Phylogenesis of lymphocyte diversity II. Characterization of agama stellio Ig-negative lymphocytes by a heterologous anti-thymocyte serum

DEVELOPMENTAL AND COMPARATIVE IMMUNOLOGY, Vol. 7, pp. 507-516, 0145-305X/83 $3.00 +.00 Printed in the USA. Copyright (c) 1983 Pergamon Press Ltd. All rights reserved.

1983.

PHYLOGENESIS OF LYMPHOCYTE DIVERSITY

II. CHARACTERIZATION OF AGAMA STELLIO Ig-NEGATIVE LYMPHOCYTES BY A HETEROLOGOUS ANTI-THYMOCYTE SERUM 1

Hoda Negm and M.H. Mansour

2

Zoology Department, Faculty of Science, Cairo University Cairo, Egypt

ABSTRACT A specific antiserum was raised in rabbits against thymocytes of the lizard Agama stellio (ATS). In indirect immunofluorescent (IF) assays, the in vitro absorbed ATS was, at any given dilution, reactable with thymocytes > peripheral blood lymphocytes (PBL) ~ splenocytes > bone marrow lymphocytes (BML). In quantitative absorption analyses the specificity of ATS for recognizing antigenic determinant(s) specific to Agama thymocytes (ASTA) and shared by other lymphocytes of probable thymic origin was indicated. In view of the inverse correlation of ASTA-bearing lymphocytes with Ig-bearing lymphocytes in lymphoid tissues, and the uniformity of results obtained with either an antiserum raised with Ig-negative thymocytes or antisera absorbed with purified y-fractions, ASTA may be a strict phenotype specific to Ig-negative thymocytes and other Ig-negative lymphocytes in the periphery. We propose a dichotomy of the T- and B-like arms of immunity in Agama, on the basis of presence or absence of ASTA.

INTRODUCTION The detection of antibodies specific for cell surface antigens is among the most important techniques of cellular immunology. Such antibodies are useful in determining the role of membrane-bound molecules in immune functio~ delineating subpopulations of lymphocytes and defining cell-surface differentiation markers (i). In higher vertebrates, while the presence of readily detectable surface Ig has served as an important marker for elucidating B-lymphocyte functions, a number of phenotypic markers with representations limited to T-lymphocytes has been also introduced (2-10). In the murine system, different subclasses of T-cells are characterized by distinct differentiation antigens of the Ly 507

508

SURFACE MARKERS ON LIZARD LYMPHOCYTES

Vol. 7, No. 3

system. The characterization of these Ly antigens has been greatly facilitated by using alloantisera generated in genetically defined strains of mice (9,11). However, due to lack of inbred strains in other animal models and humans, greater attention has been directed towards xenoantisera as an alternative approach for studying lymphocyte-specific antigens. In spite of difficulties associated with producing specific xenoantisera, defining xenoantigens on the bulk of thymus-dependent (T) lymphocytes and their heterogeneous expression among subpopulations has been described in humans (5), mouse (12), rat (6), guinea pig (7), hamster (13) and chicken (8). At the reptilian level recognition of most T-cells, via the demonstration of specific xenoantigens, has also served as a powerful tool in delineating cellular compartments within this phylogenetically primitive level (14,15). The present study deals with the characterization and distribution of A g a m a Ig-negative lymphocytes utilizing a rabbit anti-Agama thymocyte serum in indirect IF assays and quantitative absorption analyses. The previously suggested structural heterogeneity of lymphocytes in this species is further documented in the present report (16).

