Veterinary Immunology and Immunopathology, 25 (1990) 195-208 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands
195
Identification and Characterization of Monoclonal Antibodies Reactive with Bovine, Caprine and Ovine T-Lymphocyte Determinants by Flow Microfluorimetry R.A. LARSEN, M.L. MONAGHAN', Y.H. PARK, M.J. HAMILTON, J.A. ELLIS" and W.C. DAVIS*
Department of Veterinary Microbiology and Pathology, College of Veterinary Medicine, Washington St. University, Pullman, WA 99164-7040(U.S.A.) 'Department of Farm Animal Clinical Studies, Faculty of Veterinary Medicine, University College Dublin, Ballsbridge Dublin 4 (Ireland) '~Departments of Veterinary Sciences and Molecular Biology, University of Wyoming, Laramie, WY (U.S.A.) {Accepted 23 November 1989)
ABSTRACT Larsen, R.A., Monaghan, M.L., Park, Y.H., Hamilton, M.J., Ellis, J.A. and Davis, W.C., 1990. Identification and characterization of monoclonal antibodies reactive with bovine, caprine and ()vine T-lymphocyte determinants by flow microfluorimetry. Vet. Immunol. Immunopathol., 25: 195-208. Comparative flow microfluorimetric (FMF) analysis was used to identify and characterize 27 monoclonal antibodies (MoAbs) reactive with bovine T-lymphocytes. Determinants present on all circulating T-lymphocytes were recognized by 11 MoAbs, 8 of which blocked E rosette formation. Determinants present on only the BoCD4 + T-lymphocyte subset were detected by 9 MoAbs, while determinants restricted to the BoCD8 ÷ T-lymphocyte subset were recognized by 7 MoAbs. Competitive labeling experiments demonstrated that determinants recognized by subset-specific MoAbs were present on BoCD4 or BoCD8 molecules. Comparative studies revealed that some determinants, both pan-T specific and subset-specific, were conserved on homologous (orthologous) molecules expressed on leukocytes from other species of ruminants. Polymorphism was evident with several determinants.
INTRODUCTION Extensive studies of leukocyte differentiation antigens by flow microfluori-
merry (FMF) have demonstrated that many leukocyte surface molecules ex*Towhomcorrespondence shouldbe addressed.
0165-2427/90/$03.50
© 1990 Elsevier Science Publishers B.V.
196
R.A. L A R S E N E T AL.
hibit unique patterns of expression on one or more lineages of leukocytes (Lanier et al., 1983). This finding has facilitated the identification and characterization of monoclonal antibodies (MoAbs) which recognize molecules expressed by a given leukocyte lineage (Lanier et al., 1987). To identify MoAbs that potentially recognize molecules of interest prior to rigorous functional and biochemical characterization, the patterns of expression for antigens detected by uncharacterized MoAbs are compared with those of known antigens. Once identified, candidate MoAbs are paired with MoAbs of confirmed specificity and examined by dual fluorochrome FMF to establish the relationship between the leukocyte lineages expressing the detected molecules. This strategy has been successfully employed by Lanier et al. (1987) to identify MoAbs that recognize molecules expressed on human T-lymphocytes, and in our laboratory {Davis et al., 1987) to identify MoAbs specific for both bovine major histocompatability complex (BoLA) gene products and leukocyte differentiation molecules. In this report we describe a set of T-cell reactive MoAbs identified and characterized by FMF. This set includes MoAbs specific for the CD2, CD4, CD6, and CD8 homologues (orthologues)* of domestic cattle, goats and sheep. MATERIAI,S AND M E T H O D S
Animals Holstein-Friesian cows were maintained at the Washington State University (WSU) dairy facilities, Saanen goats and mixed breed sheep at the WSU College of Veterinary Medicine, and pigs at the WSU Swine Research Center. These animals served as a source of peripheral blood mononuclear cells (PBM) for immunization and flow microfluorimetric (FMF) screening and analysis. BALB/c (H-2 d) mice were obtained from colonies maintained by the College of Veterinary Medicine animal care facilities. Monoclonal antibodies A list of the MoAbs employed as standards is shown in Table 1. The MoAbs specific for BoCD2, the bovine orthologue of the sheep red blood cell receptor (CD2) have been previously described {Lewin et al., 1985; Davis et al., 1988), as have the MoAbs used to identify monocyte, B-cell, N-cell, and granulocyte populations (Davis et al., 1987, 1990). The ovine- and caprine-specific MoAbs *The term "homology" is imprecise, as it does not distinguish molecules which diverge via speciation from the products of gene duplication which are maintained side by side within a lineage. The importance of this distinction becomes obvious when large, heterogeneous groups of related molecules (e.g. the immunoglobulin gene superfamily) are considered. For the sake of clarity, we have adopted the terminology of Fitch { 1970), wherein molecules that have diverged through speciation are defined as "orthologues" while those which result from gene duplication are termed "paralogues". For a recent discussion of this problem, see Goodman et al. ( 1987 ).
MoAbs REACTIVE WITH BOVINE,CAPRINE AND OVINET-I.YMPHOCYTE DETERMINANTS
197
TABLE 1 Monoclonal antibodies reactive with known leukocyte-expressed molecules MoAb
a-TH4A a-TH14B a-H42A a-B26A4 o~-CH61A a-CH128A ~ - I L A11 o~-II, A12 a - S B U T4 (x-GC1A ~ - I L A17 (x-BAGB25A ~-SBU T8 a-PIg45A2 o~-BAQ44A o~-BTA1 a-BAQ90A a-BAQ4A a-DH59B
Isotype
IgM IgG~a IgG.~a IgM IgG1 IgG1 IgGea IgG.,a IgGza IgG2a IgG~ IgM IgG~ IgG~, IgM IgM IgG:~ IgG, IgG,
Species reactivity Cattle
Sheep
Goat
+ + + + + + + + + + + + + + + +
+ + + + + + + + + + + +
+ + + + + + + + + + + + + +
Specificity
Reference
Class II M H C Class II M H C Class I! M H C CD2 CD2 CD2 CD4 CD4 CD4 CD4 CD8 CD8 CD8 IgM B2 N1 Nl-like N2 GM1
Davis et al., 1987 Davis et al., 1987 Davis et al., 1987 Lewin et al., 1985 Davis et al., 1988 Davis et al., 1988 Baldwin et al., 1986 Baldwin et al., 1986 Mackay et al., 1986 Davis et al., 1990 Ellis et al., 1986 Davis et al., 1990 Mackay et al., 1986 Davis et al., 1987 Davis et al., 1990 Davis et al., 1987 Davis et aI., 1990 Davis et al., 1990 Davis et al., 1987
T h e acronyms denote origin of the hybridoma cell line producing the monoclonal antibody and the regimen of i m m u n i z a t i o n used to generate the cell line: B = immunized with bovine peripheral blood mononuclear cells; H = immunized with equine, canine, caprine, murine (Rattus norvegicu~ ), and lagomorph (Oryctolagus cuniculu.~ ) peripheral blood mononuclear cells; T H = immunized with bovine, caprine, equine, a n d murine thymocytes. CH, DH = immunized with bovine caprine porcine, equine, canine, mustelid ( M u s t e l a vison ), a n d h u m a n peripheral blood mononuclear cells; P i g - - i m m u n i z e d with bovine a n d porcine IgM; BAQ, BAG, BAGB -- immunized with h u m a n T lymphocyte t u m o r ( J M ) cells, bovine, caprine, ovine, porcine, and equine peripheral blood mononuclear cells; C A C T = i m m u n i z e d with bovine Concanavalin A (Con A) stimulated peripheral blood mononuclear cells; BAT = immunized with caprine, ovine, bovine, and cervid (Cervu~ elaphus ) Con A stimulated lymphocytes a n d caprine, ovine, and bovine thymocytes. T h e MoAbs a IL A11, -IL A12, a n d -IL A17 were provided by W.I. Morrison; the MoAbs a - S B U T4 and -SBU T8 were obtained from M. B r a n d o n (see Methods section). M H C = major histocompatibility complex; T = T lymphocyte; CD = cluster of differentiation; B = B lymphocyte; N = N o n T / N o n B lymphocyte; G = granulocyte; M = monocyte; - = negative; + = positive.
a-GC1A1 and -BAGB25A have been described by Davis et al. (1990). AntiBoCD4 (a-IL A l l , -IL A12) and ~-BoCD8 MoAbs (~-IL A17) were provided by W.I. Morrison (International Laboratory for Research on Animal Diseases, Nairobi, Kenya) in the form of ascites (Baldwin et al., 1986; Ellis et al., 1986). Anti-SBU T4 and -SBU T8 (Maddox et al., 1985) were purchased from M.
198
R.A LARSEN ET AL.
