- Email: [email protected]
Circulating antibodies to prothymosin α in systemic lupus erythematosus
CLINICAL
IMMUNOLOGY
Circulating
AND
IMMUNOPATHOLOGY
53, 151-160 (1989)
Antibodies to Prothymosin Lupus Erythematosus
(Y in Systemic
P. G. VLACHOYIANNOPOULOS,* S. FRILLINGOS,? A. G. TZIOUFAS,* K. SEFERIADIS,? H. M. MOUTSOPOULOS,* AND 0. TSOLAS? *Department of Internal Medicine and fLaboratory of Biological Chemistry, Medical School, University of loannina, 451 10 loannina, Greece Autoantibodies to prothymosin (x, an imnnmoactive protein that exists in a large variety of mammalian tissues, were found to be present in patients with systemic lupus erythematosus (SLE) by a new, sensitive, and specific anti-prothymosin a ELISA. The antigen was prothymosin a, purified by high-pressure liquid chromatography from goat spleen extracts. Sera from 44 SLE patients and 276 healthy individuals were screened for the presence of anti-prothymosin a activity; 18% of SLE sera were found to be positive, compared with 1.8% of control sera. This anti-prothymosin o! activity appears to be idiotypically distinct from either anti-thymosin o, activity or anti-dsDNA activity, as demonstrated by inhibition experiments. Sign&ant positive correlation exists between anti-prothymosin d and anti-dsDNA activities of SLE sera (r = +0.5%, n = 36, P < O.OOl), while no correlation was observed with the clinical activity (x2 = 1.239, 0.1 < P < 0.5) or with complement levels C, and C4. o 1989 Academic press, IIIC.
INTRODUCTION
Systemic lupus erythematosus (SLE) is a systemic autoimmune disease characterized by the presence of numerous nonorgan specific autoantibodies (1). The involvement of these autoantibodies in the pathogenesis of the disease is yet unclear; nevertheless, anti-dsDNA autoantibodies have been shown to play a central role in immune complex-mediated tissue damage (2) and are currently used as a criterion of diagnosis for the disease (2). The primary defect accounting for this autoreactivity is either impaired T-cell suppressor function or B-cell hyperactivity, known to be the main immunoregulatory abnormalities in SLE (1). Autoantibodies against lymphocytes (1) and aberrant production of T-cell differentiation and growth factors (1) have been described to occur in SLE. One set of immunoactive peptides, derived from the thymus, plays a central role in the differentiation and maturation of T-lymphocytes. Thymosin o1 (To, 3108 Da), in particular, has been shown to enhance the expression of Thy-l,2 and Lyt-1,2,3 antigens on the T-cell (3), enhance natural killer (NK) activity in vivo (4) and in vitro (5), protect sensitive or immunosuppressed mice against opportunistic infections (6), enhance the human mixed lymphocyte response (MLR) by acting on T4+ cells (7), and improve survival, in conjunction with radiotherapy of immunosuppressed lung cancer patients (8). Of particular interest is prothymosin a (ProTa), a highly acidic (pZ = 3.55) polypeptide of 12,500 Da (9), which contains the entire thymosin a, sequence at its amino terminal (IO), and an acidic stretch in the central portion of the molecule 151 0090-1229/89 $1.50 Copy&M Au rights
0 1989 by Academic Press, Inc. of rcpmdwXion in any form named.
I52
VLACHOYIANNOPOULOS
El‘
AI..
(10). Prothymosin a appears to be more immunoactive than thymosin ai in certain and in vitro (12) assays, i.e., in protecting sensitive mice against infections with Can&da alhicuns (I I) and enhancing the human autologous or allogeneic MLR (12). Prothymosin a has also been reported to restore suboptimal human MLR (12). as well as deficient MLR in lymphocytes from patients with clinically active multiple sclerosis or SLE (13, 14). Although first isolated from rat thymus (9). prothymosin (Y is now known to have a wide distribution in the tissues of mammals, such as rat (15), pig (16), goat (17). and man (18). Moreover. the wide tissue distribution of prothymosin (Y mRNA (19) gives evidence for the synthesis of prothymosin (Ynot only in thymus, but also in other tissues, especially the spleen (19-21). Incubation of SLE peripheral blood lymphocytes with thymic extracts or thymic peptides (22) increases in vitro the number and function of T-cells. These observations suggest that thymic peptide deficiency may be present in SLE. To study if this deficiency is immunologically mediated, we searched for the presence of circulating autoantibodies in the sera of patients with SLE. in vivo (II)
MATERIALS
AND METHODS
Patients and sera. Sera from 44 SLE patients and from 276 healthy individuals as controls were studied. The sera were kept at -20°C until tested. The anti-DNA antibodies were detected with an ELISA method described previously (23). Autoantibodies to nuclear and cytoplasmic antigens were tested by counterimmunoelectrophoresis (24). The levels of the complement C, and C4 were measured by radial immunodiffusion (Boehringer). SLE was considered active when the patient presented with at least one of the following manifestations: serositis, arthritis, nephritis, myositis, vasculitis, and central nervous system involvement, as well as high serum DNA binding or low levels of complement. Preparation of ProTcx. Prothymosin a was purified from the spleen of a young female goat by gel filtration on Sephacryl S-200 (Pharmacia) and reversed-phase HPLC on Altex Ultrasphere ODS-C,, columns (Beckman Instruments), following a procedure described by others (25). The purity of the peptide was confirmed by analytical isoelectric focusing (IEF) performed on a LKB Ampholine PAG plate (LKB); IEF revealed a single band at the characteristic isoelectric point for ProTo (~1 = 3.5). Quantitative determinations of the peptide, used for coating the ELISA plates, were based on amino acid analyses, performed on hydrolysates as the phenylthiocarbamyl derivatives (26). Synthetic thymosin oi used in some experiments was a generous gift from A. M. Felix of Hoffmann-La Roche (Nutley, NJ). ELZSA. Prothymosin (Yat different quantities per well was applied to polystyrene cuvettes, either precoated or not precoated with poly-L-lysine (PLL) (Sigma) 50 pg/ml in PBS. The quantities of the antigen per well, tested in several preliminary experiments, were 50, 100, 200, 400, 800, and 1600 ng. In order to reduce the nonspecific binding of serum immunoglobulin, especially in hypergammaglobulinemic sera, 10% bovine serum (BS) was used as a blocking agent (23).