MATERIAL AND METHODS Animals Adult male and female lizards were used in the present study and maintained as described earlier (16). Preparation of cells used for immunization Thymi excised from normal lizards were gently teased on ice-cold serum-free medium (199). Cell clumps were allowed to settle, and cell suspensions passed onto a gradient of 5:1 (v/v) distilled water and Verografin (Leciva, Narodni Podnic, Prague) as a lymphocyte-enrichment step. After centrifugation at 1800 rpm for 25 min at 4°C, the lymphocyte-rich layer at the interface was collected and lymphocytes washed thrice with medium (199) , counted and viability assessed by trypan blue exclusion (0.16% w/v). Aliquots of Ig-negative thymocytes were prepared by treating thymocyte suspensions with undiluted anti-Agama y-globulin serum (AAGS) (16) supplemented with 1:4 diluted fresh guinea pig complement for 45 min at 37°C. Samples of treated thymocyte suspensions revealed not a single positive cell using AAGS and fluorescein conjugated-goat anti-rabbit y-globulin (FITC-ARG, Behring Institute, Marburg, W. Germany) in indirect IF assays. Preparation of anti-thymocyte sera

(ATSI,II)

One rabbit received, at two-week intervals, three intraperitoneal injections of 80-120 x i0 viable thymocytes suspended in 1 ml serum-free medium (199). Another rabbit received the same number but of Ig-negative thymocytes, suspended in 1 ml serum-free medium (199). The two rabbits were bled a week after the last injection, and blood collected individually. Sera were separated by centrifugation at 3000 rpm for 20 min and heat inactivated at 56°C for 30 min. The resulting serum from the first rabbit was designated ATS I and that of the second, ATSII. In vitro absorption of sera was effected

Vol.

7, No. 3

SURFACE MARKERS ON LIZARD LYMPHOCYTES

509

sequentially, with thrice-washed A. stellio erythrocytes and kidney cells in p h o s p h a t e - b u f f e r e d saline (PBS) pH 7.2, each at ratio of 5:1 for 45 min at 37°C, then stored in small aliquots of 0.2 ml at -20°C until used; aliquots of antisera were left unabsorbed and stored frozen at -20°C. Normal rabbit serum (NRS) was o b t a i n e d from the same rabbits prior to immunization. Indirect immunofluorescence

(IF) assay

One to two million viable lymphocytes were incubated for 45 min with 0.i ml of serial two-fold dilutions of in vitro absorbed ATS I and ATSII at 4°C. Controls were maintained by replacing the antisera with either absorbed undiluted NRS and/or antisera absorbed with pure Agama y-globulin fractions (16). After the incubation period, lymphocytes were washed thrice with medium (199) and reincubated at 4°C for 45 min with 0.i ml of i:I0 diluted FITC-ARG. After four washes, lymphocytes were suspended in one drop of medium and then mounted on slides. Lymphocytes were scored microscopically at a magnification of 400x - 640x by using a Leitz Dialux microscope equipped with Ploemopak FITC labeling blue excitation and lamp housing 102 Z with 50 watt Hg. Percentage of positive cells was determined by counting a minimum of 200 lymphocytes and the reactions quantitated by calculating percentage of fluorescence indices (FI%) using the following formula: FI% = % of +ve lymphocytes in a n t i s e r u m - % of +ve lymphocytes in control xl00 i00 - % of +re lymphocytes in control Quantitative absorption analysis of ATS A range of 0.5 x 106 to i0 x 106 viable th~mocytes, PBL, splenocytes and BML were incubated separately, for 45 min at 37 C, with aliquots of 0.i ml of undiluted, in vitro absorbed ATS with constant agitation. BML were selectively obtained without any contaminating cell type by passing cell suspensions onto a V e r o g r a f i n gradient with a yield in the range of 0.25 - 0.85 x 106 lymphocytes/2 femur bones of an individual lizard. The absorbed aliquots were recovered by centrifugation at 1800 rpm for 5 min and then tested against thymocytes for residual reactivity. Residual reactivity (expressed as FI%) was plotted against the number of lymphocytes used for absorption to give an absorption curve. Both NRS and undiluted ATS, absorbed in vitro with Agama erythrocytes and kidney cells but not with any viable lymphocytes, were used as parallel control series. Statistical analysis of data Results are expressed as mean+ standard error. Significance of the differences between the means was calculated, when necessary, according to Student's "t" test.

