Aging 2: 347-355, 1990
ORIGINAL ARTICLES
Immune dysfunction in the elderly: Effect of thymic hormone administration on several in vivo and in vitro immune function parameters U. Fagiolo*, A. Amadori**, F. Borghesan*, R. Zamarchi**, M.L. Veronese**, G. De Silvestro***,
E. Passarella****, and G. Crepaldi*
* Institute of Internal Medicine and ** Institute of Oncology, University of Padova, Padova, *** Central Laboratory Unit of Padova General Hospital, Padova, **** Serono Pharmaceutical,
Milano, Italy.
ABSTRACT. The effects of short-term thymic hormone administration on age-associated immune function were evaluated. Two groups of individuals > 65 years of age were treated for 30 days with thymic extracts (TP1) or placebo; before and after this treatment a panel of in vitro and in vivo parameters was determined according to a very rigorous experimental protocol. In most individuals, TP1 treatment was associated with an improvement in cutaneous delayed-type response to PPD. Moreover, an increase in a circulating T cell subpopulation bearing the CD45R surface antigen ("virgin" T cells), and in NK cell cytotoxic activity was also observed in some subjects. Finally, lymphocyte responsiveness to PHA tended to increase, while no effect on lymphocyte ability to produce IL-2 following mitogen stimulation was observed. These findings suggest that TP1 treatment may influence age-related alterations in immune function parameters in some subjects. (Aging 2: 347-355, 1990) INTRODUCTION Immunodepression is associated with aging
in both animals and humans (1). Although different mechanisms probably underlie immune dysfunction in mice (2-4) and man (5-7), progressive involution of thymic function is a common feature (8, 9). Attempts to reconstitute immune function by in vivo thymic hormone administration were successful in mice (10-13), and associated with an increase in some immune function parameters in middle-aged subhuman primates (14). Studies in elderly persons are more sporadic (15, 16), but some findings indicate that such treatment may have beneficial effects in immunodeficiency patients (17-19). Longitudinal investigations into the effects of drugs on immune function are hampered in man by the high inter-person variability of the immune response, as well as the inter-test variability of in vitro assays, and time-dependent variations in immune function in individual subjects. The first two problems can be circumvented by observing a very strict experimental protocol; the last is not readily resolved, but can be absorbed by parallel studies in a matched group of untreated subjects. Under these conditions, we investigated whether short-term
Key words: Aging, thymic hormones, TP1, age-associated immune defect, immunotherapy. Correspondence: Umberto Fagiolo, M.D., Istituto di Medicina Interna, Via Giustiniani 2, 35128 Padova, Italy. Received December 18, 1989; accepted July 9, 1990.
Aging, Vol. 2, N. 4 347
U. Fagiolo, A. Amadori, F. Borghesan, et al.
thymic hormone treatment in elderly persons was associated with a change in some in vivo and in vitro immune function parameters, namely, delayed type response to PPD, cell surface marker expression, NK activity, early activation marker expression, IL-2 production and proliferative response to mitogens and antigens. METHODS
Subjects and experimental protocol All persons enrolled in this study fulfilled the EURAGE Senieur protocol criteria (20). Thirteen individuals> 65 years of age (6 males and 7 females; mean age 77 years, range 67-90) were injected i.m. with 70 mg of thymic hormone extract (TP1, kindly provided by Serono, Milano, Italy, and prepared as described in 21) daily for 7 days, and then twice a week for 3 weeks; before (day 0) and at the end of treatment (day 30) several immune function parameters were determined. A control group of 11 age- and sexmatched individuals was injected with placebo ' according to the same schedule, and analyzed , as above. Due to the low number of lymphocytes recovered in some cases, not all the assays were performed in every individual. A very strict experimental protocol was followed. Inter-test variability was circumscribed by testing equal numbers of TP1 -treated subjects and controls in a single session; moreover, the same persons forming a given group at evaluation on day 0 were re-examined all together on day 30. To assess inter-test variability, single assays were repeatedly performed in two normal young volunteers, and the standard deviation from the mean of replicate tests was calculat ed; variations observed on dav 30 in the experimental groups were considered significant only when they exceeded the 95% confidence limits of this variability. An additional control group consisted of 14 young members of the laboratory staff (mean age 27 years). Reagents Recombinant IL-2 (rIL-2) was obtained through the courtesy of Dr. F. Sinigaglia, Hoffmann-LaRoche, Basel, Switzerland. Phytohemagglutinin (pHA-P, Difco, Detroit, MI) was
348 Aging, Vol. 2, N. 4
used at 0.5 JLg/ml. Monoclonal antibody (mAb) OKT3 , kindly provided by Dr. Dunia Ramarli, Genova, Italy, was used at 10 ng/ml. Protein purified derivative (PPD), a gift of Sclavo, Siena , Italy, was used at 200 ng/ml. Staphylococcal protein A (SPA, Pharmacia, Uppsala, Sweden) was used at 100 JLg/ml. In vivo testing for cutaneous delayed-type response to recall antigens was performed using Multitest Merieux.