Brandon (University of Melbourne, Parkville, Vic., Australia) as culture supernatants. The MoAbs used as standards in rosette inhibition assays have been described by Davis et al. (1988). The MoAbs described in this report were produced by established methods (Davis et al., 1983; Davis, 1988). Briefly, splenocytes were isolated from mice previously hyperimmunized with PBM of bovine, ovine, caprine, porcine, and cervid origin, then fused with X63 Ag8.653 myeloma cells. The culture supernatants of individual hybridomas derived in four such fusions were screened by FMF for reactivity with bovine and/or caprine leukocyte subsets. Cultures producing MoAbs of interest were cloned at high dilution in the presence of thymocytes and their supernatants reevaluated by FMF. Cloned lines were expanded in vitro, then injected into BALB/ c mice for production of immune ascites. Other hybridomas producing antibodies reactive with ruminant leukocytes were cryopreserved for future analysis. Rosette inhibition assay The ability of MoAb-containing culture supernatants and immune ascites to inhibit E-rosette formation was evaluated using 2-aminoethylisothiouronium (AET)-treated sheep erythrocytes as previously described (Davis et al., 1988). Flow microfluorimetry (FMF) Fresh PBM were isolated from bovine peripheral blood as previously described (Davis et al., 1987), and used fresh or stimulated with Concanavalin A (5/~g/ml) and maintained as long term cultures in DMEM (Gibco; Grand Island, NE) supplemented with 13% bovine calf serum and recombinant bovine interleukin 2 (rBoIL-2; provided by P. Baker, Immunex Corp., Seattle, WA ). Fresh and cultured cells were purified by density gradient centrifugation on a cushion of Lympho-paque T M ( a = 1.086: Nycomed AS, Oslo, Norway), washed once in phosphate buffered saline (PBS, pH 7.2) and resuspended to 107 cells/ml in PBS containing 10 m M EDTA, 0.1% Na azide, 10% ACD and 2% gamma-globulin-free horse serum. For comparative studies, fresh caprine and ovine leukocytes were obtained as buffy coats and processed as above. For single fluorescence experiments, 50/zl aliquots of cells were incubated with one MoAb, then indirectly labeled with fluorescein-conjugated goat c~-mouse Ig (heavy and light chain specific (Tago Inc., Burlingame, CA)) as previously described (Davis et al., 1987). For dual fluorescence experiments cells were incubated with one or two MoAbs (of different isotype) as above, reacted with biotinylated isotype-specific goat anti-mouse Ig (Southern Biotechnology Associates, Birmingham, AL), then labeled indirectly with phycoerythrin-conjugated strepavidin (Becton Dickinson immunocytometry systems, Mountain View, CA) and fluoresceinated isotype-specific goat anti-mouse Ig (Southern Biotechnology Associates) as described by Davis et al. {1987). Labeled cell preparations were then fixed for 30 min in 2% buffered formaldehyde. Fixed
MoAbs REACTIVEWITH BOVINE,CAPRINE AND OVINE T-LYMPHOCYTE DETERMINANTS
199
cells were subjected to dual parameter analysis with a Hg-Arc FACS T M Analyzer (Becton Dickinson Immunocytometry Systems, Mountain View, CA), with a minimum of 1500 data points collected per sample (Davis et al., 1987). TABLE 2 Monoclonal antibodies reactive with r u m i n a n t T cells MoAbs
Isotype
Species reactivity
Specificity
Cattle
Sheep
Goat
a-BAQ95A a-CACT31A a-CACT98A a-BAT18A a-BAT42A a-BAT76A a-MUC2A a-MUCllA a-BAQ82A a-BAQ91A a-CACT141A a-CACT33A a-CACT83A a-CACT87A a-CACT91E o~-CACT93A a-CACT132D a-CACT138A a-GC17A1 a-GC50A1 a-CACT80C
IgG, IgM IgM IgG, IgG~ lgG~ IgG2. IgM IgM IgG~ IgG2b IgM IgM IgM IgM IgM IgM IgG~ IgM IgM IgG~
÷ + + + + + + + + + + + (P ) + + + + + (P) + + + +
+ + -+ + +
÷ ÷ + + + + + + -+ + +
a-CACT85A
IgG,
+
-
-
CD2* CD2* CD2 CD2* CD2* CD2* CD2 CD2 CD6* CD6* CD6* CD4 CD4* CD4* CD4 CD4 CD4 CD4* CD4 CD4* CDS* CD8
a-CACT88C
IgGa
+
-
-
CD8*
a-CACTI30A
IgG,~
+
-
-
CD8*
a-BAQI 11A
IgM
+
-
+ (P)
CD8*
a-TH82DI
IgG,
+ (P)
+
+
CD8
a-BAT82A
IgG,
+
+
+
CD8*
*Assigned specificity has been confirmed in the Workshop on Bovine, Ovine and Caprine Leucocyte Differentiation Antigens, held in Hannover, F.R.G. on 6 July 1989. T h e acronyms denote origin of the hybridoma cell line producing the MoAb and the regimen of immunization used to generate the cell line: B A Q = immunized with h u m a n T lymphocyte t u m o r ( J M ) cells, bovine, caprine, ovine, porcine, and equine PBM; CACT = immunized with bovine Concanavalin A stimulated PBM; B A T = i m m u n i z e d with caprine, ovine, bovine, and cervid (Cervus elaphus) Concanavalin A stimulated lymphocytes and caprine, ovine, and bovine thymocytes; MUC = immunized with bovine, ovine, caprine lymphocytes, Concanavalin A stimulated ovine leukocytes, and porcine thymocytes; T H = immunized with bovine, caprine, equine, and murine {Rattus norvegicus } thymocytes. CD = cluster of differentiation; P = antigen polymorphic; - = negative; + = positive; PBM = peripheral blood mononuclear cells.
@
O'~
l t.,,~,~'. ,
:,~::-."•
•
D1
,:~
_~
li --
I :~::.t ,z'.;~;~, ~-;!. "
~.
,~ ,
~.,..;,:-
~;~¢.~': ..
....~ I~:
A1 i~:"
CD2
• ~:.~ , . , , -.
.! ~.
"
C3
A3
D2
• :--:-.'b~.
• • ,,~.~:,.,.,~
~.:~
"
!: . ~:'," ,-
i~.
. • , ~.~;, .
D4
C4]
Green Fluorescence
..'.,~i~i :i~.:.!,
1
N-CELL
~i
' t~ ~,~ ~.
:
CD4
•• L;h:!, ' ~ -
.,
D5
_:~!,.!~__ :.:,~,,~,...,__ .--~-~-.~,-.
D31 , .
c2Io
:!~~!~"~:'
: ¢'~:
le:
A2t f
CLASS [[
I.
....
~&~:.:: ,~'-:~;;,',~-:,-~,:'~ .
"i!:
CD8
m----~
"61
• A6
c~-BAQ1 1 1A
~-CACT91E
o~-CACT31A
>
MoAbsREACTIVEWITHBOVINE,CAPR1NEANDOVINET-I,YMPHOCYTEDETERMINANTS
201
RESULTS
Initial identification Supernatants from 534 cell lines derived from four fusions were screened by single fluorescence FMF. The FMF profiles generated with the MoAbs were compared to profiled obtained with MoAbs of known specificity for N-cell, Tcell, and T-cell subsets (data not shown). Monoclonal antibodies from 27 lines yielded patterns of interest (Table 2 ): 11 were identical to those obtained with known BoCD2-specific MoAbs (~-CH128A, -CH61A); 9 were identical to those obtained with BoCD4-specific MoAbs (c~-IL A l l , -IL A12); and 7 were identical to that obtained with a BoCD8-specific MoAb (a-IL A17). These lines were cloned and immune ascites produced for evaluation. The 27 MoAbs identified above were paired with antibodies that recognize known antigens (Table 1 ) and subjected to dual fluorescence analysis. Determinants detected by these MoAbs were not present on n o n T / n o n B (N)-cells (Fig. 1 ), B-lymphocytes expressing surface IgM, monocytes or granulocytes (data not shown ).
Monoclonal antibodies recognizing T-lymphocytes The 11 MoAbs identified as pan T-specific were paired with BoCD2-specific MoAbs of different Ig isotype (c~-B26A4, -CH61A, or -CH128A) for dual fluorescence analysis. Four of the MoAbs (a-CACT31A, -CACT141A, -BAQ82A, and -BAQ91A) co-stained the BoCD2 ÷ population, whereas varying degrees of competition in staining were noted between ~-BoCD2 controls and the other seven MoAbs (Anti-BAQ95A, -BAT18A, -BAT76A, -CACT98A, -MUC2A, and - M U C l l A ) . Competitive effects were also noted when the latter MoAbs were paired with each other, but not when paired with ~-CACT31A, -CACT141A, -BAQ82A or -BAQ91A (not shown). Anti-CACT31A co-stained with aCACT141A and -BAQ91A. Anti-CACT141A, -BAQ82A, and -BAQ91A exhibFig. 1. Representative profiles of PBM reacted with one or two MoAbs using indirect labeling with two fluorochromes (b-GaMIg-iso/streptavidin-PE (red fluorescence) and FITC-GaMIg-iso (green fluorescence ) ) as described under Methods. Panel A1 is a profile of cells reacted with streptavidinPE in the absence of antibody. The other panels in row A (A2-A6} are profiles of cells reacted with MoAbs of known specificity and stained with b-GaMIg-iso/streptavidin-PE.The other panels in row 1 (B1, C1, D1 ) are profiles of cells reacted with MoAbs reactive with T-lymphocytes and T-subsets, and stained with FITC-GaMIg-iso. The remaining panels (B2-B6, C2-C6, D2-D6) present profiles of cells reacted with MoAbs of known specificity and MoAbs reactive with Tlymphocytes and T-subsets. These were stained with both b-GaMIg-iso/streptavidin-PEand FITCGaMIg-iso. The MoAbs of known specificity used here were: panels A2-D2, a-CH61A (a-CD2); panels A3-D3, a-H42A (a-Class II MHC }; panels A4-D4, a-BAQ4A (a-N-cell); panels A5-D5, a-IL A12 (a-CD4); panels A6-D6, a-IL A17 (a-CD8). The MoAbs reactive with T-lymphocytes and T-subsets used here were: panels B1-B6, a-CACT31A (a-CD2); panels C1-C6, o~-CACT91E (a-CD4); panels D1-D6, a-BAQ1 l l A {a-CD8).
!
, • ,:.2..
I" . ..!~i
[~
"'~ "'~
..........
~,~.. ,,~,~::,
.
i
~ '
I
~"
,:, ~~ :
':: . : ~ '
c~!i:
B1
..
~'~4!
c~-CACT91E
.
•
-'~" '"~
... "~
, . .. ....
Green Fluorescence
]#". '
~
~.,;,i':
I
:
]
~
.
,_
I
" 0~-ILA12
I
o~-ILAI I
"'i ",,
cz-CACT93A
Fig. 2. Anti-CD4 MoAbs: profiles of PBM reacted with one or two MoAbs using indirect labeling with two fluorochromes (b-GaMIg-iso/streptavidin-PE (red fluorescence) and FITC-GaMlg-iso (green fluorescence)) as described under Methods. Panel A1 is a profile of cells reacted with streptavidin-PE in the absence of antibody. The other panels in row A (A2-A5) are profiles of cells reacted with MoAbs reactive with T-subsets and stained with b-GaMIg-iso/streptavidin-PE. The other panels in row 1 (B1, C1 ) are profiles of cells reacted with c~-CI)4 MoAbs, and stained with FITC-GaMIg-iso. The remaining panels (B2-B5, C2-C5) present profiles of cells reacted with MoAbs reactive with T-subsets and with ~CD4 MoAbs. These were stained with both b-GaMIg-iso/streptavidin-PE and FITC-GaMIg-iso. The MoAbs reactive with T-subsets used here were: panels A2-C2, c~-CACT83B; panels A3-C3, o~-CACT87B; panels A4-C4, ~-CACT91E; panels A5-.C5, ~-CACT93A. The c~-CD4 MoAbs used here were: panels B1-BS, o~-IL A11; panels C1-C5, c~-IL A12. All c~-CD4 MoAb combinations not shown produced profiles similar to the panels presented with the exception of a-GC 17A and CACT138A, which co-labeled.
rr
~
~~i
~
m
~!
