16 Low, M. G. and Kinkade, P. W., Nature 1985.318: 62. 17 Fishman, W. H., in Weber, G. (Ed.), Gene Expression in Normal and Transformed Cells, Plenum Press, New York 1983, p. 403. 18 Knapp,W., Rieber, P., Dorken, B., Schmidt, R. E., Stein, H. and von dem Borne, A. E. G., Immunol. Today 1989. 10: 253. 19 Shipp, M. A.,Vijayaraghavan, J., Schmidt, E.V., Masteller, E. L., D’Adamio, L., Hersh, L. B., and Reinherz, E. L., Proc. Natl. Acad. Sci. USA 1989. 86: 297. 20 Dairiki, K., Nakamura, S., Ikegawa, S., Nakamura, M., Fujimori,T.,%maoki,N. and ’lhda, N., Zmmunogenetics 1989. 29: 235.
Eur. J. Immunol. 1990.20: 947-950
947
Short paper Carlos lMrquezov, Antonio De la Hers., Esther LeonardooA, Luis Pezziov, Andreas Strasseroo and Carlos Martinez-A.O Centro de Biologia Molecularo, Universidad Autbnoma, Madrid and Basel Institute for Immunologyo, Basel
Identity of PB76 differentiation antigen and lymphocyte alkaline phosphatase" Alkaline phosphatases (APases, EC 3.1.3.1) are ecto-enzymes bound to cell membranes by a phosphatidyl-inositol anchor. We have previously shown that APase is present on activated murine B cells and its expression correlates with the process of B cell differentiation into immunoglobulin secretion. Recently, a monoclonal antibody (mAb), G-5-2, that recognizes a 76-kDa molecule preferentially expressed on the surface of pre-B and plasma cells (PB76) was described. Some features shared by APase and PB76 differentiation antigen suggest that the G-5-2 mAb might be specific for lymphocyte APase. Here,we have analyzed this possibility and found an absolute correlation between PB76 expression in cells and their APase activity. Although PB76 has been described as a B cell-restricted marker, PB76 is also expressed on some T cells, such as theYAC-1 Tcell lymphoma, that are known to bear APase.Treatment of YAC-1 cells with phosphatidylinositol-specificphospholipase C resulted in a quantitatively correlated removal of both APase and PB76 antigens. Moreover, we demonstrate that PB76 antigen has APase activity using an enzyme-antigen immunoassay with the G-5-2 mAb. We conclude that PB76 and lymphocyte APase are one and the same antigen.
1 Introduction
We have previously shown that APase behaves as a differentiation antigen in the B cell lineage since: (a) Alkaline phosphatase (APase, orthophosphoricmonoester activated, but not resting, mouse B cells express high levels phosphohydrolase,EC 3.1.3.1) is the genetic denomination of ecto-APase activity [6] and (b) APase expression of a group of isozymes that catalyzes the hydrolysis of a correlates with mouse B cell maturation into Ig secretion, wide variety of phosphomonoesters at an alkaline pH [l]. rather than cellular proliferation [7]. The APase expressed They belong to a category of proteins that are bound to the on activated B cells belongs to the liverhonekidney type cellular membrane via a phosphatidyl-inositol anchor [2]. [7]. Along these lines, some properties of the PB76 B cell The physiological role of APase isozymes in vivo is differentiation antigen described by Strasser using the unknown. In mouse developmentAPase activity is special- G-5-2 mAb [8] suggested that this mAb could be directed ly prominent in primordial germ cells, early embryo cells against lymphocyte ecto-APase: (a) the similarity in the and placenta [3, 41. In the adult mouse, it is detectable in molecular weight reported for both molecules (- 76 kDa, most organs but high activities are found only in a few; (b) the expression of G-5-2 on pre-B cells and antibodysmall intestine, kidney and bone have specific activities secretingcells but not resting mature B or Tcells and (c) the about two orders of magnitude above those in other organs pattern of expression in several cell lines [9]. In the present [3,4].The gene encoding the isozyme expressed in placenta study we have analyzed this possibility and found that PB76 as well as liverhonekidney maps to the Akp-2 locus in and lymphocyte APase are the same differentiation antichromosome 4, whereas the isozyme occurring in the gen. intestine is structurally unrelated and is encoded by a separate locus [4,51. 