Lymphocyte isolation and culture Peripheral blood lymphocytes (PBL) were isolated from heparinized fasting blood samples collected between 8-9 a.m ., as previously described (22). Cells were resuspended in RPMI 1640 medium containing 10% fetal calf serum (FCS, Flow, Irvine, U.K.), 1% L-glutamine, 1% non-essential aminoacids, 2 x 10-sM 2-mercaptoethanol, and 50 JLg/ml gentamycin (complete RPMI). Mitogenic stimulation was accomplished by culturing PBL in 96-well U-bottom microtiter plates (Costar Data Packaging, Cambridge, MA) at a concentration of 1 x 106 cells/rnl in 0.2 ml complete RPMI in the presence of PHA or OKT3 for 72 hours at 37 -c, 5% CO2 , For antigenic stimulation, PBL were cultured for 144 hours in complete RPMI containing 10% human pooled serum in place of FCS in the presence of PPD or irradiated (3,000 R) pooled PBL from normal donors as allogeneic stimulators. Eighteen hours before termination, the cultures were pulsed with methyJ-3H-thymidine (3H-TdR, Amersham, Buckinghamshire, U.K.; sp . act. 74 GBqjmM), and processed as reported (23). Mitogenic response was expressed as cpm of incorporated radioactivity. Antigen response was represented by a stimulation index (S.I.), calculated by dividing 3H-TdR uptake in the presence of antigen by isotope uptake in its absence; subjects were scored as responsive when S.1. was> 2. IL-2 production and assay Lymphocyte IL-2 production was assessed as described (7) by stimulating PBL with SPA. Following 24-hour incubation, supernatants were recovered by low-speed centrifugation, filtered through 0.45 JLm filters (Millipore), and stored at - 20 "C until testing. Supernatant IL2 contents were evaluated using the CTLL cell
TPI effect on immune junction in aging
line as indicator cells (7) and expressed as cprn 3H-TdR uptake. Day-O and day-30 samples from individual subjects were tested for IL-2 content together; each experiment included a reference curve, obtained by testing double dilutions of the standard rIL-2 preparation.
NK cell function This parameter was evaluated by testing PBL ability to kill the NK-sensitive cell line K562 in a short-term SlCr release assay. Briefly, PBL wer e incubated at different effector to target cell ratios with a fi xed number of previously labelled K562 cells (100 IlCi Na2s1Cr04 (Amersh am ) for 90 minutes at 37 °C); after 4 hours at 37 DC , 100 III of the supernatant were collected and counted in a gamma-counter (packard). Results were expressed as percent specific SlCr release, ca lculated as follows: ex perimental cpm - backg round cpm total cpm - background cpm where background cpm corresponded to SlCr release in the absenc e of effectors, and tot al cpm to the radioactivity incorporated by K562 ce lls. Cell surface marker studies To study IL-2 receptor and DR ant igen ex-
pression following mitogen activation, PBL were cultured in 24-well tissue culture plates (Costar ) as described (7). Following stimulation with SPA , 24-hou r incubation and cell recovery , T cells were isolated as des cribed (24), and assayed by indirect immunofluorescence with an Epics C cytofluorometer (Beckton-Dickinson, Hialeah, FL). IL-2 receptor expression was evaluated with an mAb against IL-2 receptor (Ortho Diagnostics, Raritan, NJ); DR antigens were detected by indirect immunofluorescence with an mAb (25) against non -polymo rphic determinants of the MHC class II (anti-DR, a gift of Dr. S. Ferrone, Valhalla, NY). Phenotypic analysis of circulating PBL was achieved by single and double immunofluorescence using mAbs against cell surface antigens (Ortho) , according to standard methods; for CD45R and CDw29 detect ion, 2H4 and 484 (Coulter) mAbs, respectively, were used.
Statistical analysis Statistical analysis was performed on both crude data (cpm , percent specific cytotoxicity, et c.) and data expressed as percent var iation on day 30 compared to day-O values . The t test for paired dat a, the Mann-With ney, the Wilcox-
Table 1 - Phenotypic markers in peripheral blood lymphocytes from aged subjects before (day 0) and afte r (day 30) TPI treatment. Subjects
CD3
C D4
CD45R
CD w29
NHK-1
day 0 TP 1-treated
68.6 ± 12.6* (1,386 ± 265)**
43.7 ± 11.7 (892 ± 285)
13.1 ± 11.3 (246 ± 172)
28.6 ± 6.8 (593 ± 224)
21.6 ± 12.8 (512 ± 421)
Placebo-tre ated
70.5 ± 15.9 (1,322 ± 340)
40.5 ± 9.9 (786 ± 265)
10.1 ± 5.9 (204 ± 138)
26.5 ± 5.6 508 ± 188)
15.8 ± 8.7 (318 ± 222)
Young controls
71.1 ± 6.6 (1,721 ± 523)
39.3 ± 6.8 (928 ± 270)
18.3 ± 5.1 (448 ± 205)
18.1 ± 5.4 (419 ± 144)
15.6 ± 7.3 393 ± 286)
day 30 TP1·treated
67.6 ± 11.9 (1,518 ± 231)
42.7 ± 13.7 (949 ± 236)
14.5 ± 13.1 (291 ± 195)
26.8 ± 4.6 (619 ± 181)
21.6 ± 13.7 (568 ± 475)
Placebo-treat ed
67.5 ± 16.4 (1,574 ± 419)
39.0 ± 11.2 (905 ± 227)
10.4 ± 4.9 (239 ± 82)
27.2 ± 8. 1 (627 ± 191)
18.0 ± 7.3 (442 ± 194)