O c
I " ~
c~-CACT87B
i~!~. I ~ _ ~ ~.~
I t ~;~,;~: I l,:)~,:.
(z-CACT83B
m > => ~ ~:~ ~ ~
[-o
MoAbs REACTIVE WITH BOVINE, CAPRINE AND OVINE T-LYMPHOCYTE DETERMINANq,'S
203
ited variable inhibition when paired with each other. E rosette formation was blocked by a-BoCD2 controls as well as by ~-CACT31A and the seven MoAbs competitive with standard ~-BoCD2 MoAbs. Anti-CACT141A, -BAQ82A, and -BAQ91A did not inhibit E rosette formation.
Monoclonal antibodies recognizing T-lymphocyte subsets The 16 MoAbs identified as reacting with determinants present on either BoCD4 ÷ or BoCD8 + cells were paired with MoAbs (of different Ig isotype) known to recognize BoCD2, BoCD4 (a-IL A11, -IL A12 ), BoCD8 (a-IL A17 ), and each other for dual fluorescence analysis (Figs. 1-3 ). All 16 MoAbs reacted with subsets of the BoCD2 ÷ population, with those identified as specific for BoCD4 ÷ cells (a-CACT33A,-CACT83A, -CACT87A, -CACT91C,-CACT93A, -CACT132D, -CACT138A, -GC17A, and -GC50A) labeling IL All +, IL A12 ÷ cells, and those identified as specific for BoCD8 + cells (a-CACT80C, -CACT85A, -CACT88A, -CACT130A, -BAQlllA, -TH82D1, and -BAT82A) labeling IL A17 ÷ cells. These two subsets were mutually exclusive and together represented 90-100% of the BoCD2 ÷ set (Fig. 1). The nine MoAbs that recognized BoCD4 + cells were paired with the BoCD4specific MoAbs a-IL A l l and -IL A12 in dual fluorescence analysis (Fig. 2). Competition was evident in all pairs examined, but most pronounced with aCACT83B, which completely blocked staining by both a - I L A l l and -IL A12 (Fig. 2). Comparisons between MoAbs were limited, as most shared the same isotype. Of the comparisons performed, co-staining without competition was evident only in the a-GC17A and -CACT138A combination (not shown). The seven MoAbs that recognized BoCD8 ÷ cells were paired with either the BoCD8-specific MoAb a - I L A17 or each other in dual fluorescence FMF analysis {Figs. 1 and 3). Anti-CACT88A (Fig. 3) and -BAQlllA (Fig. 1) co-stained the IL 17A ÷ population. Anti-BAQ111A similarly co-stained with a-TH82D1 and a-CACT130A (not shown). Other MoAb combinations examined displayed competitive binding, most evident in the blocking of a-TH82D1 and -CACT85A by a-CACT88A and of a - I L A17 and -TH82D1 by a-CACT130A (Fig. 3).
Cross-reactivity The 27 MoAbs identified above were evaluated against caprine and ovine PBM by single fluorescence FMF {Table 2). Nine of the 11 pan T-specific MoAbs labeled caprine PBM. Profiles of the eight MoAbs reactive with bovine CD2 and one pan T (~-BAQ91A) were identical to those obtained with bovine cells. Anti-CACT141A and -BAQ82A did not react with caprine cells. Two of the pan T-specific MoAbs (o~-BAQ82A and -BAQ91A) reacted with PBM from some but not all sheep. Of the MoAbs reactive with bovine CD4 + cells, two of nine (~-GC17A1 and -GC50A1) recognized determinants present on ovine and caprine PBM. Profiles obtained with these cross-reactive MoAbs were
-
.~
i.:!.';
C1
B1
":
. . . . . . . . .
:
•
C2
B2
A2
,
.
,.
.
, ..~
..
,
,,:
C3
B3
A3
(z-TH82D1
.... ":;:":.,-
~, ..
.
~
•
.
~
~,
'.
:~:i;
,.~? ' ~ ". :•~
~-
.
I
C41 r
B4
• .
I
.1 ~:~
,'
C5
B5
(z-CACT80C ~z-CACT85A A4I I ~5
Green Fluorescence
•
(z-ILA17
(z-CACT130A
(z-CACT88A
Fig. 3. Anti-CD8 MoAbs: profiles of P B M reacted with one or two MoAbs using indirect labeling with tw() fluorochromes (FITC-GaMlg-isotv3)e specific anti b - G a M l g - i s o / s t r e p t a v i d i n - P E ) . Panel A1 is a profile of cells reacted with s t r e p t a v i d i n - P E in the absence of ant b.~dy. T h e other panels in row A (A2-AS) are profiles of cells reacted with MoAbs reactive with T-subsets or an (~-(?D8 MoAb and stained with b - ( ; a M l g - i s o / streptavidin-PE. T h e other panels m r~)w 1 (B1, C1 ) are profiles of (:ells reacted with MoAbs reactive with T-subsets. and stained with FITC(;aMlg-iso. The remaining panels ( B2-B5, C2-C5 ) I)resent profiles of cells reacted wit h two MoAbs iof different isotypc ) reactive with T-subsets or one MoAb reactive with T-subsets and one MoAb reactive with CDS. These were stained with both b - G a M I g - i s o / s t r e p t o v i d i n - P E and FITC(;aMlg-iso. The MoAb reactive with CD8 was used in panels A2 (?2 ((~-IL A17 ). T h e MoAbs reactive with T-subsets used here were: panels A3C3, (~-'FH821)l: panels A4 C4, c~-CA(!TSOC: panels A5 C5, et-('A('T85A: panels B1 BS. ~t-CA(?'I'88A; panels C1 C5. (~-('ACT13OA. Reactions n(~t shown: labeling with t~-BAQ111A was partially blocked by -(?A('T80C. whereas -('ACT88C partially bh)cked -BAQ 11 IA (both similar to panel B3 ). "l'be M()Ab (t'-BAQI 11A c(.labeled with -('A(~'I'I:10A and -'1'H821)1 l a.~ per panel B2 ~and with -1[. A17 (Fig. 1. panel [)6 ).
O
O C-
A1
.
~ ;~ ~" ~" z >~ -
MoAbs REACTIVE WITH BOVINE, CAPRINE AND OVINE T-I,YMPHOCYTE DETERMINANTS
205
identical to those produced with MoAbs specific for ovine and caprine CD4 (a-SBU T4 and -GC1A, Table 1 ). Of the MoAbs reactive with bovine CD8 + cells, four of seven (a-CACT80C, -TH82D1, -BAT82A, and -BAQlllA) reacted with caprine cells and three of seven (a-CACT80C, -TH82D1, and -BAT82A)reacted with ovine PBM. The determinant recognized by aBAQ111A was polymorphic in goats. Profiles obtained with the cross-reactive MoAbs were identical to those produced by MoAbs that recognize caprine (aBAGB25A) or caprine and ovine (a-SBU T8) CD8. Cross-reactive MoAbs were paired with MoAbs of known specificity and subjected to dual fluorescence FMF analysis with caprine and ovine PBL (not shown). Determinants detected by these MoAbs were absent from cells bearing IgM, N-cell, granulocyte or monocyte-specific determinants. Cross-reactive pan T-specific MoAbs either co-stained B26A4 + caprine cells (aBAQ91A), or competed with a-B26A4. The cross-reactive MoAbs that recognized bovine CD4 ÷ cells labeled a subset of caprine B26A4 ÷, BAQ91A ÷ cells and a subset of BAQ82A ÷, BAQ91A ÷ ovine cells, and competed with both aSBU T4 and -GC1A on both caprine and ovine cells. The cross-reactive MoAbs that recognized bovine CD8 + cells labeled a subset of caprine B26A4 ÷, BAQ91A + cells and a subset of BAQ82A ÷, BAQ91A ÷ ovine cells. These MoAbs competed with both a-SBU T8 and -BAGB25A on caprine cells, and competed with a-SBU T8 on ovine cells. The T-cell subsets recognized by these crossreactive MoAbs were mutually exclusive and together represented 90-100% of the B26A4 +, BAQ91A ÷ cells (caprine) or the BAQ82A ÷, BAQ91A ÷ cells (ovine).