2 Materials and methods [I 82081 2.1 Cultures and antibodies * This work was supported by grants from CICYT and EEC. The Basel Institute for Immunology was founded and is supported by Cells were grown in RPMT 1640 medium supplemented with 2 m L-glutamine, ~ 5 0 m ~2 - m , 1 0 m Hepes, F. Hoffmann-La Roche and Co.,Ltd., Basel. antibiotics and 5% of a selected nonstimulatory batch of Recipient of an MEC fellowship. A Recipient of an FISS fellowship. FCS. Mouse G-5-2 mAb (yzb,x) anti-mouse PB76 has been Present address: The Walter and Eliza Hall Institute of Medical described elsewhere [8]. F23.1 mAb (anti-mouseTcRVgs.2) Research,Victoria 3050, Australia was used as control mAb. Rat anti-mouse IL.2Ra (Tac) Correspondence: Carlos Martinez-A., Centro de Biologia Molecu- PC61.5 mAb [lo] was provided by Dr. M. Nabholz, and lar, Universidad Aut6noma de Madrid, Campus de Cantoblanco, used in biotin-conjugated form. FITC-labeled anti-Thy-1.2 28049 Madrid, Spain mAb and FITC (F(ab')Z goat anti-mouse IgG (GaMIg) Abbreviations: APase: Alkaline phosphatase GaMJg: Goat were purchased from Becton Dickinson (Mountain View, anti-mouse Ig PI-PLC: Phosphatidylinositol-specificphospho- CA) and Southern Biotechnology (Birmingham, AL), lipase C respectively. 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
0014-2980/90/0404-0$02.50/0
948
Eur. J. Immunol. 1990.20: 947-950
C. MBrquez, A. De la Hera, E. Leonard0 et al. APase- SP210
2.2 Treatment with phosphatidyliositol-specific phospholipase C (PI-PLC) Cells were washed twice with medium without FCS and resuspended at 5 x lo5 cells/50 pl in releasing buffer for PI-PLC (25 mM Tiis-HCL, 0 . 2 7 ~sucrose and 10 mM glucose, pH 7.4). After adding the appropiate amount of PI-PLC (Funakoshi Pharmaceutical Co. Ltd., Tokyo), tubes were incubated at 37°C for 1h with gentle shaking. Molecules released from the cells were recovered in the SN after centrifugation.
APase+ SP210
c
n
2.3 APase assays Microwells in 96-well plates contained either whole cells or the surface molecules from PI-PLC-treated cells that specifically bound to the antibodies used (G-5-2 or irrelevant mAb). In the first case, cells were attached to the plastic using 0.25 M glutaraldehyde (Merck, Darmstadt, FRG) in PBS. In the second situation, wells were sequentially: (a) coated with 100 p1 GaMIg at 100 pg/ml in buffer (50 m~ K2HP04/KH2P04, pH 8.0) for 2 h at 37°C and washed four times with PBS; (b) incubated twice with 0.5% powdered gelatin (Merck) in 200 p1 of PBS for 1 h at room temperature and extensively washed; (c) incubated for 2 h at room temperature with the indicated amounts of mAb in 100 p1 of PBS-gelatine and washed, and (d) incubated for 2 h at room temperature with 10 p1SN from PI-PLC-treated cells in 100 pl of PBS-gelatin. Finally the APase activity was evaluated measuring the absorbance of p-nitrophenol released from the p-nitrophenylphosphate substrate (Fluka, Buchs, Switzerland) as detailed elsewhere [7]. The results are expressed as the released p-nitrophenol in p~ concentration. 2.4 Quantitative FCM The procedure for direct and indirect immunofluorescence staining of the cells has been described elsewhere [ll]. PE-conjugated streptavidin was purchased from Southern Biotechnology. Quantitation of the surface fluorescence was performed analysing 104 viable cells using an EPICS Profile (Coulter Electronics, Hialeah, FL) equipped with a four-decade logarithmic amplifier. Cells that stained above the second-step reagent background were considered as positive.