* Mean perce nt values ± SD
** Mean absolute numbers /mm" ± SD
Aging, Vol. 2, N. 4 349
U. Fagiolo, A. Amadori, F. Borghesan, et al.
on, and the exact Fisher tests were used where appropriate. RESULTS
Cell surface marker studies PBL from all study participants were examined for a panel of cell surface markers on day O. No differences-were seen for CD3, CD4, and HNK-l markers (Table 1); however, compared to young controls, aged subjects showed decreased numbers of circulating CD4SR+ lymphocytes (P < O.OS for both placebo- and TPltreated subjects), as well as significantly higher percentages of circulating CDw29+ cells (P < O.OS). Re-examination on day 30 disclosed no significant change in percentages or absolute numbers of the different subpopulations in the placebo group (Table 1). TPl-treated subjects showed no change in the percentage of CD4S+ cells, but absolute cell number was increased (P = 0.034 according to the Wilcoxon test). Remaining markers, including CDw29 antigen, did not undergo significant changes from pretreatment values (Table 1).
NK activity In keeping with previous reports (26, 27), lymphocytes from aged individuals were less able to kill KS62 cells, compared to young controls (not shown). Although no differences between pre- and post-treatment values were seen in the groups as a whole, 4 TP-l treated subjects on day 30 (Fig. 1) showed an NK activity increase exceeding the 9S% confidence limits of inter-test variability (calculated by repeatedly testing the same donor) (Fig. 1); a similar increase was not found in placebo-treated individuals (p = 0.04 according to the exact Fisher's test). Citotoxicity values observed in young controls, however, were not achieved.
/L-2 receptor/DR antigen expression, and "in vitro" /L-2 production -Lymphocytes were cultured for 24 hours in the presence of SPA; culture supernatants and cells were then recovered separately, and tested for IL-2 content and IL-2 receptor/DR antigen
350 Aging, Vol. 2, N. 4
expression, respectively. In line with previous observations (7), T cells from aged subjects on day 0 did not show an impaired ability to express early activation markers, compared to young controls; following TP1 treatment, no changes were observed (not shown). As reported (S, 7), lymphocytes from elderly subjects showed poor IL-2 production following mitogenic stimulation, compared to young controls (not shown). At the end of treatment, the TP1-treated group as a whole showed no significant change in IL-2 producing ability; on an individual level (Fig. 2), only 3 TP1-treated subjects evidenced a significant variation, compared to 2 individuals receiving placebo (not significant).
.ceo_
y
~
~
c o ~
•
~
u 0 ••••••__•• _ •• -.
o
i
100_
;
i
•
qgf
•
8
••
10 J -
.....,-
--r-
TPt ·U tot e d
D IC C 3 mm at 48 hours was observed.
352 Aging, Vol. 2, N. 4
TPl effect on imm une function in aging
cell function. Nonetheless, the mechanism(s) underlying this effect is not readily understood. NK cell ontogeny does not appear influenced by thymic hormone (34); however, C03-C02+ cells also display NK activity, and NK cells are highly sensitive to the up-regulating effect of IL2 and gamma-interferon (35). We observed no correlation between NK activity increase and IL-2 production in individual subjects, but it is possible that the TP1-associated increase in NK cell activity seen in some patients might occur through an augmented lymphokine release by T lymphocytes. In this setting, studies addressing the cytotoxic potential per cell with purified cellsubpopulations would be particularly helpful. In reference to the lymphocyte proliferative response to PHA, our observations substantiate findings in monkeys (14). The mechanisms responsible for this effect are difficult to identify, as the lymphocyte PHA response involves many sequential activation steps. A likely explanation could reside in the increase in circulating C045R+ cells following completion of TP1 treatment; it is known that this T-cell subpopulation is very sensitive to PHA stimulation (36). That IL-2 production/IL-2 receptor expression is involved in this phenomenon is improbable; in fact , no changes in these parameters were ob served followingTP1 treatment completion. On the other hand, our data do not exclude the possibility of an increased IL-2 receptor affinity (37). In this context, the finding of no parallel change in lymphocyte responsiveness to OKT3 following TP1 treatment would favour this hypothesis; in fact, studies show that lymphocyte stimulation via C03 receptor probably involves different activation pathways, apparently independent from IL-2 production and utilization (38).
In line with a previous report (16), TP1 administration was associated with a striking increase in in vivo responsiveness to PPO antigen . These findings were recently confirmed in a longitudinal study including over 190 old subjects with chronic bronchitis undergoing TP1 treatment (unpublished observations). This improvement, however, was not accompanied by a parallel increase in the in vitro lymphocyte response to the relevant antigen; this apparent paradox could reflect the observed increase in
circulating CD45R+ T cells, which are very poor responders to antigen stimulation (36). Thus, it may be advanced that thymic hormone administration induces a change in the redistribution and homing of antigen-specific T cells. However , in vivo and in vitro results are not readily compared, as in vivo responses involve a much more complex cascade of events than in vitro lymphocyte proliferation; in addition, PBL findings often may simply echo the phenotypic and functional pattern of the overall lymphoid sys tem (39). Our results indicate that thymic hormone administration in aged persons can influence some immunological parameters. The actual bearing of this effect on in vivo immune function in this age group is however a matter for conjecture. Further studies with longer TP1 treatment in individuals with overt manifestations of age -associated immune deficiency are needed to ascertain the real therapeutical potential of thymic hormones in old age.