PROTHYMOSIN
a
AND
SYSTEMIC
LUPUS
ERYTHEMATOSUS
153
A strongly positive SLE serum as well as a negative control were tested in serial dilutions from I:25 to 1:3200, and the tests were repeated for all of the quantities of antigen mentioned. The reaction was measured at 15 min intervals, at 37”C, for 4 hr. The best discriminatory effect between positive and negative controls was obtained using 1600 ng of antigen (ProTar) per well, a working dilution of the sera 1:50, and incubation time with the substrate of 90 min, which is in the linear portion of the reaction. Standard conditions for ELISA were set as follows: ProTa (1600 ng per well) in coating buffer 10 mM Na,C03, 30 mM NaHCO,, pH 9.6, was incubated in polystyrene cuvettes, at 37°C for 1 hr. After washing with PBS, the nonspecific binding sites were blocked with 10% BS at 37”C, for I hr. Fifty microliters of serum diluted 150 in 10% BS was added in duplicate and incubated at 37°C for 3 hr. After washing, antibody bound to ProTa was detected with alkaline phosphatase antihuman y chain-specific antisera (Sigma), diluted 1:lOOO in 10% BS. Substrate solution (p-nitrophenyl-phosphate disodium) 1 mg/ml in 10% diethanolamine, 100 mg/liter MgC1,6H,O, pH 9.8, was added to each well and the absorbance was read at 405 nm. Sera were considered to be positive when the absorbance was higher than the mean of normals plus three standard deviations. Discrimination between anti-dsDNA and anti-ProTa activity. This was first examined by inhibition experiments which showed that the binding of strongly positive anti-ProTa sera was not inhibited by preincubation with dsDNA, and second, by affinity chromatography of a serum with both anti-DNA and anti-ProTa activities on a dsDNA-cellulose affinity column (Pharmacia). Discrimination between anti-Tcx, and anti-ProTcx activity. In a preliminary ELISA experiment, 25 SLE patients were screened for the presence of antibodies against thymosin 01, in their sera. Six of them appeared with high titer anti-Ta, activity, while their anti-ProTa activity remained at values below the cutoff (termed negative). Apart from this, sera positive in the anti-ProTq ELISA did not exhibit any anti-Ta, activity and this was further confirmed with inhibition experiments (see Results). In order to correlate the anti-ProTa activity with the immunoglobulin level, the absorbance (A) content value of every serum sample was divided by the total immunoglobulin content (Total Ig) and was expressed as the percentage of the total Ig: Total Ig = [(Ado5 x lOOO)/(Total Ig, mg)]%. Correlation of anti-ProTcx antibodies with disease activity and immunological indices. SLE activity was correlated with anti-ProTol antibodies by x2 test analysis. Linear regression analysis was performed to examine correlation of the presence of anti-ProTa antibodies with the presence of anti-dsDNA antibodies, the levels of immunoglobulins and complement components C, and C,. RESULTS
1. ELISA for Anti-ProTa (a) Amount was obtained (b) Working bance values
Antibodies
of ProTa per well. As shown in Fig. 1, a good discriminatory effect by using 1600 ng of ProTo per well. serum dilutions. Sera diluted I:50 gave the highest range of absorbetween positive and negative samples (Fig. 2).
VLACHOYIANNOPOULOS
ET AL.
- ~------._--
A
T
___0.
0’
B
0
0 8
0
n
0
000000 I
60
180
Time (min)
300
60
0 r
180 Time (mini
1
I
300
FIG. 1. Progress curve of ELISA obtained with a positive (closed symbols) and a negative (open symbols) serum. The plate was coated with: (A) 100 (O), 200 (0) or 800 ng (A) ProTa/well and (B) 1600 ng ProTa/well. Dilution of the sera was 1:25. The plate was not precoated with PLL.
(c) The effect of PLL. Figure 3 shows that precoating the plate with PLL resulted in the loss of anti-ProTcx binding to the antigen. (6) Standardization of the incubation time with substrate solution. As shown in Fig. 1, after 90 min the upper limit of the linearity of the progress curve was obtained. At this point, the reaction was stopped and the absorbance was read at 405 nm.
1
1:X
1:50
1:lOO 1:200 Dilution
1:LOO MOO of sera
1:lEOO 1:3200
FIG. 2. Titration curve of ELISA obtained with a positive (0) and a negative (A) serum. The plate was coated with 1600 ng ProTcLlwell. The set-awere applied in serial dilutions. The reaction of alkaline phosphatase was stopped at 240 min, when the substrate was exhausted in the low dilution of the positive serum.
PROTHYMOSIN
cx AND SYSTEMIC
155
LUPUS ERYTHEMATOSUS
j-
.O.
l
0 I-
...AAAAAAA I ’ I 60
120
r ,
7 I
1 ,
180
2hcl
300
TimeImin)
. 0. AAAAA ~:AAA I 7 , 1 I 60
120
160
I
2LO
’
I
300
Timelmin)
FIG. 3. Progress curve of ELISA without precoating the plate (A) or after precoating with PLL (B). The plates were coated with 1600 ng ProTa/well. Dilutions of the positive (0) and negative (A) sera were 150. In (B), the plate was precoated with 50 kg/ml PLL for 30 min.