RESULTS Absorption of anti-thymocyte serum Prior to any cellular absorption, serially two-fold diluted aliquots of ATSI, A T S i i w e r e tested thrice with Agama lymphoid (thymocytes, PBL, splenocytes and ~ML) and non-lymphoid cells (erythrocytes, kidney cells and thrombocytes) in indirect IF. More than 95% of all these cell types consistently displayed a patchy surface fluorescein stain up to i:i00 dilution;

510

SURFACE MARKERS ON LIZARD LYMPHOCYTES

Vol.

7, No.

3

however, slightly higher titers (1:125) were necessary for thymic lymphocytes. The relatively diminished titers obtained could be attributed to the comparatively small numbers of immunogens used and to the absence of adjuvant during immunizations° In this respect, no significant differences between the resulting batches of antisera were observed, and both were distributed thereafter into small aliquots. By a single in vitro absorption of the antisera with A g a m a v e r o g r a f i n - p u r i f i e d erythrocytes and kidney cells, the reactivity of both ATS I and ATSII to non-lymphoid cells was entirely abolished. The p e r s i s t i n g fluorescence, on the other hand, was totally confined to lymphocytes. Immunofluorescent

staining

capacity

in in vitro absorbed ATS on lymphocytes

The in vitro absorbed ATS I and ATSII were tested for residual IF staining reactivity against pooled lymphocyte-enriched suspensions of thymus, p e r i p h e r a l blood (PB), spleen and bone marrow (BM). When test performance was either at 4°C or 37°C, no mark of heterogeneity of the staining intensity was observed among positive lymphocytes at any given dilution of ATS I or ATSII. Positivity of lymphocytes was always displayed in patchy fluorescent patterns. By contrast, not a single positively stained lymphocyte was observed with absorbed NRS. Consistently similar results were o b t a i n e d with either ATS I or ATSII and are shown in Fig. 1 as the mean fI%~ standard error from seven separate experiments. Although ATS reactivity was markedly diminished to a titre of 1:16, yet within this range of dilution, a discriminatory reactivity of ATS was apparent among lymphocytes of various lymphoid tissues and blood. At any given dilution, ATS reactibility was always significantly (P > 0.001 - P < 0.01) higher with thymocytes > PBL > Splenocytes > BML. However, among thymic lymphocytes, an increasing level of reduction in positivity by 55.88%, 75.13% and 96.5% was observed in parallel with the increasing levels of ATS dilution by 4, 8 and 16 times, respectively. Similarly, a linear reduction in the positivity of PB and splenic lymphocytes by 38% 51%, 61% - 72% and 71% - 82% accompanied the successive increase of ATS dilution by 4, 8, 16 times, respectively. This greatly participated in similar trends of either PB or splenic reactivity curves with respect to the thymic curve using ATS, suggesting antigenic correlation between thymic lymphocytes and a p o p u l a t i o n of PB and splenic lymphocytes, to which specific antibody (ies) in the in vitro absorbed ATS was (were) specifically directed° As depicted in the BML curve of reactivity with ATS, the trend of the curve was quite irrelevant to that of thymic, PB and splenic lymphocyte curves. However, the p u t a t i v e - s h a r e d antigenicity of a fraction of BML with thymic lymphocytes was not excluded since a definite percentage of about 9.5% of BML consistently displayed the positive pattern of fluorescent staining up to 1:8 dilution of ATS. It is noteworthy that only with the highly concentrated ATS, a significantly high level of differences between the mean percentage positive thymic lymphocytes and BML was observed. With the undiluted, 1:2 and 1:4 diluted ATS, the difference was very highly significant (P < 0.001) and thereafter reduced to a lower level (P < 0.01) with the 1:8 diluted ATSo This may suggest that, by increasing the ATS concentration, its reactivity tended to discriminate better between the organs. Subsequently, this would favor the maximum discriminatory reaction to be attained with the undiluted ATS.

Vol.

7, No.

3

SURFACE MARKERS ON LIZARD LYMPHOCYTES

(3~

T h y mo cymes

~ ~

Pe¢lpherol

blood

Lym~h~y~es

Solenocyees

~ --

I00

511

~

Bone morrow

Lymohocytes

SO

i zo

60

L

5o-

40.