Polymorphic determinants Profiles generated using PBM from 30 individual cattle revealed that at least three of the determinants detected (CACT33A, CACT132D, and TH82D1) were polymorphic. The screening of a - B A Q l l l A on caprine PBM revealed that its respective determinant is polymorphic in goats. Examination of over 50 animals has provided no evidence for a similar polymorphism in cattle. DISCUSSION Comparative analysis of FMF profiles obtained with MoAbs from primary cultures of hybridomas with those obtained using MoAbs of known specificity is an effective technique for distinguishing MoAbs that react with the same or new molecules. As demonstrated here, 27 MoAbs reactive with bovine T cell determinants were correctly identified 534 supernatants containing antibodies reactive with bovine leukocytes. This technique has also been used to identify MoAbs reactive with bovine N cells and B cells (Davis et al., 1987) and with goat and sheep PBM subsets (Davis et al., 1990). In this study 11 MoAbs were identified that reacted with determinants pres-
206
R.A. I ~ R S E N E T AI,.
ent on all mature T-cells. Competitive binding and inhibition of E rosette formation demonstrated that eight of these MoAbs (ce-BAT18A, -BAT42A, BAT76A, -BAQ95A, -CACT31A, -CACT98A, -MUC2A and - M U C l l A ) recognize determinants present on the BoCD2 molecule. All of these MoAbs displayed competitive binding when paired with other ce-BoCD2 MoAbs in dual fluorescence FMF. The finding that c~-MUCllA is not reactive with the caprine CD2 orthologue indicates that the determinant detected by this MoAb is structurally distinct from the determinants detected by the other ~-BoCD2 MoAbs. Three pan T MoAbs (c~-BAQ82A, -BAQ91A and -CACT141A) did not inhibit E rosette formation or compete with a-BoCD2 MoAbs. Although these MoAbs compete with each other, their patterns of cross-reactivity indicate that each MoAb recognizes a different determinant on the same molecule or molecular complex (Table 2). Antibodies that recognize human CD2, but do not block E rosette formation have been described {Meuer et al., 1984), however, Baldwin (personal communication) has found that c~-BAQ91A and -CACT141A compete with the MoAbs c~-IL A27 and -IL A28, which indicates that this set of determinants is present on the orthologue of CD6. This supposition is supported by data obtained at a recent workshop (see below). Nine MoAbs were identified that reacted with determinants restricted to BoCD4 ÷ cells (~-CACT33A,-CACT83A,-CACT87A,-CACT91C,-CACT93A, -CACT132D, -CACT138A, oGC17A, and -GC50A). All MoAbs in this set displayed competitive behavior in one or more combinations when paired with each other or with MoAbs known to react with BoCD4 (c~-IL A l l and -IL A12), indicating that determinants recognized by these MoAbs are present on the BoCD4 molecule. Two of these determinants (GC17A and GC50A) were also present on caprine and ovine T-lymphocyte subsets. This pattern of crossreactivity indicates that determinants are distinct from those recognized by other members of the set and from the GC1A and SBU T4 determinants, which are absent in cattle (Tables 1, 2). Seven MoAbs were identified that reacted with determinants restricted to BoCD8 ÷ cells (c~-BAQ111A,-CACT80C,-CACT85A,-CACT88C,-CACT130A, -TH82D1, and BAT82A). The MoAbs in this set either competed with ~-IL A17 or each other in one or more combinations, indicating the determinants recognized by these MoAbs are present on the BoCD8 molecule. Two MoAbs, c~-BAQlllA and -CACT88A, co-stained with c~-IL A17 and -TH82D1, indicating that their respective determinants are spatially removed from the IL A17 and TH82D1 determinants. Because CD8 is a heterodimeric molecule in rats (Johnson et al., 1985), humans (Barclay et al., 1987; Spurr et al., 1988) and perhaps cattle {Ellis et al., 1986), there exists the additional possibility that the determinants detected by a-BAQ111A and -CACT88A are not present on the peptide bearing the determinants detected by c~-IL A17 and -TH82D1. Regardless of their location, the B A Q l l l A and CACT88A determinants are not identical, as indicated by conservation of the BAQlllA, but not the
MoAbs REACTIVE WITH BOVINE, CAPRINE AND OVINE T-LYMPHOCYTE DETERMINANTS
207
CACT88A determinant in goats (Table 2). Similarly, patterns of cross-reactivity indicate that CACT80C and BAT82A are distinct from other determinants, as is TH82D1 - based on both cross-reactivity and polymorphism. All determinants recognized by this MoAb set are distinct from BAGB25A, which has only been detected on caprine lymphocytes {Table 2 ). Monoclonal antibodies which define BoCD4 (a-IL A11, -IL A12) and BoCD8 (a-IL A17) have been rigorously characterized (Baldwin et al., 1986; Ellis et al., 1986). The determinants recognized by these MoAbs are not broadly conserved among ruminants. Maddox et al. (1985) have produced two MoAbs (aSBU T4, -SBU T8) which identify the orthologous molecules in sheep. Both determinants are present in goats, but only one (SBU T8) occurs in cattle (Mackay et al., 1986). The cross-reactivity patterns of MoAbs described in this report and of MoAbs used here and described elsewhere {Davis et al., 1989) support the contention (Mackay, 1988) that the molecules identified as BoT4 in cattle and as SBU T4 in sheep and goats are orthologues, as are the molecules identified as BoT8 in cattle, and as SBU T8 in sheep and goats. Seventeen of the 27 MoAbs described in this study (Table 2 ) were submitted to the Workshop on Bovine, Ovine and Caprine Leucocyte Differentiation Antigens {summary in preparation). Cluster analysis and determinations of apparent molecular weights indicate that the specificities for all 17 MoAbs submitted were identical to those we had assigned based on the information presented in this study. Preliminary studies indicate that the strategy employed in the present study will be useful in identifying orthologous leukocyte differentiation molecules of other domestic and wild species {Davis et al., 1990). ACKNOWLEDGEMENTS The technical assistance of J. Davis, L. Froseth, L. Huston and S. Vibber is acknowledged. The authors thank Dr. David Reduker for his critical comments during the preparation of the manuscript. This research was supported in part by grants from the USDA-SEA (82-CRSR-2-2045, 83-CRSR-2-2281, 84-CRSR2-2457, 86-CRSR-2-2913, 86-CRCR-1-2241 ), USDA-CSRS (89-37265-4537), NIH (CA50141-01), WSU-ARC (project 3073), the Washington Technology Center, and the Carnation Research Farm. R.A. Larsen was supported by a National Services award (A106 07025 ) from the National Institutes of Allergy and Infectious Diseases. Y.H. Park was supported by the exchange scientist student program of the Rockefeller Foundation. REFERENCES
Baldwin, C.L., Teale, A.J., Naessens, J.G., Goddeeris, B.M., MacHugh, N.D. and Morrison, W.l., 1986. Characterization of a subset of bovine T lymphocytes that express BoT4 by monoclonal antibodies and function: Similarity to lymphocytes defined by human T4 and murine L3T4. J. Immunol., 136: 4385-4391.
208
R.A.LARSENET AI..
Barclay, A.N., Johnson, P., McCaughan, G.W. and Williams, A.F., 1987. Immunoglobulin-related structures associated with vertebrate cell surfaces. In: T.W. Mak (Editor), The T Cell Receptors. Plenum Press, NY, pp. 53-87. Davis, W.C., 1988. Enhancement of myeloma-B-cell hybridoma outgrowth in primary cultures with B cell mitogens. Period. Biol., 90: 367-374. Davis, W.C., McGuire, T.C. and Perryman, L.E., 1983. Biomedical and biological application of monoclonal antibody technology in developing countries. Period. Biol., 85: 259-282. Davis, W.C., Marusic, S., Lewin, H.A., Splitter, G.A., Perryman, L.E., McGuire, T.C. and Gorham, J.R., 1987. The development and analysis of species-specific and cross-reactive monoclonal antibodies to leukocyte differentiation antigens and antigens of the major histocompatibility complex for use in the study of the immune system in cattle and other species. VeL Immunol. Immunopathol., 15: 337-376. Davis, W.C., Ellis, J.A., MacHugh, N.D. and Baldwin, C.L., 1988. Bovine pan T-cell monoclonal antibodies reactive with a molecule similar to CD2. Immunology, 62: 165--167. Davis, W.C.. Larsen, R.A. and Monaghan, M.L., 1990. Genetic markers identified by immunogenetic methods. International Symposium and Educational Workshop on Fish-marking Techniques, in press. Ellis, J.A., Baldwin, C.L., MacHugh, N.D., Bensaid, A., Teale, A.J., Goddeeris, B.M. and Morrison, W.I., 1986. Characterization by a monoclonal antibody and functional analysis of a subset of bovine T lymphocytes that express BoT8, a molecule analogous to human CD8. Immunology, 58: 351-358. Fitch, W.M., 1970. Distinguishing homologous from analogous proteins. Syst. ZooL, 19: 99-113. Goodman, M., Miyamoto, M.M., Czelusniak, J., 1987. Pattern and process in vertebrate phylogeny revealed by coevolution of molecules and morphologies. In: C. Patterson (Editor), Molecules and Morphology in Evolution: Conflict or Compromise? Cambridge University Press, Cambridge, pp. 142- 176. Johnson, P., Gagnon, J., Barclay, A.N. and Williams, A.F., 1985. Purification, chain separation and sequence of the MRC OX-8 antigen, a marker of rat cytotoxic T lymphocytes. EMBO J.. 4: 2539-2545. I,anier, I,.l,., Engleman, E.G., Gatenby, P., Babcock, G.F., Warner, N.L. and Herzenberg, I,.A.. 1983. Correlation of functional properties of human lymphoid cell subsets in surface marker phenotypes using multiparameter analysis and flow cytometry. Immunol. Reviews.. 74: 143160. Lanier, L.L., Le, A.M., Ding, A.H. and Evans, E.I,., 1987. Analysis of the workshop T-cell monoclonal antibodies by 'indirect two-colour immunofluorescence' and multiparameter flow cytometry. In: A.J. McMichael (Editor), I~ukocyte Typing III. Oxford University Press, New York, NY, pp. 62-65. I,ewin, H.A., Davis, W.C. and Bernoco, D., 1985. Monoclonal antibodies that distinguish bovine T and B lymphocytes. Vet. Immunol. Immunopathol., 9: 87-102. Mackay, C.R., 1988. Sheep leukocyte molecules: A review of their distribution, structure and possible function. Vet. Immunol. Immunopathol., 19: 1-20. Mackay, C.R., Maddox, J.F. and Brandon, M.R., 1986. Three distinct subpopulations of sheep T lymphocytes. Eur. J. Immunol., 16: 19-25. Maddox, F., Mackay, C.R. and Brandon, M.R., 1985. Surface antigens, SBU-T4 and SBU-T8, of sheep T lymphocyte subsets defined by monoclonal antibodies. Immunology, 55: 739-748. Meuer, S.C., Hussey, R.E., Fabbi, M., Fox, D., Acuto, D., Fitzgerald, K.A., Hodgdon, J.C., Protentis, J.P., Schlossman, S.F. and Reinherz, E.L., 1984. An alternative pathway of T-cell activation: A functional role for the 50 kD T11 sheep erythrocyte receptor protein. Cell, 36: 89790x. Spurt, N.K., Goodfelh)w, P.N. and Johnson, P., 1988. CD8B, the human equivalent of the mouse Ly-3 gene is localized on chromosome 2. Immunogenetics, 27: 70-72.