2. The Specificity
of ELISA
(a) The purity of the antigen. Prothymosin (Ywas purified from the spleen of a young female goat, as described under Materials and Methods. The purity of the isolated peptide was then confirmed with IEF (pZ = 3.5) and the quantity was calculated on the basis of amino acid analysis. (b) Inhibition experiments. Sera with other specificities, like anti-dsDNA and anti-thymosin (Y, activity, were negative in anti-ProTa ELISA. In addition, a strongly positive anti-ProTol serum was preincubated with different amounts of ProTa, thymosin CQ,and dsDNA, ranging from 1 r&ml to 250 kg/ml, and then the serum was tested in the anti-proTa ELISA. Figure 4 shows that preincubation of the serum with ProTa completely inhibits the reaction, while no inhibition is observed on preincubation with dsDNA or thymosin cx,. 3. Anti-ProTol Antibodies and Anti-dsDNA Antibodies Are Idiotypically Distinct A serum with strong anti-DNA and anti-ProTa activity was passed through a dsDNA-cellulose column. Two peaks were obtained: the first, in the flowthrough volume, represents the unbound proteins and the second, eluted with 1 M NaCl, represents antibodies to dsDNA. Both peaks were dialysed overnight against PBS, and were tested in anti-dsDNA and anti-ProTa ELISA. Figure 5 shows that the first peak (unbound proteins) has only anti-ProTa activity, while the second peak (anti-DNA) has only anti-dsDNA activity. 4. Frequency of Anti-ProTcx Antibodies in SLE Patients: Correlation with the Clinical Picture Eight out of 44 SLE patients (18%), as well as 5 out of 276 healthy individuals (1.8%), were positive for anti-ProTa antibodies (Fig. 6). Linear regression analysis to study the correlation between anti-ProTo and anti-dsDNA activity in the sera of SLE patients revealed a significant correlation (r = +0.5%, n = 36, P <
156
VLACHOYIANNOPOULOS lOOr---------
El' AL. - ~--~~~~~-
/
4
15
60
250
Antigen,pg/ml
FIG. 4. Lack of inhibition of anti-ProTa by dsDNA (W) or To, (A). Control: ProTa (0). The plate was coated with 1600 ng ProTa/well. The positive serum was preincubated at 37”C, for 2 hr, with different concentrations of dsDNA, Tar, or ProTa, and was then added to the ELISA plate. Final dilution of the serum was 150. The reaction was stopped at 90 min. The inhibition was expressed according to (23) Inhibition (%) = 1 - ‘(“A:ykIf 1 :AAb,@“,d:
x
100.
background
Concentrations of DNA, Ta,. or ProTa lower
than 1 &ml
did not inhibit the ELISA.
0.001) (Fig. 7). No significant positive or negative correlation with the levels of serum immunoglobulins and the levels of C, or C, was found between anti-ProTa positive and negative individuals. Furthermore, no significant association between the anti-ProTol activity of the
Ill
1116
114
Dilution
of
serum
Dilution
of
serum
5. Discrimination between anti-ProTa and anti-DNA activity in the positive serum. (A) Anti-ProTa binding in the whole serum and in unbound and bound peaks to a dsDNA-cellulose column. (B) The same experiments performed with anti-dsDNA binding. FIG.
PROTHYMOSIN
cx AND
SYSTEMIC
LUPUS
ERYTHEMATOSUS
157
.. .
3SD 1 SD
-.Normals
SLE (n:LL)
In=2761
FIG. 6. Distribution of anti-ProTa antibodies in the sera of 276 normal individuals and 44 SLE patients. The cutoff point (A,, = 0.227) was calculated by taking three standard deviations above the mean of the normal values.
serum and the disease activity was found by the x2 test (x2 = 1.239, 0.1 < P < 0.5). 5. Anti-ProTa
Antibody Divided by Serum Ig
The normal range of the percentage of Protcu specific antibody was O-5%. All of the sera from patients with SLE found to be positive exhibited ratios that were two or more times higher than the upper limit of the normal range. The results clearly indicated that the anti-ProTo activity is not influenced by the immunoglobulin level. DISCUSSION
Using the ELBA method described, we have investigated the sera of SLE patients for the presence of antibodies against two immunoactive thymic peptides, namely thymosin o1 and prothymosin ct. Sera with high anti-Ta, activity were not positive in the anti-ProTa ELBA. They exhibited. however, low anti-ProTa titers, ascribed to the expected cross-reactivity with ProTa (27). Sera with elevated anti-ProTa activity did not exhibit either anti-Ta, activity or inhibition by Ta,, suggesting that these antibodies recognize an epitope on the
158
VLACHOYIANNOPOULOS
ET AL.
.
I
. . . I ov
=, 0
FIG. 7. Linear correlation patients tested.
1
500
I
1000 Anti-DNA(Binding
between anti-ProTa
I
I
1500 2ooo index)
and anti-dsDNA
2500
antibodies in the sera of SLE
C-terminal portion of ProTol (10). Although the immunoenhancing role of To, (corresponding to the N-terminal sequence of ProTo) is well known and established (3, 5, 7, 8), several lines of evidence recently suggest that the C-terminal portion of ProTo is also of importance in determining the biological activity of the molecule (11, 12, 18). Furthermore, the existence of ProTo mRNA in human fibroblasts (19), lymphocytes (19), and spleen cells (20), supports the physiological relevance of ProTa rather than To,. These data made us focus on the study of anti-ProTo antibodies. Prothymosin OLis a molecule that shares in common some important biochemical properties with DNA: Acidity (lo), poor immunogenicity (2, 27), and existence in a large variety of tissues (H-19). Thus, it is not surprising that there may exist certain analogies between the ProTa and the DNA antibody-antigen systems. The significant positive correlation found between anti-ProTo and anti-dsDNA activity in the sera of SLE patients is not due to any cross-reactivity of the anti-DNA antibodies with acid protein material (28), because both inhibition experiments and affinity chromatography studies demonstrated that anti-ProTol antibodies and anti-dsDNA antibodies are totally distinct from each other. The fact that high titer anti-ProTol antibodies cannot be readily induced by immunization (27) and the finding that low levels of anti-ProTa activity are also present in a small proportion of normal individuals argue for the genesis of anti-ProTo autoantibodies by polyclonal or oligoclonal B-cell activation, rather than by antigen presentation alone. Similar observations have been made for anti-native DNA antibodies (2). Nevertheless, the question of whether the autoimmune response results from a polyclonal B-cell activation or an unusual antigen presentation is a matter of controversy. In fact, serum autoantibodies to ubiquitin, a heat shock protein frequently found in SLE patients, have been sug-
PROTHYMOSIN
(Y AND SYSTEMIC
LUPUS ERYTHEMATOSUS
159
gested to be formed against ubiquitin-protein conjugates present during cellular injury and this represents a major characteristic of immune response in SLE (29). As with DNA and ProTo, antibodies to ubiquitin cannot be readily induced by immunization. Even though anti-dsDNA antibodies are currently used as a marker for SLE activity (2, 23) and despite the good correlation between anti-dsDNA and anti-ProTa activities, our results do not indicate any clinical significance for the presence of anti-ProTo antibodies in SLE patients. It is also known that the reactions between DNA antibody and antigen in serum have the capacity to fix complement (2). Thus, the anti-DNA activity of sera with low complement levels could explain a possible negative correlation between anti-ProTo and the levels of complement components C, or C,; in fact, such a correlation was not observed. Nevertheless, deposition of antigen-antibody complexes in peripheral tissues resulting in local tissue damage should not be excluded. Anti-ProTa antibodies in SLE may either modulate the immune system by blocking the ProTo activity or cause immune complex-mediated organ lesions. Their possible pathogenetic role remains to be elucidated. Serial studies in high titer SLE patients, as well as attempts to produce experimental models of the disease by injection of antigen-antibody complexes, may be considered. In addition, anti-ProTa antibodies found in the sera of SLE patients may prove important tools for biochemical studies on prothymosin (Y because of their high titers and unique specificity. ACKNOWLEDGMENT We thank Ms. E. E. Papanikolaou for excellent secretarial assistance.