30

20-

IO

Undiluted

1:2 A T S

1:4

1:8

dilutions

Fig. i. Reactivity of undiluted and serially two-fold diluted, in vitro absorbed ATS against thymic, peripheral blood, splenic and bone marrow lymphocytes. Each point on the curves represents the FI% and the vertical bars indicate the standard errors of seven separate experiments.

Additive staining capacity of undiluted, in vitro absorbed ATS and undiluted AAGS on lymphocytes in indirect IF The percentage of positive lymphocytes detected by the undiluted ATS was compared with those detected by undiluted AAGS (16) (Fig. 2). ATS-reactive lymphocytes comprised 92 + 2.68%, 77.71 + 2.4%, 68.57 + 3.94% and 9.71 + 1.9% of thymic, PB, splenic ~nd BM lymphocy~es. By contra--st, undiluted AAGS reacted with 7.2 + 0.5%, 25.4 + 1%, 29.2 + 1.2% and 44% of the lymphocytes of the same organs" Except fo--r the BM, The sum of lymphocytes obtained using ATS and AAGS within a given tissue seemed to account for the majority, if not all, of lymphocytes represented. In this respect, similar additivity was not found between lymphocytes d e t e c t e d by undiluted AAGS and any other dilution of ATS. However, consistently similar complementarity was observed with undiluted ATS and ATS absorbed with pure A~ama y-globulin fractions I I . and this would account for t{e entlre absence of anti-Ig antibodies in either types of ATS.

512

SURFACE MARKERS ON LIZARD LYMPHOCYTES

[~ 9O

Vol.

ATS - reoclive kymphocytes Anti - ~ serum reactive

7, No. 3

I

tymphocytes !

i / l / i i i i

eo i / / i

///I //// //// /ii/ I/// I///

T /.//.'1

// // / A ,~

"///,

///.d 1//i

/ / 1 1

Y//

Y// Thymocytls

PeripheroL blood t ymphocytes

S pLenocytes

Bo.e morrow Lymphocytes

Fig. 2. Histograms showing additivity of absorbed undiluted AAGSreactable and in vitro absorbed undiluted ATS-reactable thymic, peripheral blood, splenic and bone marrow lymphocytes. Vertical bars indicate standard errors.

Quantitative absorption of in vitro absorbed, undiluted ATS Quantitative absorption was used to investigate the degree to which antigenic determinant(s) was (were) held in common between thymic lymphocytes and lymphocytes of other lymphoid tissues and blood. The results obtained are represented as the mean FI% of two separate series of experiments with ATS I and ATS and the residual reactivity was corrected with respect to I reactivity o~ undiluted in vitro absorbed ATS (taken as 100% instead of 92%) (Fig. 3). When 50% reduction in ATS reactivity (absorption index) was taken as a measure of absorption capacity, 0.65 x 106 thymic lymphocytes were required to achieve this level, whereas 1.4 x 106 and 1.65 x 106 of PBL and splenic lymphocytes were required, respectively. BML, in contrast, seemed to be extremely beyond these levels, since up to i0 x 106 BML were only capable of reducing ATS reactivity by 45%. With respect to content of antigenic determinant(s) (to which antibody(ies) in ATS is specifically related), thymic lymphocytes seemed to express the highest amounts of these determinants, if compared to any other organ. If thymic lymphocytes would be the least heterogeneous population with respect to these determinants, BML would represent a contradictory issue in either being the lymphocytes with the least content thereof or in being the most heterogeneous population with this particular antigenic determinant(s).

Vol.

7, No. 3

SURFACE MARKERS ON LIZARD LYMPHOCYTES

0

o

Thymocyles Peripheral blood Lymphocytes

¢

=

Sptenocytes

•

•

8one

513

morrow [ymphocyfes

80~

\

40-

~O-

ZO-

I0"

I N~

of

Lymphocyfes

used for absorption

Fig. 3. Residual immunofluorescent staining capacity of undiluted in vitro absorbed ATS, after sequential absorptions with increasing numbers of thymic, peripheral blood, splenic and bone marrow lymphocytes, against thymic lymphocytes as targets. Each point on the curves represents the mean value of two separate series of experiments.