195
Identification and Characterization of Monoclonal Antibodies Reactive with Bovine, Caprine and Ovine T-Lymphocyte Determinants by Flow Microfluorimetry R.A. LARSEN, M.L. MONAGHAN', Y.H. PARK, M.J. HAMILTON, J.A. ELLIS" and W.C. DAVIS*
Department of Veterinary Microbiology and Pathology, College of Veterinary Medicine, Washington St. University, Pullman, WA 99164-7040(U.S.A.) 'Department of Farm Animal Clinical Studies, Faculty of Veterinary Medicine, University College Dublin, Ballsbridge Dublin 4 (Ireland) '~Departments of Veterinary Sciences and Molecular Biology, University of Wyoming, Laramie, WY (U.S.A.) {Accepted 23 November 1989)
ABSTRACT Larsen, R.A., Monaghan, M.L., Park, Y.H., Hamilton, M.J., Ellis, J.A. and Davis, W.C., 1990. Identification and characterization of monoclonal antibodies reactive with bovine, caprine and ()vine T-lymphocyte determinants by flow microfluorimetry. Vet. Immunol. Immunopathol., 25: 195-208. Comparative flow microfluorimetric (FMF) analysis was used to identify and characterize 27 monoclonal antibodies (MoAbs) reactive with bovine T-lymphocytes. Determinants present on all circulating T-lymphocytes were recognized by 11 MoAbs, 8 of which blocked E rosette formation. Determinants present on only the BoCD4 + T-lymphocyte subset were detected by 9 MoAbs, while determinants restricted to the BoCD8 ÷ T-lymphocyte subset were recognized by 7 MoAbs. Competitive labeling experiments demonstrated that determinants recognized by subset-specific MoAbs were present on BoCD4 or BoCD8 molecules. Comparative studies revealed that some determinants, both pan-T specific and subset-specific, were conserved on homologous (orthologous) molecules expressed on leukocytes from other species of ruminants. Polymorphism was evident with several determinants.
INTRODUCTION Extensive studies of leukocyte differentiation antigens by flow microfluori-
merry (FMF) have demonstrated that many leukocyte surface molecules ex*Towhomcorrespondence shouldbe addressed.
0165-2427/90/$03.50
© 1990 Elsevier Science Publishers B.V.
196
R.A. L A R S E N E T AL.
hibit unique patterns of expression on one or more lineages of leukocytes (Lanier et al., 1983). This finding has facilitated the identification and characterization of monoclonal antibodies (MoAbs) which recognize molecules expressed by a given leukocyte lineage (Lanier et al., 1987). To identify MoAbs that potentially recognize molecules of interest prior to rigorous functional and biochemical characterization, the patterns of expression for antigens detected by uncharacterized MoAbs are compared with those of known antigens. Once identified, candidate MoAbs are paired with MoAbs of confirmed specificity and examined by dual fluorochrome FMF to establish the relationship between the leukocyte lineages expressing the detected molecules. This strategy has been successfully employed by Lanier et al. (1987) to identify MoAbs that recognize molecules expressed on human T-lymphocytes, and in our laboratory {Davis et al., 1987) to identify MoAbs specific for both bovine major histocompatability complex (BoLA) gene products and leukocyte differentiation molecules. In this report we describe a set of T-cell reactive MoAbs identified and characterized by FMF. This set includes MoAbs specific for the CD2, CD4, CD6, and CD8 homologues (orthologues)* of domestic cattle, goats and sheep. MATERIAI,S AND M E T H O D S
Animals Holstein-Friesian cows were maintained at the Washington State University (WSU) dairy facilities, Saanen goats and mixed breed sheep at the WSU College of Veterinary Medicine, and pigs at the WSU Swine Research Center. These animals served as a source of peripheral blood mononuclear cells (PBM) for immunization and flow microfluorimetric (FMF) screening and analysis. BALB/c (H-2 d) mice were obtained from colonies maintained by the College of Veterinary Medicine animal care facilities. Monoclonal antibodies A list of the MoAbs employed as standards is shown in Table 1. The MoAbs specific for BoCD2, the bovine orthologue of the sheep red blood cell receptor (CD2) have been previously described {Lewin et al., 1985; Davis et al., 1988), as have the MoAbs used to identify monocyte, B-cell, N-cell, and granulocyte populations (Davis et al., 1987, 1990). The ovine- and caprine-specific MoAbs *The term "homology" is imprecise, as it does not distinguish molecules which diverge via speciation from the products of gene duplication which are maintained side by side within a lineage. The importance of this distinction becomes obvious when large, heterogeneous groups of related molecules (e.g. the immunoglobulin gene superfamily) are considered. For the sake of clarity, we have adopted the terminology of Fitch { 1970), wherein molecules that have diverged through speciation are defined as "orthologues" while those which result from gene duplication are termed "paralogues". For a recent discussion of this problem, see Goodman et al. ( 1987 ).
MoAbs REACTIVE WITH BOVINE,CAPRINE AND OVINET-I.YMPHOCYTE DETERMINANTS
197
TABLE 1 Monoclonal antibodies reactive with known leukocyte-expressed molecules MoAb
a-TH4A a-TH14B a-H42A a-B26A4 o~-CH61A a-CH128A ~ - I L A11 o~-II, A12 a - S B U T4 (x-GC1A ~ - I L A17 (x-BAGB25A ~-SBU T8 a-PIg45A2 o~-BAQ44A o~-BTA1 a-BAQ90A a-BAQ4A a-DH59B
Isotype
IgM IgG~a IgG.~a IgM IgG1 IgG1 IgGea IgG.,a IgGza IgG2a IgG~ IgM IgG~ IgG~, IgM IgM IgG:~ IgG, IgG,
Species reactivity Cattle
Sheep
Goat
+ + + + + + + + + + + + + + + +
+ + + + + + + + + + + +
+ + + + + + + + + + + + + +
Specificity
Reference
Class II M H C Class II M H C Class I! M H C CD2 CD2 CD2 CD4 CD4 CD4 CD4 CD8 CD8 CD8 IgM B2 N1 Nl-like N2 GM1
Davis et al., 1987 Davis et al., 1987 Davis et al., 1987 Lewin et al., 1985 Davis et al., 1988 Davis et al., 1988 Baldwin et al., 1986 Baldwin et al., 1986 Mackay et al., 1986 Davis et al., 1990 Ellis et al., 1986 Davis et al., 1990 Mackay et al., 1986 Davis et al., 1987 Davis et al., 1990 Davis et al., 1987 Davis et aI., 1990 Davis et al., 1990 Davis et al., 1987
T h e acronyms denote origin of the hybridoma cell line producing the monoclonal antibody and the regimen of i m m u n i z a t i o n used to generate the cell line: B = immunized with bovine peripheral blood mononuclear cells; H = immunized with equine, canine, caprine, murine (Rattus norvegicu~ ), and lagomorph (Oryctolagus cuniculu.~ ) peripheral blood mononuclear cells; T H = immunized with bovine, caprine, equine, a n d murine thymocytes. CH, DH = immunized with bovine caprine porcine, equine, canine, mustelid ( M u s t e l a vison ), a n d h u m a n peripheral blood mononuclear cells; P i g - - i m m u n i z e d with bovine a n d porcine IgM; BAQ, BAG, BAGB -- immunized with h u m a n T lymphocyte t u m o r ( J M ) cells, bovine, caprine, ovine, porcine, and equine peripheral blood mononuclear cells; C A C T = i m m u n i z e d with bovine Concanavalin A (Con A) stimulated peripheral blood mononuclear cells; BAT = immunized with caprine, ovine, bovine, and cervid (Cervu~ elaphus ) Con A stimulated lymphocytes a n d caprine, ovine, and bovine thymocytes. T h e MoAbs a IL A11, -IL A12, a n d -IL A17 were provided by W.I. Morrison; the MoAbs a - S B U T4 and -SBU T8 were obtained from M. B r a n d o n (see Methods section). M H C = major histocompatibility complex; T = T lymphocyte; CD = cluster of differentiation; B = B lymphocyte; N = N o n T / N o n B lymphocyte; G = granulocyte; M = monocyte; - = negative; + = positive.
a-GC1A1 and -BAGB25A have been described by Davis et al. (1990). AntiBoCD4 (a-IL A l l , -IL A12) and ~-BoCD8 MoAbs (~-IL A17) were provided by W.I. Morrison (International Laboratory for Research on Animal Diseases, Nairobi, Kenya) in the form of ascites (Baldwin et al., 1986; Ellis et al., 1986). Anti-SBU T4 and -SBU T8 (Maddox et al., 1985) were purchased from M.
198
R.A LARSEN ET AL.