REFERENCES 1. Steinberg, A. D., Raveche, E. S., La&in, C. A., Smith, H. R.. Santon. T., Miller, M. L., and Plotz, P. H., Ann. In?. Med. 100, 714, 1984. 2. Tan, E. M., Adv. Immunol. 33, 177. 1982. 3. Ahmed, A. D., Wong, D., Thurman, G. B., Low, T. L. K.. Goldstein, A. L., Sharkis, S. J., and Goldschneider, I., Ann. N. Y. Acad. Sci. 332, 81, 1979. 4. Favalli, C., Terri, T., Mastino, A., Rinaldi-Garaci, C., and Garaci, E., Cancer Immunol. Immunother. 20, 189, 1985. 5. Serrate, S. A., Schulof, R. S., Leondaridis, L., Goldstein, A., and Sztein. M. B., .I. Immunol. 139, 2338, 1987. 6. Ishitsuka, H., Umeda, Y., Nakamura, J., and Yagi, Y., Cancer Zmmunol. Immunother. 14, 145. 1983. 7. Baxevanis, C. N., Reclos, G. J., Perez, S., Kokkinopoulos, D., and Papamichail, M., Zmmunopharmacology 13, 133, 1987. 8. Schulof. R. S., Lloyd, M. L., Cleary, P. A., Palaszynski, S. R., Mai, D. A., Cox, J. W., Jr., Alabaster, O., and Goldstein, A. L., J. Biol. Response Modif. 4, 147, 1985. 9. Haritos, A. A., Goodall, G. J., and Horecker, B. L., Proc. Nutl. Acad. Sci. USA 81, 1008, 1984. 10. Haritos, A. A., Blather, R.. Stein, S., Caldarella, J., and Horecker, B. L., Proc. Narl. Acad. Sci. USA 82, 343, 1985. 11. Haritos, A. A., Salvin, S. B., Blather, R., Stein, S., and Horecker, B. L., Proc. Natl. Acad. Sci. USA 82, 1050, 1985. 12. Baxevanis, C. N., Reclos, G. J., Panneerselvam, C., and Papamichail, M., Immunopharmacology 15, 73, 1988.
160
VLACHOYIANNOPOULOS
E7
AL
13. Reclos. G. J.. Baxevanis, C. N., Sfagos, C., Papageorgiou, C.. Tsokos. G. C., and Papamichail. M., Clin. Exp. Immunol. IQ, 336. 1987. 14. Baxevanis, C. N., Reclos. G. J., Papamichail, M.. and Tsokos, G. C.. Immunophrmacd.[mmunotoxicol. 9, 426, 1987. 1.5. Haritos, A. A., Tsolas, 0.. and Horecker, B. L., Proc. Nat/. Acad. ,Sci. uSA 81, 1391, 1984. 16. Economou, M., Seferiadis, K.. Frangou-Lazaridis. M., Horecker. B. L., and Tsolas. O., f%R,‘$ Lett. 233, 342, 1988. 17. Frillingos, S., Frangou-Lazaridis, M., Seferiadis, K., Horecker. B. L., and Tsolas, O., Bid/. Chem. Hoppe-Seyler 367, Suppl., 252, 1986. 18. Pan, L. X., Haritos, A. A., Wideman, J.. Komiyama, T., Chang, M., Stem, S., Salvtn, S. B., and Horecker, B. L., Arch. Biochem. Biophys. 250, 197. 1986. 19. Eschenfeldt, W. H., and Berger, S. L., Proc. Natl. Acad. Sci. USA 83, 9403, 1986. 20. Goodall, G. J., Dominguez, F., and Horecker, B. L.. froc. Nat/. Acad. Sci. US.4 83, 8926, 1986. 21. Frangou-Lazaridis, M., Clinton. M., Goodall, G. J.. and Horeck, B. L.. Arch. B&hem, Bjophys. 263, 305, 1988. 22. Moutsopoulos, H. M., Fye, K. H., Sawada, S., Becker, M. J., Goldstein, A., and Tala), N., C/in. Exp. Immunol. 26, S63. 1976. 23. Tzioufas, A. G., Manoussakis, M. N.. Drosos, A. A., Silis, G.. Gharavi. A. E.. and Moutsopoulos, H. M., Clin. Exp. Rheumatol. 5, 247, 1987. 24. Kurata, N., and Tan. E. M., Arthritis Rheum. 19, 574, 1976. 25. Komiyama, T.. Pan, L. X.. Haritos, A. A., Wideman, J. W., Pan, Y.-C. E.. Chang, M., Rogers, I., and Horecker, B. L., Proc. Nutl. Acad. Sci. USA 83, 1242, 1986. 26. Heinrikson, R. L.. and Meredith, S. C., Anal. Biochem. 136, 65, 1984. 27. Haritos, A. A., and Horecker, B. L., J. Immunol. Methods 81, 199, 1985. 28. Jacob, L.. Lety, M. A., Bach, J. F., and Louvard, D., In “Eurorheumatology” (A. A. Andrianakos, I. Kappou. M. Mavrikakis. and H. M. Moutsopoulos, Eds.), pp. 13-18, Tagas and Son Press, Athens, 1987. 29. Muller. S., Briand. J.-P., and Van Regenmortel, M. H. V.. Proc. Nat/. Acud. Sci. UsA 85, 8176, 1988. Received October 25, 1988; accepted with revision April 19, 1989
IMMUNOLOGY
Circulating
AND
IMMUNOPATHOLOGY
53, 151-160 (1989)
Antibodies to Prothymosin Lupus Erythematosus
(Y in Systemic
P. G. VLACHOYIANNOPOULOS,* S. FRILLINGOS,? A. G. TZIOUFAS,* K. SEFERIADIS,? H. M. MOUTSOPOULOS,* AND 0. TSOLAS? *Department of Internal Medicine and fLaboratory of Biological Chemistry, Medical School, University of loannina, 451 10 loannina, Greece Autoantibodies to prothymosin (x, an imnnmoactive protein that exists in a large variety of mammalian tissues, were found to be present in patients with systemic lupus erythematosus (SLE) by a new, sensitive, and specific anti-prothymosin a ELISA. The antigen was prothymosin a, purified by high-pressure liquid chromatography from goat spleen extracts. Sera from 44 SLE patients and 276 healthy individuals were screened for the presence of anti-prothymosin a activity; 18% of SLE sera were found to be positive, compared with 1.8% of control sera. This anti-prothymosin o! activity appears to be idiotypically distinct from either anti-thymosin o, activity or anti-dsDNA activity, as demonstrated by inhibition experiments. Sign&ant positive correlation exists between anti-prothymosin d and anti-dsDNA activities of SLE sera (r = +0.5%, n = 36, P < O.OOl), while no correlation was observed with the clinical activity (x2 = 1.239, 0.1 < P < 0.5) or with complement levels C, and C4. o 1989 Academic press, IIIC.