ACKNOWLEDGEMENT We are indebted to Dr. Giovanni Forza for the invaluable help in statistical analysis.
REFERENCES 1. Weksler M.E.: Senesc ence of the immune syste m. Med . Clin. North Am. 67: 263-272,1983. 2. Bruley-Rosset M., Vergnon I.: Interleukin-1 synthesis and activity in aged mice. Mech. Ageing Dev. 24: 247264,1984. 3. Thoman M.L., Weigle W.O.: Cell-mediated immunity in aged mice: an underlying lesion in IL-2 synthesis. J. Immunol. 128: 2358-2361, 1982. 4. Thoman M.L., Weigle W.O.: Reconstitution of in vivo cell-mediated Iympholysis responses in aged mice with interleukin-2. J. Immunol. 134: 949-952, 1985. 5. GillisS., Kozak R., Durante M., Weksler M.E.: Immunological studies of aging. Decreased production of and response to T cell growth factor by lymphocytes from aged humans. J. Clin. Invest. 67: 937-942, 1981. 6. Gutowsky J.K., Innes J.8., Weksler M.E., Cohen S.: Impaired nuclear responsiveness to cytoplasmic signals in lymphocytes from elderly humans with depressed proliferative response . J. Clin. Invest. 78: 4043, 1986.
Aging, Vol. 2, N. 4 353
U. Fagiolo, A. Amadori, F. Borghesan, et al.
7. Amador i A , Zano vello P., Co zzi E., Ciminale V., Borghesan F., Fagiolo U., Crepaldi G.: Study of some ea rly immunolgical parameters in aging humans. Gerontology 34: 277-283, 1988. 8. Zat z M.M., Goldstein A.L.: Th ymosins, lymphok ines and the immunolog y of aging. Gerontology 31: 263277, 1985. 9. Weksler ME : The thymus gland and aging. A nn. Intern. Med. 98: 105-107, 1983. 10. Wek sler M.E., Innes J .B., Goldstein A L.: Immuno logical studies of aging. IV. The contribution of thymic involution to the immune deficien cies of aging mice, and reve rsa l with thymopoietin. J. Exp . Med. 148: 996-1006, 1978. II. Frasca D., Garavin i M., Dor ia G.: Recovery of T-cell functions in aged mice inject ed with synthetic thymosin alpha one . Cell. Imm uno/. 72: 384-391, 1982. 12. Ershler W.B., He bert J.e., Blow AJ ., Granter SF , Lynch J.: Effect of thyrnosin alpha one on specific antibody response and susceptib ility to infection in young and aged mice. Int. J. Immunopharmacol. 7: 465-471, 1985. 13. Frasca D., Ador ini L., Mancini C., Doria G.: Reconstitution of T-cell function s in aging mice by thymosin alpha one. Immunopharmacology 11: 155-163, 1986. 14. Ershler W.B., Co e e.L. , Laugh lin N., Klopp RG., Gravenst ein S., Roecker E.8 ., Schult z KT.: Aging and immunity in non-human primates. II. Lymphocyte response in thymosin treated middle-aged monkeys . J. Geronto/. 43: B142-146, 1988. 15. Gravenst ein S., Miller BA , Dut hir E.H., Drinka P., Prathipati K., Odiet F.M., Schicker J .M., Siewart M.E., Ershler W.B.: Enhancement of the anti-influenza antibody resp onse rate in elde rly men by thymosin alpha o ne. Clin. Res. 35: 346, 1987 (Abstract) . 16. Meroni P.L., Barcellini W., Frasca D., Sguotti G., Borghi M.a., De Bartolo G., Dor ia G. , Zanussi e. : In vivo immunopotentiating ac tivity of thymopoietin in aging humans: increase of IL-2 production. Clin. Immun ol. Imm unopathol. 42: 151-159, 1987. 17. Lin C.Y., Hsu e.H., Liu K'C; Ch en e.L., Hsu H.C: Serial immunologic and histologic studies in the treatment of necrotizing fasciitis with combined immunodeficiency by a bovine thimic ex trac t (thymost imulin). J. Pediatr. Surg . 21: 1000·1004, 1986. 18. Lin c.v., Kuo v.c., Lin e.e. , au B.R: Enhancement of interleukin-2 and gamm a-interferon production in vitro on co rd blood lymphocytes and in vivo on primary ce llular immunodeficiency pat ients with thymic extrac ts (thymostimulin). J. Clin. Immunol. 8: 103107,1988. 19. C hisesi T., Capnist G., Ranc an L., Pellizzari G., Vespignani M.: The effect of thymic substances on circulating T cells of patients treated for Hodgkin 's disease. J. Bioi. Regu/. Hom eost . Agents 2: 193-198, 1988. 20. Lighthart G.J ., Co urberaud J.X, Fourni er G., Galanaud P., Hijmans W., Kenn es B., Muller-HermeIink
354
Aging, Vol. 2, N. 4
2I.