In the present study, a heterologous antiserum was raised in rabbits using Agama thymic lymphocytes as immunogens -- the majority being Ig-negative lymphocytes (16). Prior to any cellular absorption, the indirect IF studies disclosed that unabsorbed ATS contained antibodies reactive to the membrane of all thymic, PB, splenic and BM lymphocytes, as well as the nonlymphoid erythrocytes, thrombocytes and kidney cells; slightly higher titers were observed against thymic lymphocyteso By a single in vitro absorption with Agama erythrocytes and kidney cells, reactivity of ATS was confined to cells of the lymphocytic series. This suggests that a single absorption was sufficient to remove all nondiscriminating antibodies in ATS which are directed to putative antigens expressed on all Agama cells. In previous attempts, however, various heterologous antisera required extensive absorptions, either in viva or in vitro, in order to be rendered l y m p h o c y t e - s p e c i ~ ic (reviewed in 5)° At any given dilution, in vitro absorbed ATS reactivity was always higher towards thymocytes than lymphocytes of other lymphoid organs~ However, the similar trends attained by thymic, PB and splenic lymphocyte reactivity curves provide a clue to an antigenic correlation between thymocytes and a proportion of other lymphocytes in the periphery.

514

SURFACE MARKERS ON LIZARD LYMPHOCYTES

Vol.

7, No. 3

In view of the uniform reactivity of ATS consistently obtained in every experiment (S.E. never exceeded 5%) with lymphocytes pooled from several individual lizards, it seems most unlikely that the antigen(s) recognized by ATS would be a product of a polymorphic genetic system (e.g., histocompatibility antigens). In this regard, the thymocyte-lymphocyte antigenic determinant(s) identified by ATS absorbed in vitro seems to be representative of Agama stellio as species, rather than of some members of the species; subsequently, the designation Agama-specific thymic lymphocyte antigen (ASTA) is suggested for these determinants. However, it is difficult to ascertain whether ASTA indicates one or several antigens since it lacks allelic variation and, thus, is not amenable to conventional tests of genetic determination. In quantitative absorption analysis, cross-reactivity was clearly indicated by the initial parallel absorption kinetics of thymic, PB and splenic lymphocytes in reducing the anti-ASTA activity of ATS. However, the specific flexions obtained at higher cellular concentrations of PB and splenic absorption curves indicated heterogeneity of these organs with respect to their ASTA content. BML by contrast were least capable of absorbing anti-ASTA activity of ATS, and this was reflected by the completely irrelevant trend of its absorption curve compared to that of thymic lymphocytes, suggesting a low content of ASTA among BML. A major observation inferred from studies with ATS is that ASTA and Ig determinants do not coexist on Agama lymphocyte surfaces and that ASTA appears to be a strict phenotype, associated selectively with the surfaces of Ig-negative thymocytes and other lymphocytes in the periphery. Several pieces of evidence support this suggestion. First, the distribution of Igbearing lymphocytes was inversely related to that of ASTA-bearing lymphocytes in all lymphoid tissues and blood, when either ATS I or ATSII (absorbed or unabsorbed with purified y-fraction) was used. Secondly, lymphocytes b e a r i n g ASTA and surface-bound Ig (16) within a given tissue always accounted for the majority, if not all, of the lymphocytes represented; yet such additivity did not apply to the BM. Finally, ATS was not capable of detecting the minority of thymic lymphocytes that bear Ig determinants (16), although they were included in the thymic population used as an immunogen. Although it is p r e d i c t e d that this minority might not be adequate for raising anti-Ig antibodies in ATS, the plateau of residual reactivity, depicted in the thymocyte absorption curve, would indicate the inability of this Ig-positive minority to absorb anti-ASTA activity of ATS. In summary, the bulk of Ig-negative lymphocytes of Agama appears to express ASTA which seems to be analogous to the various xenoantigens described previously (5-8, 12-15) in being a specific differentiation antigen for Agama thymocytes and other lymphocytes of probable thymic origin. In structural terms, it is conceivable that dichotomy of the T- and B-like arms of immunity in Agama might also be documented on the basis of presence or absence of the ASTA presented in this study.