INTRODUCTION
Systemic lupus erythematosus (SLE) is a systemic autoimmune disease characterized by the presence of numerous nonorgan specific autoantibodies (1). The involvement of these autoantibodies in the pathogenesis of the disease is yet unclear; nevertheless, anti-dsDNA autoantibodies have been shown to play a central role in immune complex-mediated tissue damage (2) and are currently used as a criterion of diagnosis for the disease (2). The primary defect accounting for this autoreactivity is either impaired T-cell suppressor function or B-cell hyperactivity, known to be the main immunoregulatory abnormalities in SLE (1). Autoantibodies against lymphocytes (1) and aberrant production of T-cell differentiation and growth factors (1) have been described to occur in SLE. One set of immunoactive peptides, derived from the thymus, plays a central role in the differentiation and maturation of T-lymphocytes. Thymosin o1 (To, 3108 Da), in particular, has been shown to enhance the expression of Thy-l,2 and Lyt-1,2,3 antigens on the T-cell (3), enhance natural killer (NK) activity in vivo (4) and in vitro (5), protect sensitive or immunosuppressed mice against opportunistic infections (6), enhance the human mixed lymphocyte response (MLR) by acting on T4+ cells (7), and improve survival, in conjunction with radiotherapy of immunosuppressed lung cancer patients (8). Of particular interest is prothymosin a (ProTa), a highly acidic (pZ = 3.55) polypeptide of 12,500 Da (9), which contains the entire thymosin a, sequence at its amino terminal (IO), and an acidic stretch in the central portion of the molecule 151 0090-1229/89 $1.50 Copy&M Au rights
0 1989 by Academic Press, Inc. of rcpmdwXion in any form named.
I52
VLACHOYIANNOPOULOS
El‘
AI..
(10). Prothymosin a appears to be more immunoactive than thymosin ai in certain and in vitro (12) assays, i.e., in protecting sensitive mice against infections with Can&da alhicuns (I I) and enhancing the human autologous or allogeneic MLR (12). Prothymosin a has also been reported to restore suboptimal human MLR (12). as well as deficient MLR in lymphocytes from patients with clinically active multiple sclerosis or SLE (13, 14). Although first isolated from rat thymus (9). prothymosin (Y is now known to have a wide distribution in the tissues of mammals, such as rat (15), pig (16), goat (17). and man (18). Moreover. the wide tissue distribution of prothymosin (Y mRNA (19) gives evidence for the synthesis of prothymosin (Ynot only in thymus, but also in other tissues, especially the spleen (19-21). Incubation of SLE peripheral blood lymphocytes with thymic extracts or thymic peptides (22) increases in vitro the number and function of T-cells. These observations suggest that thymic peptide deficiency may be present in SLE. To study if this deficiency is immunologically mediated, we searched for the presence of circulating autoantibodies in the sera of patients with SLE. in vivo (II)
MATERIALS
AND METHODS
Patients and sera. Sera from 44 SLE patients and from 276 healthy individuals as controls were studied. The sera were kept at -20°C until tested. The anti-DNA antibodies were detected with an ELISA method described previously (23). Autoantibodies to nuclear and cytoplasmic antigens were tested by counterimmunoelectrophoresis (24). The levels of the complement C, and C4 were measured by radial immunodiffusion (Boehringer). SLE was considered active when the patient presented with at least one of the following manifestations: serositis, arthritis, nephritis, myositis, vasculitis, and central nervous system involvement, as well as high serum DNA binding or low levels of complement. Preparation of ProTcx. Prothymosin a was purified from the spleen of a young female goat by gel filtration on Sephacryl S-200 (Pharmacia) and reversed-phase HPLC on Altex Ultrasphere ODS-C,, columns (Beckman Instruments), following a procedure described by others (25). The purity of the peptide was confirmed by analytical isoelectric focusing (IEF) performed on a LKB Ampholine PAG plate (LKB); IEF revealed a single band at the characteristic isoelectric point for ProTo (~1 = 3.5). Quantitative determinations of the peptide, used for coating the ELISA plates, were based on amino acid analyses, performed on hydrolysates as the phenylthiocarbamyl derivatives (26). Synthetic thymosin oi used in some experiments was a generous gift from A. M. Felix of Hoffmann-La Roche (Nutley, NJ). ELZSA. Prothymosin (Yat different quantities per well was applied to polystyrene cuvettes, either precoated or not precoated with poly-L-lysine (PLL) (Sigma) 50 pg/ml in PBS. The quantities of the antigen per well, tested in several preliminary experiments, were 50, 100, 200, 400, 800, and 1600 ng. In order to reduce the nonspecific binding of serum immunoglobulin, especially in hypergammaglobulinemic sera, 10% bovine serum (BS) was used as a blocking agent (23).