Log Fluorescence Intensity
Figure 1. Immunofluorescence profiles of APase+ and APaseSp2/0 cells stained with the PB76-specific G-5-2 mAb. Parental APase+ Sp2/0 cells and an APase- mutant subclone were analyzed for surface expression of the PB76 antigen and for APase activity. PB76 expression was quantitated using indirect immunofluorescence with mAb G-5-2 and FITC-GaMIg and an EPICS-Profile flow cytometer. Background (FITC-GaMIg) fluorescence limit is indicated in the upper panel. APase activity was assayed by measuring the release of p-nitrophenol (pM) by 3 X 10" cells in 20 min. Values of APase activity in six independent experiments were 780 f 160and < 5 p~ for the APase+ and APase- Sp2/0cells, respectively.
Interestingly, cells in a representative APase- Sp2/0 mutant subclone do not express either PB76 surface antigen or APase activity. 3.2 PB76 is expressed in T cell lines and, like APase, is bound to the cell membrane by a PI anchor PB76 was originally defined as a B cell lineage-specific surface antigen which is preferentially expressed by pre-B and plasma cells, but not T and myeloid lineages. High levels of APase are also present in pre-B and plasma cell but not on myeloid cell lines [12]. APase is, however, expressed in some T cells [13-151. We have studied this apparent discrepancy between the distribution of PB76 and APase in T cells by analyzing whether YAC-1 cells, a mouse T cell lymphoma known t o express APase [13, 141, bears PB76. Control
PB76
3 Results 3.1 The lack of APase activity correlates with the loss of PB76 surface expression in a APase- mutant B cell hybridoma Previously we have characterized a number of subclones derived from the Sp2/0 B cell hybridoma for the lack of APase expression [9].We have now analyzed in parallel the parental APase+ Sp2/0 cells and the APase- Sp2/0 mutants for the correlated surface expression of PB76 differentiation antigen, defined by the G-5-2 mAb, and APase activity. As shown in Fig. 1, cells in the parental Sp2/0 line are recognized by the G-5-2 mAb and bear APase activity as revealed by FCM and enzymatic analyses, respectively.
Log Fluorescence Intensity Figure2. The T cell lymphoma YAC-1 expresses both PB76 differentiation antigen and APase activity. Cells were assayed for PB76 expression and for APase activity as indicated in Fig. 1. APase activity in six independent experiments was 160 k 35 pM. Background (FITC-GaMIg) fluorescence limit is indicated by a dashed line.
m l
Eur. J. Immunol. 1990.20: 947-950 CONTROL-FITC
CONTROL-FITC
G-5-2 antibody recognizes alkaline phosphatase in lymphocytes
(A)
CONTROL-PE
i
I L2-RcK 98.3x
90 %
f+p91.4 %
no
y--t--t"i" 35.9 %
0.002 u
98.5%
* 10.5 x
1 2 3
PI-PLC
99.9%
95.1x
1 2 3
0.01
u
2 3 0.05U
Loglo Flu0rescence >
100
-L 90 I-
949
As shown in Fig. 2,YAC-1 cells specifically bind the G-5-2 mAb as quantitated by immunofluorescence and FCM. APase is inserted in the cytoplasmic membrane of some cell types via a PI anchor that substitutes for the conventional transmembrane protein domain [2]. Given the absolute correlation in the distribution of PB76 expression and APase activity in several other cell lines (C. MBrquez, unpublished results), we examined whether the APase activity and the surface PB76 antigen expression were similarly affected by treatment with PI-PLC. For that purpose,YAC-1 lymphoma cells were treated with different doses of PI-PLC and both the surface expression of PB76 and the APase activity remaining in each cell group were measured using immunofluorescence and APase assays, respectively. Thy-1 and I L 2 R a molecules were taken as positive and negative internal controls (i.e., Thy-1 was chosen because it is another PL-linked glycoprotein [16], while LL2Ra is not). Fig. 3 A shows the staining for these markers in treated and untreated YAC-1 cells. There is a marked dose-dependent reduction of both the intensity of fluorescence for G-5-2 and Thy-1 (note the logarithmic scale) as well as in the number of positive cells for G-5-2. The effect is specific because IL 2R a antigen density is unaltered by the PI-PLC treatment. Interestingly, the parallel studies of the APase enzymatic activity, Fig. 3B, showed that YAC-1 cells also lose APase expression as a consequence of PI-PLC treatment. Moreover, the plots of remaining amount of APase activity and of PB76 antigen are tightly correlated.