'22.
23.
24.
25.
26.
27.
28.
29.
30.
3I.
32.
33.
H.K, St einmann G.G.: Admission criteria for immunogerontological studies in man : the SENIEUR pro tocol. Mech . Ageing Dev . 28: 47-55, 1984. Falchetti R , Bergesi A, Eshkol A , Caliero e., Adorini L., Caprino L.: Pharmacological and biological properties of a calf thymus extr act (TP 1). Drugs Exp. Clin. Res. 3: 39-47, 1977. Amadori A , De Rossi A, Faulkner-Valle G.P., ChiecoBianchi L.: Spontaneous in vitro production of virusspecific antibody by lymphocytes from HIV-infected subjects. Clin. Immuno/. Immunopatho/. 46: 342-351, 1988. Amadori A , Faulkner-Valle G.P., De Rossi A., Zano vella P., Collavo D., Chieco-Bianch i L.: HIV-mediated immunodepression: in vitro inhibition of T-Iymphocyte pro liferative response by UV-irradiated virus . Clin. Immunol. Immunopathol. 46: 37·54, 1988. Romagnani S., Maggi E., Amadori A , Giudizi M.G., Ricci M.: Cooperation bet ween T and B lymphocytes from human tonsil in the response to mitogens and antigens. Clin. Exp. Im muno/. 28: 332-338, 1977. Qu arant a V., Tanigaki N., Ferron e S .: Distribution of antigenic determinants recognized by thr ee monocl onal antibodies (Q 2/70, Q 5/ 6 and Q 5/ 13) on human Ia-like alloantigens and on their subunits . Immunogenetics 12: 175-182, 198I. Pross HF, Baines M.G.: St udies of human natural killer cells. I. In vivo parameters affecting norm al cytotox ic function. Int. J. Cancer 29: 383-390, 1982. Mysliwska J ., Mysliwsk i A , Witkowski J .: Age-depen dent decline of natural killer and 'antibody-dependent cell mediated cytotoxicity activity of hum an lymphocytes is connecte d with decrease of their acid phosp hatase activity. Mech . Ageing Dev. 31: 1-11, 1985: Wek sler ME , Hutt eroth T.H.: Impaired lymphocyte function in aged humans . J. Clin. Invest. 53: 99-104, 1974. Hefton J.M ., Darlington G.J ., Casaz za B.A, Weksler ME : Immuno logic studies of aging. V. Impaired proliferation of PHA responsive human hymphocytes in culture. J. Immuno/. 125: 1007·1010, 1980. De Paoli B., Battistin S., Santini G.F.: Age-related changes in human lymphocyt e subsets: progressive reduction of the CD4 CD4 5R (suppressor inducer) population. Clin. Immunol. Immu nop athol. 48: 290296,1988. Akbar AN. , Terry L., Timms A , Beverley Pol ., Janossy G .: Loss of CD45R and gain of UCHLl reac tivity is a feature of pr imed T ce lls. J. Immuno/. 140: 2171·2178,1988. Pilar ski L.M., Deans J .P. : Selective expression of CD45 isoforms and of maturation antigens during hum an thymocyte differentiation : observation and hypothesis .lmmuno/. Lett . 21: 187-198, 1989. Sobel RA , Hafler D.A, Castro E.E., Morimoto e.,
TPl effect on immune function in aging
Wein er H.L.: Th e 2H4 (C D45R) antige n is se lec tively decreased in multiple scle rosis lesions. J. Immunol. 140: 2210·22 14, 1988. 34. Cant or H., Kasai M., Sh en F.W. , Leclerc J .e. , Glimch er L.: Immunogen etic analysis of "natural killer " activity in the mou se. Immunol. Rev. 44: 3-12, 1979. 35. Brooks e.G., Urdal D.L. , Hen ney e.S.: Lyrnphokinedriven differentiation of cyto toxic T-cell clones into ce lls with NK-like specificity: co rre lations with display of mac rom olecules . lmmunol. Reu. 72: 43-72, 1983. 36. Morim oto G., Letv in N.L. , Dist aso J.A., Aldrich W.R., Schlossman S.F. : Th e isolation a nd cha rac te riza tion
of the human suppressor inducer T cell s ubse t. J. Immunol. 134: 1508-1515, 1985. 37. Shimi zu A., Kondo S., Sa be H. , Ishid a N., Honjo T. : Structure and function of the interleukin-2 receptor: affi nity co nve rsion mod els. Immun ol. Rev. 92: 103120, 1986. 38. Laing T.J. , Weiss A.: Eviden ce for IL-2 inde pende nt pr oliferation in human T ce lls. J. lmmunol. 140: 10561062, 1988. 39. Ern st D.N. , Weigle W.O., Th oman M.L.: Retention of T cell react ivity to mitogen s a nd alloantigen s by Peyer pat ch ce lls of aged mice. J. lmmunol. 138: 26-31, 1987.
Aging, Vol. 2, N. 4 355
ORIGINAL ARTICLES
Immune dysfunction in the elderly: Effect of thymic hormone administration on several in vivo and in vitro immune function parameters U. Fagiolo*, A. Amadori**, F. Borghesan*, R. Zamarchi**, M.L. Veronese**, G. De Silvestro***,
E. Passarella****, and G. Crepaldi*
* Institute of Internal Medicine and ** Institute of Oncology, University of Padova, Padova, *** Central Laboratory Unit of Padova General Hospital, Padova, **** Serono Pharmaceutical,
Milano, Italy.