PROTHYMOSIN
a
AND
SYSTEMIC
LUPUS
ERYTHEMATOSUS
153
A strongly positive SLE serum as well as a negative control were tested in serial dilutions from I:25 to 1:3200, and the tests were repeated for all of the quantities of antigen mentioned. The reaction was measured at 15 min intervals, at 37”C, for 4 hr. The best discriminatory effect between positive and negative controls was obtained using 1600 ng of antigen (ProTar) per well, a working dilution of the sera 1:50, and incubation time with the substrate of 90 min, which is in the linear portion of the reaction. Standard conditions for ELISA were set as follows: ProTa (1600 ng per well) in coating buffer 10 mM Na,C03, 30 mM NaHCO,, pH 9.6, was incubated in polystyrene cuvettes, at 37°C for 1 hr. After washing with PBS, the nonspecific binding sites were blocked with 10% BS at 37”C, for I hr. Fifty microliters of serum diluted 150 in 10% BS was added in duplicate and incubated at 37°C for 3 hr. After washing, antibody bound to ProTa was detected with alkaline phosphatase antihuman y chain-specific antisera (Sigma), diluted 1:lOOO in 10% BS. Substrate solution (p-nitrophenyl-phosphate disodium) 1 mg/ml in 10% diethanolamine, 100 mg/liter MgC1,6H,O, pH 9.8, was added to each well and the absorbance was read at 405 nm. Sera were considered to be positive when the absorbance was higher than the mean of normals plus three standard deviations. Discrimination between anti-dsDNA and anti-ProTa activity. This was first examined by inhibition experiments which showed that the binding of strongly positive anti-ProTa sera was not inhibited by preincubation with dsDNA, and second, by affinity chromatography of a serum with both anti-DNA and anti-ProTa activities on a dsDNA-cellulose affinity column (Pharmacia). Discrimination between anti-Tcx, and anti-ProTcx activity. In a preliminary ELISA experiment, 25 SLE patients were screened for the presence of antibodies against thymosin 01, in their sera. Six of them appeared with high titer anti-Ta, activity, while their anti-ProTa activity remained at values below the cutoff (termed negative). Apart from this, sera positive in the anti-ProTq ELISA did not exhibit any anti-Ta, activity and this was further confirmed with inhibition experiments (see Results). In order to correlate the anti-ProTa activity with the immunoglobulin level, the absorbance (A) content value of every serum sample was divided by the total immunoglobulin content (Total Ig) and was expressed as the percentage of the total Ig: Total Ig = [(Ado5 x lOOO)/(Total Ig, mg)]%. Correlation of anti-ProTcx antibodies with disease activity and immunological indices. SLE activity was correlated with anti-ProTol antibodies by x2 test analysis. Linear regression analysis was performed to examine correlation of the presence of anti-ProTa antibodies with the presence of anti-dsDNA antibodies, the levels of immunoglobulins and complement components C, and C,. RESULTS
1. ELISA for Anti-ProTa (a) Amount was obtained (b) Working bance values
Antibodies
of ProTa per well. As shown in Fig. 1, a good discriminatory effect by using 1600 ng of ProTo per well. serum dilutions. Sera diluted I:50 gave the highest range of absorbetween positive and negative samples (Fig. 2).
VLACHOYIANNOPOULOS
ET AL.
- ~------._--
A
T
___0.
0’
B
0
0 8
0
n
0
000000 I
60
180
Time (min)
300
60
0 r
180 Time (mini
1
I
300
FIG. 1. Progress curve of ELISA obtained with a positive (closed symbols) and a negative (open symbols) serum. The plate was coated with: (A) 100 (O), 200 (0) or 800 ng (A) ProTa/well and (B) 1600 ng ProTa/well. Dilution of the sera was 1:25. The plate was not precoated with PLL.
(c) The effect of PLL. Figure 3 shows that precoating the plate with PLL resulted in the loss of anti-ProTcx binding to the antigen. (6) Standardization of the incubation time with substrate solution. As shown in Fig. 1, after 90 min the upper limit of the linearity of the progress curve was obtained. At this point, the reaction was stopped and the absorbance was read at 405 nm.
1
1:X
1:50
1:lOO 1:200 Dilution
1:LOO MOO of sera
1:lEOO 1:3200
FIG. 2. Titration curve of ELISA obtained with a positive (0) and a negative (A) serum. The plate was coated with 1600 ng ProTcLlwell. The set-awere applied in serial dilutions. The reaction of alkaline phosphatase was stopped at 240 min, when the substrate was exhausted in the low dilution of the positive serum.
PROTHYMOSIN
cx AND SYSTEMIC
155
LUPUS ERYTHEMATOSUS
j-
.O.
l
0 I-
...AAAAAAA I ’ I 60
120
r ,
7 I
1 ,
180
2hcl
300
TimeImin)
. 0. AAAAA ~:AAA I 7 , 1 I 60
120
160
I
2LO
’
I
300
Timelmin)
FIG. 3. Progress curve of ELISA without precoating the plate (A) or after precoating with PLL (B). The plates were coated with 1600 ng ProTa/well. Dilutions of the positive (0) and negative (A) sera were 150. In (B), the plate was precoated with 50 kg/ml PLL for 30 min.