80
a Q)
70
v)
8
30 0
60
U
0 50
z
250
z40
U
30
a
f
LL
2o
8
10
0
200
=L
n
a
5
150
0
N O . ' . 0 0 2 .01 .05
N O ".002.01 . 0 5
PI-PLC UNI T S I A S S A Y Figure3. Correlated reduction of the G-5-2 mAb binding and ecto-APase expression after PI-PLC treatment of YAC-1 cells. YAC-1 cells were incubated with or without titrated amounts of PI-PLC in releasing buffer (5 x 105 cells/50 pl). Thereafter cells from each group were either stained with mAb specific for PB76, Thy-1 or IL 2R a differentiation antigens (A) or assayed for APase activity (B). Anti-Thy-1 antibody was directly labeled with FITC. Second-step reagents were FITC-GaMIg (anti-PB76, G-5-2) or PE-conjugated streptavidin (anti-IL 2R a, PC61.5). Accordingly, background controls were FITC-Gah4Ig for G-5-2 and anti-Thy-1 mAb, an irrelevant biotin-conjugated mAb plus PE-conjugated sptreptavidin for IL 2R a (top panels). Results represent either the percentage of cells labeled above the threshold for positive staining indicated by the vertical bars (A) or the percent remaining APase activity and PB76 expression, defined using untreated cells as 100% control (B). Surface fluorescence and APase activity were quantitated after treating the cells with the indicated amounts of PI-PLC.
U
100
50
4016
8
4
2
1
m Ab DILUTIONS-'
Figure 4. In an EAIA assay the G-5-2 mAb recognizes the APase released from lymphocyte surface by PI-PLC treatment. Microwells in 96-well plates were coated with GaMIg and, sequentially with twofold serial dilutions of either no antibody (0),G-5-2 mAb SN [ ( O ) , starting from 50 pUwell] or F23.1 purified mAb [(a), starting from 2 pg/well] and, SN from PI-PLC-treated YAC-1cells. After extensive washing the APase activity retained in the wells was evaluated measuring the amount of p-nitrophenol (p-NP, p ~ ) released in 20 min.
950
Eur. J. Immunol. 1990.20: 947-950
C. Mhquez, A. De la Hera, E. Leonard0 et al.
3.3 PB76 is a lymphocyte APase The previous results suggest that the G-5-2 mAb might recognize the ecto-APase isozyme expressed in mouse lymphocytes. To address this possibility directly, we examined whether the PB76 antigen exhibits APase activity. An enzyme-antigen immunoassay (EAIA) applicable to the specific detection of mAb binding t o APase isozymes has been described. The EAIA offers the advantage that neither purified antibody nor purified enzyme is required [17].We reasoned that this assay could be used to study the putative APase activity of the PB76 antigen released from cells by PI-PLC treatment and specifically recognized by the purified G-5-2 mAb. As shown in Fig. 4, APase activity was detected, in a dose-dependent fashion, only in those wells coated with the G-5-2 mAb and not in those wells coated with an irrelevant mAb (F23.1).
Little is known about the regulation of APase, the physiologic substrates and the role of the distinct isozymes expressed in different tissues. We hope that the present report may help those interested in these problems. The finding that APase expression is tightly regulated during T and B cell differentiation and activation, may provide a suitable model for the analysis of these questions in mammalian cells, and, at the same time, be useful for a better characterization of the phenomena of B and T cell activation and differentiation. We thank Drs. J. Kauffman and G.-K. Simforthe critical reading of the manuscript.
Received December 21, 1989.