ABSTRACT. The effects of short-term thymic hormone administration on age-associated immune function were evaluated. Two groups of individuals > 65 years of age were treated for 30 days with thymic extracts (TP1) or placebo; before and after this treatment a panel of in vitro and in vivo parameters was determined according to a very rigorous experimental protocol. In most individuals, TP1 treatment was associated with an improvement in cutaneous delayed-type response to PPD. Moreover, an increase in a circulating T cell subpopulation bearing the CD45R surface antigen ("virgin" T cells), and in NK cell cytotoxic activity was also observed in some subjects. Finally, lymphocyte responsiveness to PHA tended to increase, while no effect on lymphocyte ability to produce IL-2 following mitogen stimulation was observed. These findings suggest that TP1 treatment may influence age-related alterations in immune function parameters in some subjects. (Aging 2: 347-355, 1990) INTRODUCTION Immunodepression is associated with aging
in both animals and humans (1). Although different mechanisms probably underlie immune dysfunction in mice (2-4) and man (5-7), progressive involution of thymic function is a common feature (8, 9). Attempts to reconstitute immune function by in vivo thymic hormone administration were successful in mice (10-13), and associated with an increase in some immune function parameters in middle-aged subhuman primates (14). Studies in elderly persons are more sporadic (15, 16), but some findings indicate that such treatment may have beneficial effects in immunodeficiency patients (17-19). Longitudinal investigations into the effects of drugs on immune function are hampered in man by the high inter-person variability of the immune response, as well as the inter-test variability of in vitro assays, and time-dependent variations in immune function in individual subjects. The first two problems can be circumvented by observing a very strict experimental protocol; the last is not readily resolved, but can be absorbed by parallel studies in a matched group of untreated subjects. Under these conditions, we investigated whether short-term
Key words: Aging, thymic hormones, TP1, age-associated immune defect, immunotherapy. Correspondence: Umberto Fagiolo, M.D., Istituto di Medicina Interna, Via Giustiniani 2, 35128 Padova, Italy. Received December 18, 1989; accepted July 9, 1990.
Aging, Vol. 2, N. 4 347
U. Fagiolo, A. Amadori, F. Borghesan, et al.
thymic hormone treatment in elderly persons was associated with a change in some in vivo and in vitro immune function parameters, namely, delayed type response to PPD, cell surface marker expression, NK activity, early activation marker expression, IL-2 production and proliferative response to mitogens and antigens. METHODS
Subjects and experimental protocol All persons enrolled in this study fulfilled the EURAGE Senieur protocol criteria (20). Thirteen individuals> 65 years of age (6 males and 7 females; mean age 77 years, range 67-90) were injected i.m. with 70 mg of thymic hormone extract (TP1, kindly provided by Serono, Milano, Italy, and prepared as described in 21) daily for 7 days, and then twice a week for 3 weeks; before (day 0) and at the end of treatment (day 30) several immune function parameters were determined. A control group of 11 age- and sexmatched individuals was injected with placebo ' according to the same schedule, and analyzed , as above. Due to the low number of lymphocytes recovered in some cases, not all the assays were performed in every individual. A very strict experimental protocol was followed. Inter-test variability was circumscribed by testing equal numbers of TP1 -treated subjects and controls in a single session; moreover, the same persons forming a given group at evaluation on day 0 were re-examined all together on day 30. To assess inter-test variability, single assays were repeatedly performed in two normal young volunteers, and the standard deviation from the mean of replicate tests was calculat ed; variations observed on dav 30 in the experimental groups were considered significant only when they exceeded the 95% confidence limits of this variability. An additional control group consisted of 14 young members of the laboratory staff (mean age 27 years). Reagents Recombinant IL-2 (rIL-2) was obtained through the courtesy of Dr. F. Sinigaglia, Hoffmann-LaRoche, Basel, Switzerland. Phytohemagglutinin (pHA-P, Difco, Detroit, MI) was
348 Aging, Vol. 2, N. 4
used at 0.5 JLg/ml. Monoclonal antibody (mAb) OKT3 , kindly provided by Dr. Dunia Ramarli, Genova, Italy, was used at 10 ng/ml. Protein purified derivative (PPD), a gift of Sclavo, Siena , Italy, was used at 200 ng/ml. Staphylococcal protein A (SPA, Pharmacia, Uppsala, Sweden) was used at 100 JLg/ml. In vivo testing for cutaneous delayed-type response to recall antigens was performed using Multitest Merieux.