2. The Specificity
of ELISA
(a) The purity of the antigen. Prothymosin (Ywas purified from the spleen of a young female goat, as described under Materials and Methods. The purity of the isolated peptide was then confirmed with IEF (pZ = 3.5) and the quantity was calculated on the basis of amino acid analysis. (b) Inhibition experiments. Sera with other specificities, like anti-dsDNA and anti-thymosin (Y, activity, were negative in anti-ProTa ELISA. In addition, a strongly positive anti-ProTol serum was preincubated with different amounts of ProTa, thymosin CQ,and dsDNA, ranging from 1 r&ml to 250 kg/ml, and then the serum was tested in the anti-proTa ELISA. Figure 4 shows that preincubation of the serum with ProTa completely inhibits the reaction, while no inhibition is observed on preincubation with dsDNA or thymosin cx,. 3. Anti-ProTol Antibodies and Anti-dsDNA Antibodies Are Idiotypically Distinct A serum with strong anti-DNA and anti-ProTa activity was passed through a dsDNA-cellulose column. Two peaks were obtained: the first, in the flowthrough volume, represents the unbound proteins and the second, eluted with 1 M NaCl, represents antibodies to dsDNA. Both peaks were dialysed overnight against PBS, and were tested in anti-dsDNA and anti-ProTa ELISA. Figure 5 shows that the first peak (unbound proteins) has only anti-ProTa activity, while the second peak (anti-DNA) has only anti-dsDNA activity. 4. Frequency of Anti-ProTcx Antibodies in SLE Patients: Correlation with the Clinical Picture Eight out of 44 SLE patients (18%), as well as 5 out of 276 healthy individuals (1.8%), were positive for anti-ProTa antibodies (Fig. 6). Linear regression analysis to study the correlation between anti-ProTo and anti-dsDNA activity in the sera of SLE patients revealed a significant correlation (r = +0.5%, n = 36, P <
156
VLACHOYIANNOPOULOS lOOr---------
El' AL. - ~--~~~~~-
/
4
15
60
250
Antigen,pg/ml
FIG. 4. Lack of inhibition of anti-ProTa by dsDNA (W) or To, (A). Control: ProTa (0). The plate was coated with 1600 ng ProTa/well. The positive serum was preincubated at 37”C, for 2 hr, with different concentrations of dsDNA, Tar, or ProTa, and was then added to the ELISA plate. Final dilution of the serum was 150. The reaction was stopped at 90 min. The inhibition was expressed according to (23) Inhibition (%) = 1 - ‘(“A:ykIf 1 :AAb,@“,d:
x
100.
background
Concentrations of DNA, Ta,. or ProTa lower
than 1 &ml
did not inhibit the ELISA.
0.001) (Fig. 7). No significant positive or negative correlation with the levels of serum immunoglobulins and the levels of C, or C, was found between anti-ProTa positive and negative individuals. Furthermore, no significant association between the anti-ProTol activity of the
Ill
1116
114
Dilution
of
serum
Dilution
of
serum
5. Discrimination between anti-ProTa and anti-DNA activity in the positive serum. (A) Anti-ProTa binding in the whole serum and in unbound and bound peaks to a dsDNA-cellulose column. (B) The same experiments performed with anti-dsDNA binding. FIG.
PROTHYMOSIN
cx AND
SYSTEMIC
LUPUS
ERYTHEMATOSUS
157
.. .
3SD 1 SD
-.Normals
SLE (n:LL)
In=2761
FIG. 6. Distribution of anti-ProTa antibodies in the sera of 276 normal individuals and 44 SLE patients. The cutoff point (A,, = 0.227) was calculated by taking three standard deviations above the mean of the normal values.
serum and the disease activity was found by the x2 test (x2 = 1.239, 0.1 < P < 0.5). 5. Anti-ProTa
Antibody Divided by Serum Ig
The normal range of the percentage of Protcu specific antibody was O-5%. All of the sera from patients with SLE found to be positive exhibited ratios that were two or more times higher than the upper limit of the normal range. The results clearly indicated that the anti-ProTo activity is not influenced by the immunoglobulin level. DISCUSSION
Using the ELBA method described, we have investigated the sera of SLE patients for the presence of antibodies against two immunoactive thymic peptides, namely thymosin o1 and prothymosin ct. Sera with high anti-Ta, activity were not positive in the anti-ProTa ELBA. They exhibited. however, low anti-ProTa titers, ascribed to the expected cross-reactivity with ProTa (27). Sera with elevated anti-ProTa activity did not exhibit either anti-Ta, activity or inhibition by Ta,, suggesting that these antibodies recognize an epitope on the
158
VLACHOYIANNOPOULOS
ET AL.
.
I
. . . I ov
=, 0
FIG. 7. Linear correlation patients tested.
1
500
I
1000 Anti-DNA(Binding
between anti-ProTa
I
I
1500 2ooo index)
and anti-dsDNA
2500
antibodies in the sera of SLE
C-terminal portion of ProTol (10). Although the immunoenhancing role of To, (corresponding to the N-terminal sequence of ProTo) is well known and established (3, 5, 7, 8), several lines of evidence recently suggest that the C-terminal portion of ProTo is also of importance in determining the biological activity of the molecule (11, 12, 18). Furthermore, the existence of ProTo mRNA in human fibroblasts (19), lymphocytes (19), and spleen cells (20), supports the physiological relevance of ProTa rather than To,. These data made us focus on the study of anti-ProTo antibodies. Prothymosin OLis a molecule that shares in common some important biochemical properties with DNA: Acidity (lo), poor immunogenicity (2, 27), and existence in a large variety of tissues (H-19). Thus, it is not surprising that there may exist certain analogies between the ProTa and the DNA antibody-antigen systems. The significant positive correlation found between anti-ProTo and anti-dsDNA activity in the sera of SLE patients is not due to any cross-reactivity of the anti-DNA antibodies with acid protein material (28), because both inhibition experiments and affinity chromatography studies demonstrated that anti-ProTol antibodies and anti-dsDNA antibodies are totally distinct from each other. The fact that high titer anti-ProTol antibodies cannot be readily induced by immunization (27) and the finding that low levels of anti-ProTa activity are also present in a small proportion of normal individuals argue for the genesis of anti-ProTo autoantibodies by polyclonal or oligoclonal B-cell activation, rather than by antigen presentation alone. Similar observations have been made for anti-native DNA antibodies (2). Nevertheless, the question of whether the autoimmune response results from a polyclonal B-cell activation or an unusual antigen presentation is a matter of controversy. In fact, serum autoantibodies to ubiquitin, a heat shock protein frequently found in SLE patients, have been sug-
PROTHYMOSIN
(Y AND SYSTEMIC
LUPUS ERYTHEMATOSUS
159
gested to be formed against ubiquitin-protein conjugates present during cellular injury and this represents a major characteristic of immune response in SLE (29). As with DNA and ProTo, antibodies to ubiquitin cannot be readily induced by immunization. Even though anti-dsDNA antibodies are currently used as a marker for SLE activity (2, 23) and despite the good correlation between anti-dsDNA and anti-ProTa activities, our results do not indicate any clinical significance for the presence of anti-ProTo antibodies in SLE patients. It is also known that the reactions between DNA antibody and antigen in serum have the capacity to fix complement (2). Thus, the anti-DNA activity of sera with low complement levels could explain a possible negative correlation between anti-ProTo and the levels of complement components C, or C,; in fact, such a correlation was not observed. Nevertheless, deposition of antigen-antibody complexes in peripheral tissues resulting in local tissue damage should not be excluded. Anti-ProTa antibodies in SLE may either modulate the immune system by blocking the ProTo activity or cause immune complex-mediated organ lesions. Their possible pathogenetic role remains to be elucidated. Serial studies in high titer SLE patients, as well as attempts to produce experimental models of the disease by injection of antigen-antibody complexes, may be considered. In addition, anti-ProTa antibodies found in the sera of SLE patients may prove important tools for biochemical studies on prothymosin (Y because of their high titers and unique specificity. ACKNOWLEDGMENT We thank Ms. E. E. Papanikolaou for excellent secretarial assistance.