Brandon (University of Melbourne, Parkville, Vic., Australia) as culture supernatants. The MoAbs used as standards in rosette inhibition assays have been described by Davis et al. (1988). The MoAbs described in this report were produced by established methods (Davis et al., 1983; Davis, 1988). Briefly, splenocytes were isolated from mice previously hyperimmunized with PBM of bovine, ovine, caprine, porcine, and cervid origin, then fused with X63 Ag8.653 myeloma cells. The culture supernatants of individual hybridomas derived in four such fusions were screened by FMF for reactivity with bovine and/or caprine leukocyte subsets. Cultures producing MoAbs of interest were cloned at high dilution in the presence of thymocytes and their supernatants reevaluated by FMF. Cloned lines were expanded in vitro, then injected into BALB/ c mice for production of immune ascites. Other hybridomas producing antibodies reactive with ruminant leukocytes were cryopreserved for future analysis. Rosette inhibition assay The ability of MoAb-containing culture supernatants and immune ascites to inhibit E-rosette formation was evaluated using 2-aminoethylisothiouronium (AET)-treated sheep erythrocytes as previously described (Davis et al., 1988). Flow microfluorimetry (FMF) Fresh PBM were isolated from bovine peripheral blood as previously described (Davis et al., 1987), and used fresh or stimulated with Concanavalin A (5/~g/ml) and maintained as long term cultures in DMEM (Gibco; Grand Island, NE) supplemented with 13% bovine calf serum and recombinant bovine interleukin 2 (rBoIL-2; provided by P. Baker, Immunex Corp., Seattle, WA ). Fresh and cultured cells were purified by density gradient centrifugation on a cushion of Lympho-paque T M ( a = 1.086: Nycomed AS, Oslo, Norway), washed once in phosphate buffered saline (PBS, pH 7.2) and resuspended to 107 cells/ml in PBS containing 10 m M EDTA, 0.1% Na azide, 10% ACD and 2% gamma-globulin-free horse serum. For comparative studies, fresh caprine and ovine leukocytes were obtained as buffy coats and processed as above. For single fluorescence experiments, 50/zl aliquots of cells were incubated with one MoAb, then indirectly labeled with fluorescein-conjugated goat c~-mouse Ig (heavy and light chain specific (Tago Inc., Burlingame, CA)) as previously described (Davis et al., 1987). For dual fluorescence experiments cells were incubated with one or two MoAbs (of different isotype) as above, reacted with biotinylated isotype-specific goat anti-mouse Ig (Southern Biotechnology Associates, Birmingham, AL), then labeled indirectly with phycoerythrin-conjugated strepavidin (Becton Dickinson immunocytometry systems, Mountain View, CA) and fluoresceinated isotype-specific goat anti-mouse Ig (Southern Biotechnology Associates) as described by Davis et al. {1987). Labeled cell preparations were then fixed for 30 min in 2% buffered formaldehyde. Fixed
MoAbs REACTIVEWITH BOVINE,CAPRINE AND OVINE T-LYMPHOCYTE DETERMINANTS
199
cells were subjected to dual parameter analysis with a Hg-Arc FACS T M Analyzer (Becton Dickinson Immunocytometry Systems, Mountain View, CA), with a minimum of 1500 data points collected per sample (Davis et al., 1987). TABLE 2 Monoclonal antibodies reactive with r u m i n a n t T cells MoAbs
Isotype
Species reactivity
Specificity
Cattle
Sheep
Goat
a-BAQ95A a-CACT31A a-CACT98A a-BAT18A a-BAT42A a-BAT76A a-MUC2A a-MUCllA a-BAQ82A a-BAQ91A a-CACT141A a-CACT33A a-CACT83A a-CACT87A a-CACT91E o~-CACT93A a-CACT132D a-CACT138A a-GC17A1 a-GC50A1 a-CACT80C
IgG, IgM IgM IgG, IgG~ lgG~ IgG2. IgM IgM IgG~ IgG2b IgM IgM IgM IgM IgM IgM IgG~ IgM IgM IgG~
÷ + + + + + + + + + + + (P ) + + + + + (P) + + + +
+ + -+ + +
÷ ÷ + + + + + + -+ + +
a-CACT85A
IgG,
+
-
-
CD2* CD2* CD2 CD2* CD2* CD2* CD2 CD2 CD6* CD6* CD6* CD4 CD4* CD4* CD4 CD4 CD4 CD4* CD4 CD4* CDS* CD8
a-CACT88C
IgGa
+
-
-
CD8*
a-CACTI30A
IgG,~
+
-
-
CD8*
a-BAQI 11A
IgM
+
-
+ (P)
CD8*
a-TH82DI
IgG,
+ (P)
+
+
CD8
a-BAT82A
IgG,
+
+
+
CD8*
*Assigned specificity has been confirmed in the Workshop on Bovine, Ovine and Caprine Leucocyte Differentiation Antigens, held in Hannover, F.R.G. on 6 July 1989. T h e acronyms denote origin of the hybridoma cell line producing the MoAb and the regimen of immunization used to generate the cell line: B A Q = immunized with h u m a n T lymphocyte t u m o r ( J M ) cells, bovine, caprine, ovine, porcine, and equine PBM; CACT = immunized with bovine Concanavalin A stimulated PBM; B A T = i m m u n i z e d with caprine, ovine, bovine, and cervid (Cervus elaphus) Concanavalin A stimulated lymphocytes and caprine, ovine, and bovine thymocytes; MUC = immunized with bovine, ovine, caprine lymphocytes, Concanavalin A stimulated ovine leukocytes, and porcine thymocytes; T H = immunized with bovine, caprine, equine, and murine {Rattus norvegicus } thymocytes. CD = cluster of differentiation; P = antigen polymorphic; - = negative; + = positive; PBM = peripheral blood mononuclear cells.
@
O'~
l t.,,~,~'. ,
:,~::-."•
•
D1
,:~
_~
li --
I :~::.t ,z'.;~;~, ~-;!. "
~.
,~ ,
~.,..;,:-
~;~¢.~': ..
....~ I~:
A1 i~:"
CD2
• ~:.~ , . , , -.
.! ~.
"
C3
A3
D2
• :--:-.'b~.
• • ,,~.~:,.,.,~
~.:~
"
!: . ~:'," ,-
i~.
. • , ~.~;, .
D4
C4]
Green Fluorescence
..'.,~i~i :i~.:.!,
1
N-CELL
~i
' t~ ~,~ ~.
:
CD4
•• L;h:!, ' ~ -
.,
D5
_:~!,.!~__ :.:,~,,~,...,__ .--~-~-.~,-.
D31 , .
c2Io
:!~~!~"~:'
: ¢'~:
le:
A2t f
CLASS [[
I.
....
~&~:.:: ,~'-:~;;,',~-:,-~,:'~ .
"i!:
CD8
m----~
"61
• A6
c~-BAQ1 1 1A
~-CACT91E
o~-CACT31A
>
MoAbsREACTIVEWITHBOVINE,CAPR1NEANDOVINET-I,YMPHOCYTEDETERMINANTS
201
RESULTS
Initial identification Supernatants from 534 cell lines derived from four fusions were screened by single fluorescence FMF. The FMF profiles generated with the MoAbs were compared to profiled obtained with MoAbs of known specificity for N-cell, Tcell, and T-cell subsets (data not shown). Monoclonal antibodies from 27 lines yielded patterns of interest (Table 2 ): 11 were identical to those obtained with known BoCD2-specific MoAbs (~-CH128A, -CH61A); 9 were identical to those obtained with BoCD4-specific MoAbs (c~-IL A l l , -IL A12); and 7 were identical to that obtained with a BoCD8-specific MoAb (a-IL A17). These lines were cloned and immune ascites produced for evaluation. The 27 MoAbs identified above were paired with antibodies that recognize known antigens (Table 1 ) and subjected to dual fluorescence analysis. Determinants detected by these MoAbs were not present on n o n T / n o n B (N)-cells (Fig. 1 ), B-lymphocytes expressing surface IgM, monocytes or granulocytes (data not shown ).
Monoclonal antibodies recognizing T-lymphocytes The 11 MoAbs identified as pan T-specific were paired with BoCD2-specific MoAbs of different Ig isotype (c~-B26A4, -CH61A, or -CH128A) for dual fluorescence analysis. Four of the MoAbs (a-CACT31A, -CACT141A, -BAQ82A, and -BAQ91A) co-stained the BoCD2 ÷ population, whereas varying degrees of competition in staining were noted between ~-BoCD2 controls and the other seven MoAbs (Anti-BAQ95A, -BAT18A, -BAT76A, -CACT98A, -MUC2A, and - M U C l l A ) . Competitive effects were also noted when the latter MoAbs were paired with each other, but not when paired with ~-CACT31A, -CACT141A, -BAQ82A or -BAQ91A (not shown). Anti-CACT31A co-stained with aCACT141A and -BAQ91A. Anti-CACT141A, -BAQ82A, and -BAQ91A exhibFig. 1. Representative profiles of PBM reacted with one or two MoAbs using indirect labeling with two fluorochromes (b-GaMIg-iso/streptavidin-PE (red fluorescence) and FITC-GaMIg-iso (green fluorescence ) ) as described under Methods. Panel A1 is a profile of cells reacted with streptavidinPE in the absence of antibody. The other panels in row A (A2-A6} are profiles of cells reacted with MoAbs of known specificity and stained with b-GaMIg-iso/streptavidin-PE.The other panels in row 1 (B1, C1, D1 ) are profiles of cells reacted with MoAbs reactive with T-lymphocytes and T-subsets, and stained with FITC-GaMIg-iso. The remaining panels (B2-B6, C2-C6, D2-D6) present profiles of cells reacted with MoAbs of known specificity and MoAbs reactive with Tlymphocytes and T-subsets. These were stained with both b-GaMIg-iso/streptavidin-PEand FITCGaMIg-iso. The MoAbs of known specificity used here were: panels A2-D2, a-CH61A (a-CD2); panels A3-D3, a-H42A (a-Class II MHC }; panels A4-D4, a-BAQ4A (a-N-cell); panels A5-D5, a-IL A12 (a-CD4); panels A6-D6, a-IL A17 (a-CD8). The MoAbs reactive with T-lymphocytes and T-subsets used here were: panels B1-B6, a-CACT31A (a-CD2); panels C1-C6, o~-CACT91E (a-CD4); panels D1-D6, a-BAQ1 l l A {a-CD8).
!
, • ,:.2..
I" . ..!~i
[~
"'~ "'~
..........
~,~.. ,,~,~::,
.
i
~ '
I
~"
,:, ~~ :
':: . : ~ '
c~!i:
B1
..
~'~4!
c~-CACT91E
.
•
-'~" '"~
... "~
, . .. ....
Green Fluorescence
]#". '
~
~.,;,i':
I
:
]
~
.
,_
I
" 0~-ILA12
I
o~-ILAI I
"'i ",,
cz-CACT93A
Fig. 2. Anti-CD4 MoAbs: profiles of PBM reacted with one or two MoAbs using indirect labeling with two fluorochromes (b-GaMIg-iso/streptavidin-PE (red fluorescence) and FITC-GaMlg-iso (green fluorescence)) as described under Methods. Panel A1 is a profile of cells reacted with streptavidin-PE in the absence of antibody. The other panels in row A (A2-A5) are profiles of cells reacted with MoAbs reactive with T-subsets and stained with b-GaMIg-iso/streptavidin-PE. The other panels in row 1 (B1, C1 ) are profiles of cells reacted with c~-CI)4 MoAbs, and stained with FITC-GaMIg-iso. The remaining panels (B2-B5, C2-C5) present profiles of cells reacted with MoAbs reactive with T-subsets and with ~CD4 MoAbs. These were stained with both b-GaMIg-iso/streptavidin-PE and FITC-GaMIg-iso. The MoAbs reactive with T-subsets used here were: panels A2-C2, c~-CACT83B; panels A3-C3, o~-CACT87B; panels A4-C4, ~-CACT91E; panels A5-.C5, ~-CACT93A. The c~-CD4 MoAbs used here were: panels B1-BS, o~-IL A11; panels C1-C5, c~-IL A12. All c~-CD4 MoAb combinations not shown produced profiles similar to the panels presented with the exception of a-GC 17A and CACT138A, which co-labeled.
rr
~
~~i
~
m
~!