Lymphocyte isolation and culture Peripheral blood lymphocytes (PBL) were isolated from heparinized fasting blood samples collected between 8-9 a.m ., as previously described (22). Cells were resuspended in RPMI 1640 medium containing 10% fetal calf serum (FCS, Flow, Irvine, U.K.), 1% L-glutamine, 1% non-essential aminoacids, 2 x 10-sM 2-mercaptoethanol, and 50 JLg/ml gentamycin (complete RPMI). Mitogenic stimulation was accomplished by culturing PBL in 96-well U-bottom microtiter plates (Costar Data Packaging, Cambridge, MA) at a concentration of 1 x 106 cells/rnl in 0.2 ml complete RPMI in the presence of PHA or OKT3 for 72 hours at 37 -c, 5% CO2 , For antigenic stimulation, PBL were cultured for 144 hours in complete RPMI containing 10% human pooled serum in place of FCS in the presence of PPD or irradiated (3,000 R) pooled PBL from normal donors as allogeneic stimulators. Eighteen hours before termination, the cultures were pulsed with methyJ-3H-thymidine (3H-TdR, Amersham, Buckinghamshire, U.K.; sp . act. 74 GBqjmM), and processed as reported (23). Mitogenic response was expressed as cpm of incorporated radioactivity. Antigen response was represented by a stimulation index (S.I.), calculated by dividing 3H-TdR uptake in the presence of antigen by isotope uptake in its absence; subjects were scored as responsive when S.1. was> 2. IL-2 production and assay Lymphocyte IL-2 production was assessed as described (7) by stimulating PBL with SPA. Following 24-hour incubation, supernatants were recovered by low-speed centrifugation, filtered through 0.45 JLm filters (Millipore), and stored at - 20 "C until testing. Supernatant IL2 contents were evaluated using the CTLL cell
TPI effect on immune junction in aging
line as indicator cells (7) and expressed as cprn 3H-TdR uptake. Day-O and day-30 samples from individual subjects were tested for IL-2 content together; each experiment included a reference curve, obtained by testing double dilutions of the standard rIL-2 preparation.
NK cell function This parameter was evaluated by testing PBL ability to kill the NK-sensitive cell line K562 in a short-term SlCr release assay. Briefly, PBL wer e incubated at different effector to target cell ratios with a fi xed number of previously labelled K562 cells (100 IlCi Na2s1Cr04 (Amersh am ) for 90 minutes at 37 °C); after 4 hours at 37 DC , 100 III of the supernatant were collected and counted in a gamma-counter (packard). Results were expressed as percent specific SlCr release, ca lculated as follows: ex perimental cpm - backg round cpm total cpm - background cpm where background cpm corresponded to SlCr release in the absenc e of effectors, and tot al cpm to the radioactivity incorporated by K562 ce lls. Cell surface marker studies To study IL-2 receptor and DR ant igen ex-
pression following mitogen activation, PBL were cultured in 24-well tissue culture plates (Costar ) as described (7). Following stimulation with SPA , 24-hou r incubation and cell recovery , T cells were isolated as des cribed (24), and assayed by indirect immunofluorescence with an Epics C cytofluorometer (Beckton-Dickinson, Hialeah, FL). IL-2 receptor expression was evaluated with an mAb against IL-2 receptor (Ortho Diagnostics, Raritan, NJ); DR antigens were detected by indirect immunofluorescence with an mAb (25) against non -polymo rphic determinants of the MHC class II (anti-DR, a gift of Dr. S. Ferrone, Valhalla, NY). Phenotypic analysis of circulating PBL was achieved by single and double immunofluorescence using mAbs against cell surface antigens (Ortho) , according to standard methods; for CD45R and CDw29 detect ion, 2H4 and 484 (Coulter) mAbs, respectively, were used.
Statistical analysis Statistical analysis was performed on both crude data (cpm , percent specific cytotoxicity, et c.) and data expressed as percent var iation on day 30 compared to day-O values . The t test for paired dat a, the Mann-With ney, the Wilcox-
Table 1 - Phenotypic markers in peripheral blood lymphocytes from aged subjects before (day 0) and afte r (day 30) TPI treatment. Subjects
CD3
C D4
CD45R
CD w29
NHK-1
day 0 TP 1-treated
68.6 ± 12.6* (1,386 ± 265)**
43.7 ± 11.7 (892 ± 285)
13.1 ± 11.3 (246 ± 172)
28.6 ± 6.8 (593 ± 224)
21.6 ± 12.8 (512 ± 421)
Placebo-tre ated
70.5 ± 15.9 (1,322 ± 340)
40.5 ± 9.9 (786 ± 265)
10.1 ± 5.9 (204 ± 138)
26.5 ± 5.6 508 ± 188)
15.8 ± 8.7 (318 ± 222)
Young controls
71.1 ± 6.6 (1,721 ± 523)
39.3 ± 6.8 (928 ± 270)
18.3 ± 5.1 (448 ± 205)
18.1 ± 5.4 (419 ± 144)
15.6 ± 7.3 393 ± 286)
day 30 TP1·treated
67.6 ± 11.9 (1,518 ± 231)
42.7 ± 13.7 (949 ± 236)
14.5 ± 13.1 (291 ± 195)
26.8 ± 4.6 (619 ± 181)
21.6 ± 13.7 (568 ± 475)
Placebo-treat ed
67.5 ± 16.4 (1,574 ± 419)
39.0 ± 11.2 (905 ± 227)
10.4 ± 4.9 (239 ± 82)
27.2 ± 8. 1 (627 ± 191)
18.0 ± 7.3 (442 ± 194)