REFERENCES 1. Steinberg, A. D., Raveche, E. S., La&in, C. A., Smith, H. R.. Santon. T., Miller, M. L., and Plotz, P. H., Ann. In?. Med. 100, 714, 1984. 2. Tan, E. M., Adv. Immunol. 33, 177. 1982. 3. Ahmed, A. D., Wong, D., Thurman, G. B., Low, T. L. K.. Goldstein, A. L., Sharkis, S. J., and Goldschneider, I., Ann. N. Y. Acad. Sci. 332, 81, 1979. 4. Favalli, C., Terri, T., Mastino, A., Rinaldi-Garaci, C., and Garaci, E., Cancer Immunol. Immunother. 20, 189, 1985. 5. Serrate, S. A., Schulof, R. S., Leondaridis, L., Goldstein, A., and Sztein. M. B., .I. Immunol. 139, 2338, 1987. 6. Ishitsuka, H., Umeda, Y., Nakamura, J., and Yagi, Y., Cancer Zmmunol. Immunother. 14, 145. 1983. 7. Baxevanis, C. N., Reclos, G. J., Perez, S., Kokkinopoulos, D., and Papamichail, M., Zmmunopharmacology 13, 133, 1987. 8. Schulof. R. S., Lloyd, M. L., Cleary, P. A., Palaszynski, S. R., Mai, D. A., Cox, J. W., Jr., Alabaster, O., and Goldstein, A. L., J. Biol. Response Modif. 4, 147, 1985. 9. Haritos, A. A., Goodall, G. J., and Horecker, B. L., Proc. Nutl. Acad. Sci. USA 81, 1008, 1984. 10. Haritos, A. A., Blather, R.. Stein, S., Caldarella, J., and Horecker, B. L., Proc. Narl. Acad. Sci. USA 82, 343, 1985. 11. Haritos, A. A., Salvin, S. B., Blather, R., Stein, S., and Horecker, B. L., Proc. Natl. Acad. Sci. USA 82, 1050, 1985. 12. Baxevanis, C. N., Reclos, G. J., Panneerselvam, C., and Papamichail, M., Immunopharmacology 15, 73, 1988.
160
VLACHOYIANNOPOULOS
E7
AL
13. Reclos. G. J.. Baxevanis, C. N., Sfagos, C., Papageorgiou, C.. Tsokos. G. C., and Papamichail. M., Clin. Exp. Immunol. IQ, 336. 1987. 14. Baxevanis, C. N., Reclos. G. J., Papamichail, M.. and Tsokos, G. C.. Immunophrmacd.[mmunotoxicol. 9, 426, 1987. 1.5. Haritos, A. A., Tsolas, 0.. and Horecker, B. L., Proc. Nat/. Acad. ,Sci. uSA 81, 1391, 1984. 16. Economou, M., Seferiadis, K.. Frangou-Lazaridis. M., Horecker. B. L., and Tsolas. O., f%R,‘$ Lett. 233, 342, 1988. 17. Frillingos, S., Frangou-Lazaridis, M., Seferiadis, K., Horecker. B. L., and Tsolas, O., Bid/. Chem. Hoppe-Seyler 367, Suppl., 252, 1986. 18. Pan, L. X., Haritos, A. A., Wideman, J.. Komiyama, T., Chang, M., Stem, S., Salvtn, S. B., and Horecker, B. L., Arch. Biochem. Biophys. 250, 197. 1986. 19. Eschenfeldt, W. H., and Berger, S. L., Proc. Natl. Acad. Sci. USA 83, 9403, 1986. 20. Goodall, G. J., Dominguez, F., and Horecker, B. L.. froc. Nat/. Acad. Sci. US.4 83, 8926, 1986. 21. Frangou-Lazaridis, M., Clinton. M., Goodall, G. J.. and Horeck, B. L.. Arch. B&hem, Bjophys. 263, 305, 1988. 22. Moutsopoulos, H. M., Fye, K. H., Sawada, S., Becker, M. J., Goldstein, A., and Tala), N., C/in. Exp. Immunol. 26, S63. 1976. 23. Tzioufas, A. G., Manoussakis, M. N.. Drosos, A. A., Silis, G.. Gharavi. A. E.. and Moutsopoulos, H. M., Clin. Exp. Rheumatol. 5, 247, 1987. 24. Kurata, N., and Tan. E. M., Arthritis Rheum. 19, 574, 1976. 25. Komiyama, T.. Pan, L. X.. Haritos, A. A., Wideman, J. W., Pan, Y.-C. E.. Chang, M., Rogers, I., and Horecker, B. L., Proc. Nutl. Acad. Sci. USA 83, 1242, 1986. 26. Heinrikson, R. L.. and Meredith, S. C., Anal. Biochem. 136, 65, 1984. 27. Haritos, A. A., and Horecker, B. L., J. Immunol. Methods 81, 199, 1985. 28. Jacob, L.. Lety, M. A., Bach, J. F., and Louvard, D., In “Eurorheumatology” (A. A. Andrianakos, I. Kappou. M. Mavrikakis. and H. M. Moutsopoulos, Eds.), pp. 13-18, Tagas and Son Press, Athens, 1987. 29. Muller. S., Briand. J.-P., and Van Regenmortel, M. H. V.. Proc. Nat/. Acud. Sci. UsA 85, 8176, 1988. Received October 25, 1988; accepted with revision April 19, 1989