ACKNOWLEDGEMENTS lo

Support by United States Public Health Service Grant No. 03-035N.

2.

Present address: Department of Anatomy, School of Medicine, The Center for Health Sciences, University of California, Los Angeles, California 90024.

Vol.

7, No. 3

SURFACE MARKERS ON LIZARD LYMPHOCYTES

515

REFERENCES io

Beverley, P.C.L. Lymphocyte heterogeneity. In: B and T Cells In Immune Recognition, F. Loor and G.E. Roelants (eds.). London, John Wiley, 1977, p. 35.

2.

Shen, F., Boyse, E.A. and Cantor, H. Preparation and use of lymphocyte antisera. Immunogenetics 2, 591, 1975.

3.

Brown, Go and Greaves, M.F. Enumeration of absolute numbers of T- and B-lymphocytes in human blood. Scand. J. Immunol. 3, 161, 1974.

4.

Smith, R.W., Terry, W.D., Buell, D.W. and Sell, K.W. An antigenic marker for human thymic lymphocytes. J. Immunol. ii0, 884, 1973.

5.

Balch, G.M. and Ades, E.W. Heterogeneity of cell surface xeno-antigens on human T-lymphocytes. J. Reticuloendoth. S o c 25, 635, 1979.

6.

Ishii, Y., Koshiba, H., Yamaoka, H. and Kikuchi, K. Rat T-lymphocytespecific antigens and their cross-reactivity with mouse T cells. J. Immunol. 117, 497, 1976.

7.

Godfrey, H.P., Greczy, A.F., Gell, P.G.H. and Rubin, B. Induction of specific anti-guinea pig T cell sera in rabbits. J. Immunol. Meth. 9, 211, 1976.

8.

Potworowski, E.F., Richer, G., Borduas, A.G. and Forget, A. Selective inhibition of plaque-forming cells by bursa-specific antisera. J. Immunol. 106, 1416, 1971.

9.

Cantor, F. and Boyse, E°A. Functional subclass of T-lymphocytes carrying different ly Ags. I. The generation of functionally distinct T cell subclasses is a differentiative process independent of antigen. J. Exp. Med. 141, 1376, 1975.

i0.

Brochier, J., Abou-Hamed, Y.A. Gueho, J.P. and Revillard, J.P~ Study of human T- and B-lymphocytes with heterologous antisera. I. Preparation, specificity and properties of antiserao Immunology 31, 749, 1976.

ii.

Cantor, H. and Weissman, I. Development and function of subpopulations of thymocytes and T-l>~phocytes. Prog. Allergy 20, i, 1976.

12.

Shigeno, N., Hammerling, U., Arples, C., Boyse, E.A~ and Old, L.J. Preparation of lymphocyte-specific antibody from anti-lymphocyte serum. Lancet 2, 320, 1968.

13.

Blasecki, J.W. and Houston, K.J. Identification, functional characterization and partial purification of thymus-derived lymphocytes in inbred hamsters. Immunology 35, i, 1978.

14.

Pitchappan, R.M. and Muthukkaruppan, V.R. heterologous anti-lizard thymocyte serum. 85B, I, 1977.

15.

Mansour, M.H., E1 Ridi, R. and Badir, N. Surface markers of lymphocytes in the snake Spalerosophis diadema. I. Investigation of lymphocyte surface markers. Immunology 40, 605, 1980.

16.

In vitro properties of Proc. Indian Acad. Sci.

Negm, H. and Mansour, M.H. Phylogenesis of lymphocyte diversity. IImmunoglobulin determinants on the lymphocyte surface of the lizard Agama stellio. Devel. Comp. Immulol. 62, 519, 1982. Receiced : January, 1982 Accepted : May, 1982