O c
I " ~
c~-CACT87B
i~!~. I ~ _ ~ ~.~
I t ~;~,;~: I l,:)~,:.
(z-CACT83B
m > => ~ ~:~ ~ ~
[-o
MoAbs REACTIVE WITH BOVINE, CAPRINE AND OVINE T-LYMPHOCYTE DETERMINANq,'S
203
ited variable inhibition when paired with each other. E rosette formation was blocked by a-BoCD2 controls as well as by ~-CACT31A and the seven MoAbs competitive with standard ~-BoCD2 MoAbs. Anti-CACT141A, -BAQ82A, and -BAQ91A did not inhibit E rosette formation.
Monoclonal antibodies recognizing T-lymphocyte subsets The 16 MoAbs identified as reacting with determinants present on either BoCD4 ÷ or BoCD8 + cells were paired with MoAbs (of different Ig isotype) known to recognize BoCD2, BoCD4 (a-IL A11, -IL A12 ), BoCD8 (a-IL A17 ), and each other for dual fluorescence analysis (Figs. 1-3 ). All 16 MoAbs reacted with subsets of the BoCD2 ÷ population, with those identified as specific for BoCD4 ÷ cells (a-CACT33A,-CACT83A, -CACT87A, -CACT91C,-CACT93A, -CACT132D, -CACT138A, -GC17A, and -GC50A) labeling IL All +, IL A12 ÷ cells, and those identified as specific for BoCD8 + cells (a-CACT80C, -CACT85A, -CACT88A, -CACT130A, -BAQlllA, -TH82D1, and -BAT82A) labeling IL A17 ÷ cells. These two subsets were mutually exclusive and together represented 90-100% of the BoCD2 ÷ set (Fig. 1). The nine MoAbs that recognized BoCD4 + cells were paired with the BoCD4specific MoAbs a-IL A l l and -IL A12 in dual fluorescence analysis (Fig. 2). Competition was evident in all pairs examined, but most pronounced with aCACT83B, which completely blocked staining by both a - I L A l l and -IL A12 (Fig. 2). Comparisons between MoAbs were limited, as most shared the same isotype. Of the comparisons performed, co-staining without competition was evident only in the a-GC17A and -CACT138A combination (not shown). The seven MoAbs that recognized BoCD8 ÷ cells were paired with either the BoCD8-specific MoAb a - I L A17 or each other in dual fluorescence FMF analysis {Figs. 1 and 3). Anti-CACT88A (Fig. 3) and -BAQlllA (Fig. 1) co-stained the IL 17A ÷ population. Anti-BAQ111A similarly co-stained with a-TH82D1 and a-CACT130A (not shown). Other MoAb combinations examined displayed competitive binding, most evident in the blocking of a-TH82D1 and -CACT85A by a-CACT88A and of a - I L A17 and -TH82D1 by a-CACT130A (Fig. 3).
Cross-reactivity The 27 MoAbs identified above were evaluated against caprine and ovine PBM by single fluorescence FMF {Table 2). Nine of the 11 pan T-specific MoAbs labeled caprine PBM. Profiles of the eight MoAbs reactive with bovine CD2 and one pan T (~-BAQ91A) were identical to those obtained with bovine cells. Anti-CACT141A and -BAQ82A did not react with caprine cells. Two of the pan T-specific MoAbs (o~-BAQ82A and -BAQ91A) reacted with PBM from some but not all sheep. Of the MoAbs reactive with bovine CD4 + cells, two of nine (~-GC17A1 and -GC50A1) recognized determinants present on ovine and caprine PBM. Profiles obtained with these cross-reactive MoAbs were
-
.~
i.:!.';
C1
B1
":
. . . . . . . . .
:
•
C2
B2
A2
,
.
,.
.
, ..~
..
,
,,:
C3
B3
A3
(z-TH82D1
.... ":;:":.,-
~, ..
.
~
•
.
~
~,
'.
:~:i;
,.~? ' ~ ". :•~
~-
.
I
C41 r
B4
• .
I
.1 ~:~
,'
C5
B5
(z-CACT80C ~z-CACT85A A4I I ~5
Green Fluorescence
•
(z-ILA17
(z-CACT130A
(z-CACT88A
Fig. 3. Anti-CD8 MoAbs: profiles of P B M reacted with one or two MoAbs using indirect labeling with tw() fluorochromes (FITC-GaMlg-isotv3)e specific anti b - G a M l g - i s o / s t r e p t a v i d i n - P E ) . Panel A1 is a profile of cells reacted with s t r e p t a v i d i n - P E in the absence of ant b.~dy. T h e other panels in row A (A2-AS) are profiles of cells reacted with MoAbs reactive with T-subsets or an (~-(?D8 MoAb and stained with b - ( ; a M l g - i s o / streptavidin-PE. T h e other panels m r~)w 1 (B1, C1 ) are profiles of (:ells reacted with MoAbs reactive with T-subsets. and stained with FITC(;aMlg-iso. The remaining panels ( B2-B5, C2-C5 ) I)resent profiles of cells reacted wit h two MoAbs iof different isotypc ) reactive with T-subsets or one MoAb reactive with T-subsets and one MoAb reactive with CDS. These were stained with both b - G a M I g - i s o / s t r e p t o v i d i n - P E and FITC(;aMlg-iso. The MoAb reactive with CD8 was used in panels A2 (?2 ((~-IL A17 ). T h e MoAbs reactive with T-subsets used here were: panels A3C3, (~-'FH821)l: panels A4 C4, c~-CA(!TSOC: panels A5 C5, et-('A('T85A: panels B1 BS. ~t-CA(?'I'88A; panels C1 C5. (~-('ACT13OA. Reactions n(~t shown: labeling with t~-BAQ111A was partially blocked by -(?A('T80C. whereas -('ACT88C partially bh)cked -BAQ 11 IA (both similar to panel B3 ). "l'be M()Ab (t'-BAQI 11A c(.labeled with -('A(~'I'I:10A and -'1'H821)1 l a.~ per panel B2 ~and with -1[. A17 (Fig. 1. panel [)6 ).
O
O C-
A1
.
~ ;~ ~" ~" z >~ -
MoAbs REACTIVE WITH BOVINE, CAPRINE AND OVINE T-I,YMPHOCYTE DETERMINANTS
205
identical to those produced with MoAbs specific for ovine and caprine CD4 (a-SBU T4 and -GC1A, Table 1 ). Of the MoAbs reactive with bovine CD8 + cells, four of seven (a-CACT80C, -TH82D1, -BAT82A, and -BAQlllA) reacted with caprine cells and three of seven (a-CACT80C, -TH82D1, and -BAT82A)reacted with ovine PBM. The determinant recognized by aBAQ111A was polymorphic in goats. Profiles obtained with the cross-reactive MoAbs were identical to those produced by MoAbs that recognize caprine (aBAGB25A) or caprine and ovine (a-SBU T8) CD8. Cross-reactive MoAbs were paired with MoAbs of known specificity and subjected to dual fluorescence FMF analysis with caprine and ovine PBL (not shown). Determinants detected by these MoAbs were absent from cells bearing IgM, N-cell, granulocyte or monocyte-specific determinants. Cross-reactive pan T-specific MoAbs either co-stained B26A4 + caprine cells (aBAQ91A), or competed with a-B26A4. The cross-reactive MoAbs that recognized bovine CD4 ÷ cells labeled a subset of caprine B26A4 ÷, BAQ91A ÷ cells and a subset of BAQ82A ÷, BAQ91A ÷ ovine cells, and competed with both aSBU T4 and -GC1A on both caprine and ovine cells. The cross-reactive MoAbs that recognized bovine CD8 + cells labeled a subset of caprine B26A4 ÷, BAQ91A + cells and a subset of BAQ82A ÷, BAQ91A ÷ ovine cells. These MoAbs competed with both a-SBU T8 and -BAGB25A on caprine cells, and competed with a-SBU T8 on ovine cells. The T-cell subsets recognized by these crossreactive MoAbs were mutually exclusive and together represented 90-100% of the B26A4 +, BAQ91A ÷ cells (caprine) or the BAQ82A ÷, BAQ91A ÷ cells (ovine).