5 References 4 Discussion We have shown that PB76 is an antigen bound t o the cellular membrane by a PI anchor and that the solubilized molecule has APase activity. Other ecto-enzymes are also expressed on the B cell lineage as differentiation antigens (i.e. human CDlO and CD73/5’-nucleotidase [18]). The surface expression of the CDlO in pre-B cells as well as in some activated B cells was defined using mAb long before its recent identification as neutral endopeptidase [19]. Along these lines, the G-5-2 mAb becomes a useful tool to study the distribution and regulation of APase expression in mouse lymphocytes. We show here that PB76 expression is not restricted to the B cell lineage. The finding that the mouse PB76 antigen recognized by the G-5-2 mAb is the same as, at least, the T a n d B lymphocyte ecto-APase isozyme allows us to integrate data from separate research lines with no apparent relationship. Recently, a mouse lymphocyte alloantigen (Ly-31.1) was described which is an APase not restricted to B cells but also expressed in thymocytes, and in nonlymphoid tissues (preferentially in bone and kidney [20]). These results are consistent with our finding that lymphocyte APase shows the inhibition pattern of the isozyme group termed liverhonekidney [7]. They are also not in contradiction with the distribution of PB76 antigen in normal T cells. We have observed that the low levels of APase expression in activated Tcells [7] are not due t o B cell contamination but correlate with the detection of a subset of PB76+ T cells. A subset of thymocytes also bears PB76 antigen (C. Marquez, unpublished results). Available anti-Ly-31mAb recognize only the Ly-31.1allele while the G-5-2 mAb recognizes lymphocyte APase in both Ly-31.1 and Ly-31.2 strains ([8, 201 and this report). In summary, the lymphocyte APase isozymePB76 antigen can be now visualized as the same differentiation antigen on both B cells, preferentially pre-B cells and plasma cells, and on some subpopulations of thymocytes and Tcells that are still to be defined.
McComb, R. B., Bowers, G. N., Jr. and Posen, S., Alkaline phosphatases, Plenum Press, New York 1979. Low, M. G. and Saltier, A. R., Science 1988. 239: 268. Mog, R., in Celis, J. E. and Bravo, R. (Eds.), Biochemistry of Animal Development, vol. I, Academic Press, New York 1965, p. 307. 4 Terao, M. and Mintz, B., Proc. Natl. Acad. Sci. USA 1987.84: 7051. 5 Terao, M., Pravtcheva, D., Ruddle, F. H. and Mintz, B., Somatic Cell. and Mol. Genet. 1988. 14: 211. 6 Garcia-Rozas, C., Plaza, A., Diaz-Espada, F., Kreisler, M. and Martinez-A., C., J. Zmmunol. 1982.129: 52. 7 MBrquez, C. ,Toribio, M. L., Marcos, M. A. R., De la Hera, A., Barcena, A., Peui, L. and Martinez-A., C., J. Immunol. 1989. 142: 3187. 8 Strasser, A., Eur. J. Immunol. 1988. 18: 1803. 9 MArquez, C., PhD. Thesis, Universidad Autonoma, Madrid, 1989. 10 Ceredig, R., Lowenthal, J., Nabholtz, M. and MacDonald, R., Nature 1985. 314: 98. 11 Toribio, M. L., Martinez-A., C., Marcos, M. A. R., Mhquez, C., Cabrero, E. and De la Hera, A., Proc. Natl. Acad. Sci. USA 1986. 83: 6985. 12 Culvenor, J. G., Hams,W. A., Mande1,T. E.,Whitelaw, A. and Ferber, E., J. Immunol. 1981. 126: 1974. 13 Neumann, H., Klein, E., Hauck-Granoth, R.,Yachnin, S. and Ben-Bassat, H., Proc. Natl. Acad. Sci. USA 1976. 73: 1432. 14 Mosbach-Ozmen, L., Lehuen-Renard, A., Gaveriaux, C. and Loor, F., Ann. Inst. Pasteur 1986. 1370: 109. 15 Goffinet, G., Houben-Defresne, M. P. and Bonivier, J., J. Cancer Res. 1983. 43: 5416. 16 Low, M. G. and Kinkade, P. W., Nature 1985.318: 62. 17 Fishman, W. H., in Weber, G. (Ed.), Gene Expression in Normal and Transformed Cells, Plenum Press, New York 1983, p. 403. 18 Knapp,W., Rieber, P., Dorken, B., Schmidt, R. E., Stein, H. and von dem Borne, A. E. G., Immunol. Today 1989. 10: 253. 19 Shipp, M. A.,Vijayaraghavan, J., Schmidt, E.V., Masteller, E. L., D’Adamio, L., Hersh, L. B., and Reinherz, E. L., Proc. Natl. Acad. Sci. USA 1989. 86: 297. 20 Dairiki, K., Nakamura, S., Ikegawa, S., Nakamura, M., Fujimori,T.,%maoki,N. and ’lhda, N., Zmmunogenetics 1989. 29: 235.