Polymorphic determinants Profiles generated using PBM from 30 individual cattle revealed that at least three of the determinants detected (CACT33A, CACT132D, and TH82D1) were polymorphic. The screening of a - B A Q l l l A on caprine PBM revealed that its respective determinant is polymorphic in goats. Examination of over 50 animals has provided no evidence for a similar polymorphism in cattle. DISCUSSION Comparative analysis of FMF profiles obtained with MoAbs from primary cultures of hybridomas with those obtained using MoAbs of known specificity is an effective technique for distinguishing MoAbs that react with the same or new molecules. As demonstrated here, 27 MoAbs reactive with bovine T cell determinants were correctly identified 534 supernatants containing antibodies reactive with bovine leukocytes. This technique has also been used to identify MoAbs reactive with bovine N cells and B cells (Davis et al., 1987) and with goat and sheep PBM subsets (Davis et al., 1990). In this study 11 MoAbs were identified that reacted with determinants pres-
206
R.A. I ~ R S E N E T AI,.
ent on all mature T-cells. Competitive binding and inhibition of E rosette formation demonstrated that eight of these MoAbs (ce-BAT18A, -BAT42A, BAT76A, -BAQ95A, -CACT31A, -CACT98A, -MUC2A and - M U C l l A ) recognize determinants present on the BoCD2 molecule. All of these MoAbs displayed competitive binding when paired with other ce-BoCD2 MoAbs in dual fluorescence FMF. The finding that c~-MUCllA is not reactive with the caprine CD2 orthologue indicates that the determinant detected by this MoAb is structurally distinct from the determinants detected by the other ~-BoCD2 MoAbs. Three pan T MoAbs (c~-BAQ82A, -BAQ91A and -CACT141A) did not inhibit E rosette formation or compete with a-BoCD2 MoAbs. Although these MoAbs compete with each other, their patterns of cross-reactivity indicate that each MoAb recognizes a different determinant on the same molecule or molecular complex (Table 2). Antibodies that recognize human CD2, but do not block E rosette formation have been described {Meuer et al., 1984), however, Baldwin (personal communication) has found that c~-BAQ91A and -CACT141A compete with the MoAbs c~-IL A27 and -IL A28, which indicates that this set of determinants is present on the orthologue of CD6. This supposition is supported by data obtained at a recent workshop (see below). Nine MoAbs were identified that reacted with determinants restricted to BoCD4 ÷ cells (~-CACT33A,-CACT83A,-CACT87A,-CACT91C,-CACT93A, -CACT132D, -CACT138A, oGC17A, and -GC50A). All MoAbs in this set displayed competitive behavior in one or more combinations when paired with each other or with MoAbs known to react with BoCD4 (c~-IL A l l and -IL A12), indicating that determinants recognized by these MoAbs are present on the BoCD4 molecule. Two of these determinants (GC17A and GC50A) were also present on caprine and ovine T-lymphocyte subsets. This pattern of crossreactivity indicates that determinants are distinct from those recognized by other members of the set and from the GC1A and SBU T4 determinants, which are absent in cattle (Tables 1, 2). Seven MoAbs were identified that reacted with determinants restricted to BoCD8 ÷ cells (c~-BAQ111A,-CACT80C,-CACT85A,-CACT88C,-CACT130A, -TH82D1, and BAT82A). The MoAbs in this set either competed with ~-IL A17 or each other in one or more combinations, indicating the determinants recognized by these MoAbs are present on the BoCD8 molecule. Two MoAbs, c~-BAQlllA and -CACT88A, co-stained with c~-IL A17 and -TH82D1, indicating that their respective determinants are spatially removed from the IL A17 and TH82D1 determinants. Because CD8 is a heterodimeric molecule in rats (Johnson et al., 1985), humans (Barclay et al., 1987; Spurr et al., 1988) and perhaps cattle {Ellis et al., 1986), there exists the additional possibility that the determinants detected by a-BAQ111A and -CACT88A are not present on the peptide bearing the determinants detected by c~-IL A17 and -TH82D1. Regardless of their location, the B A Q l l l A and CACT88A determinants are not identical, as indicated by conservation of the BAQlllA, but not the
MoAbs REACTIVE WITH BOVINE, CAPRINE AND OVINE T-LYMPHOCYTE DETERMINANTS
207
CACT88A determinant in goats (Table 2). Similarly, patterns of cross-reactivity indicate that CACT80C and BAT82A are distinct from other determinants, as is TH82D1 - based on both cross-reactivity and polymorphism. All determinants recognized by this MoAb set are distinct from BAGB25A, which has only been detected on caprine lymphocytes {Table 2 ). Monoclonal antibodies which define BoCD4 (a-IL A11, -IL A12) and BoCD8 (a-IL A17) have been rigorously characterized (Baldwin et al., 1986; Ellis et al., 1986). The determinants recognized by these MoAbs are not broadly conserved among ruminants. Maddox et al. (1985) have produced two MoAbs (aSBU T4, -SBU T8) which identify the orthologous molecules in sheep. Both determinants are present in goats, but only one (SBU T8) occurs in cattle (Mackay et al., 1986). The cross-reactivity patterns of MoAbs described in this report and of MoAbs used here and described elsewhere {Davis et al., 1989) support the contention (Mackay, 1988) that the molecules identified as BoT4 in cattle and as SBU T4 in sheep and goats are orthologues, as are the molecules identified as BoT8 in cattle, and as SBU T8 in sheep and goats. Seventeen of the 27 MoAbs described in this study (Table 2 ) were submitted to the Workshop on Bovine, Ovine and Caprine Leucocyte Differentiation Antigens {summary in preparation). Cluster analysis and determinations of apparent molecular weights indicate that the specificities for all 17 MoAbs submitted were identical to those we had assigned based on the information presented in this study. Preliminary studies indicate that the strategy employed in the present study will be useful in identifying orthologous leukocyte differentiation molecules of other domestic and wild species {Davis et al., 1990). ACKNOWLEDGEMENTS The technical assistance of J. Davis, L. Froseth, L. Huston and S. Vibber is acknowledged. The authors thank Dr. David Reduker for his critical comments during the preparation of the manuscript. This research was supported in part by grants from the USDA-SEA (82-CRSR-2-2045, 83-CRSR-2-2281, 84-CRSR2-2457, 86-CRSR-2-2913, 86-CRCR-1-2241 ), USDA-CSRS (89-37265-4537), NIH (CA50141-01), WSU-ARC (project 3073), the Washington Technology Center, and the Carnation Research Farm. R.A. Larsen was supported by a National Services award (A106 07025 ) from the National Institutes of Allergy and Infectious Diseases. Y.H. Park was supported by the exchange scientist student program of the Rockefeller Foundation. REFERENCES
Baldwin, C.L., Teale, A.J., Naessens, J.G., Goddeeris, B.M., MacHugh, N.D. and Morrison, W.l., 1986. Characterization of a subset of bovine T lymphocytes that express BoT4 by monoclonal antibodies and function: Similarity to lymphocytes defined by human T4 and murine L3T4. J. Immunol., 136: 4385-4391.
208
R.A.LARSENET AI..
Barclay, A.N., Johnson, P., McCaughan, G.W. and Williams, A.F., 1987. Immunoglobulin-related structures associated with vertebrate cell surfaces. In: T.W. Mak (Editor), The T Cell Receptors. Plenum Press, NY, pp. 53-87. Davis, W.C., 1988. Enhancement of myeloma-B-cell hybridoma outgrowth in primary cultures with B cell mitogens. Period. Biol., 90: 367-374. Davis, W.C., McGuire, T.C. and Perryman, L.E., 1983. Biomedical and biological application of monoclonal antibody technology in developing countries. Period. Biol., 85: 259-282. Davis, W.C., Marusic, S., Lewin, H.A., Splitter, G.A., Perryman, L.E., McGuire, T.C. and Gorham, J.R., 1987. The development and analysis of species-specific and cross-reactive monoclonal antibodies to leukocyte differentiation antigens and antigens of the major histocompatibility complex for use in the study of the immune system in cattle and other species. VeL Immunol. Immunopathol., 15: 337-376. Davis, W.C., Ellis, J.A., MacHugh, N.D. and Baldwin, C.L., 1988. Bovine pan T-cell monoclonal antibodies reactive with a molecule similar to CD2. Immunology, 62: 165--167. Davis, W.C.. Larsen, R.A. and Monaghan, M.L., 1990. Genetic markers identified by immunogenetic methods. International Symposium and Educational Workshop on Fish-marking Techniques, in press. Ellis, J.A., Baldwin, C.L., MacHugh, N.D., Bensaid, A., Teale, A.J., Goddeeris, B.M. and Morrison, W.I., 1986. Characterization by a monoclonal antibody and functional analysis of a subset of bovine T lymphocytes that express BoT8, a molecule analogous to human CD8. Immunology, 58: 351-358. Fitch, W.M., 1970. Distinguishing homologous from analogous proteins. Syst. ZooL, 19: 99-113. Goodman, M., Miyamoto, M.M., Czelusniak, J., 1987. Pattern and process in vertebrate phylogeny revealed by coevolution of molecules and morphologies. In: C. Patterson (Editor), Molecules and Morphology in Evolution: Conflict or Compromise? Cambridge University Press, Cambridge, pp. 142- 176. Johnson, P., Gagnon, J., Barclay, A.N. and Williams, A.F., 1985. Purification, chain separation and sequence of the MRC OX-8 antigen, a marker of rat cytotoxic T lymphocytes. EMBO J.. 4: 2539-2545. I,anier, I,.l,., Engleman, E.G., Gatenby, P., Babcock, G.F., Warner, N.L. and Herzenberg, I,.A.. 1983. Correlation of functional properties of human lymphoid cell subsets in surface marker phenotypes using multiparameter analysis and flow cytometry. Immunol. Reviews.. 74: 143160. Lanier, L.L., Le, A.M., Ding, A.H. and Evans, E.I,., 1987. Analysis of the workshop T-cell monoclonal antibodies by 'indirect two-colour immunofluorescence' and multiparameter flow cytometry. In: A.J. McMichael (Editor), I~ukocyte Typing III. Oxford University Press, New York, NY, pp. 62-65. I,ewin, H.A., Davis, W.C. and Bernoco, D., 1985. Monoclonal antibodies that distinguish bovine T and B lymphocytes. Vet. Immunol. Immunopathol., 9: 87-102. Mackay, C.R., 1988. Sheep leukocyte molecules: A review of their distribution, structure and possible function. Vet. Immunol. Immunopathol., 19: 1-20. Mackay, C.R., Maddox, J.F. and Brandon, M.R., 1986. Three distinct subpopulations of sheep T lymphocytes. Eur. J. Immunol., 16: 19-25. Maddox, F., Mackay, C.R. and Brandon, M.R., 1985. Surface antigens, SBU-T4 and SBU-T8, of sheep T lymphocyte subsets defined by monoclonal antibodies. Immunology, 55: 739-748. Meuer, S.C., Hussey, R.E., Fabbi, M., Fox, D., Acuto, D., Fitzgerald, K.A., Hodgdon, J.C., Protentis, J.P., Schlossman, S.F. and Reinherz, E.L., 1984. An alternative pathway of T-cell activation: A functional role for the 50 kD T11 sheep erythrocyte receptor protein. Cell, 36: 89790x. Spurt, N.K., Goodfelh)w, P.N. and Johnson, P., 1988. CD8B, the human equivalent of the mouse Ly-3 gene is localized on chromosome 2. Immunogenetics, 27: 70-72.