* Mean perce nt values ± SD
** Mean absolute numbers /mm" ± SD
Aging, Vol. 2, N. 4 349
U. Fagiolo, A. Amadori, F. Borghesan, et al.
on, and the exact Fisher tests were used where appropriate. RESULTS
Cell surface marker studies PBL from all study participants were examined for a panel of cell surface markers on day O. No differences-were seen for CD3, CD4, and HNK-l markers (Table 1); however, compared to young controls, aged subjects showed decreased numbers of circulating CD4SR+ lymphocytes (P < O.OS for both placebo- and TPltreated subjects), as well as significantly higher percentages of circulating CDw29+ cells (P < O.OS). Re-examination on day 30 disclosed no significant change in percentages or absolute numbers of the different subpopulations in the placebo group (Table 1). TPl-treated subjects showed no change in the percentage of CD4S+ cells, but absolute cell number was increased (P = 0.034 according to the Wilcoxon test). Remaining markers, including CDw29 antigen, did not undergo significant changes from pretreatment values (Table 1).
NK activity In keeping with previous reports (26, 27), lymphocytes from aged individuals were less able to kill KS62 cells, compared to young controls (not shown). Although no differences between pre- and post-treatment values were seen in the groups as a whole, 4 TP-l treated subjects on day 30 (Fig. 1) showed an NK activity increase exceeding the 9S% confidence limits of inter-test variability (calculated by repeatedly testing the same donor) (Fig. 1); a similar increase was not found in placebo-treated individuals (p = 0.04 according to the exact Fisher's test). Citotoxicity values observed in young controls, however, were not achieved.
/L-2 receptor/DR antigen expression, and "in vitro" /L-2 production -Lymphocytes were cultured for 24 hours in the presence of SPA; culture supernatants and cells were then recovered separately, and tested for IL-2 content and IL-2 receptor/DR antigen
350 Aging, Vol. 2, N. 4
expression, respectively. In line with previous observations (7), T cells from aged subjects on day 0 did not show an impaired ability to express early activation markers, compared to young controls; following TP1 treatment, no changes were observed (not shown). As reported (S, 7), lymphocytes from elderly subjects showed poor IL-2 production following mitogenic stimulation, compared to young controls (not shown). At the end of treatment, the TP1-treated group as a whole showed no significant change in IL-2 producing ability; on an individual level (Fig. 2), only 3 TP1-treated subjects evidenced a significant variation, compared to 2 individuals receiving placebo (not significant).
.ceo_
y
~
~
c o ~
•
~
u 0 ••••••__•• _ •• -.
o
i
100_
;
i
•
qgf
•
8
••
10 J -
.....,-
--r-
TPt ·U tot e d
D IC C 3 mm at 48 hours was observed.
352 Aging, Vol. 2, N. 4
TPl effect on imm une function in aging
cell function. Nonetheless, the mechanism(s) underlying this effect is not readily understood. NK cell ontogeny does not appear influenced by thymic hormone (34); however, C03-C02+ cells also display NK activity, and NK cells are highly sensitive to the up-regulating effect of IL2 and gamma-interferon (35). We observed no correlation between NK activity increase and IL-2 production in individual subjects, but it is possible that the TP1-associated increase in NK cell activity seen in some patients might occur through an augmented lymphokine release by T lymphocytes. In this setting, studies addressing the cytotoxic potential per cell with purified cellsubpopulations would be particularly helpful. In reference to the lymphocyte proliferative response to PHA, our observations substantiate findings in monkeys (14). The mechanisms responsible for this effect are difficult to identify, as the lymphocyte PHA response involves many sequential activation steps. A likely explanation could reside in the increase in circulating C045R+ cells following completion of TP1 treatment; it is known that this T-cell subpopulation is very sensitive to PHA stimulation (36). That IL-2 production/IL-2 receptor expression is involved in this phenomenon is improbable; in fact , no changes in these parameters were ob served followingTP1 treatment completion. On the other hand, our data do not exclude the possibility of an increased IL-2 receptor affinity (37). In this context, the finding of no parallel change in lymphocyte responsiveness to OKT3 following TP1 treatment would favour this hypothesis; in fact, studies show that lymphocyte stimulation via C03 receptor probably involves different activation pathways, apparently independent from IL-2 production and utilization (38).
In line with a previous report (16), TP1 administration was associated with a striking increase in in vivo responsiveness to PPO antigen . These findings were recently confirmed in a longitudinal study including over 190 old subjects with chronic bronchitis undergoing TP1 treatment (unpublished observations). This improvement, however, was not accompanied by a parallel increase in the in vitro lymphocyte response to the relevant antigen; this apparent paradox could reflect the observed increase in
circulating CD45R+ T cells, which are very poor responders to antigen stimulation (36). Thus, it may be advanced that thymic hormone administration induces a change in the redistribution and homing of antigen-specific T cells. However , in vivo and in vitro results are not readily compared, as in vivo responses involve a much more complex cascade of events than in vitro lymphocyte proliferation; in addition, PBL findings often may simply echo the phenotypic and functional pattern of the overall lymphoid sys tem (39). Our results indicate that thymic hormone administration in aged persons can influence some immunological parameters. The actual bearing of this effect on in vivo immune function in this age group is however a matter for conjecture. Further studies with longer TP1 treatment in individuals with overt manifestations of age -associated immune deficiency are needed to ascertain the real therapeutical potential of thymic hormones in old age.
ACKNOWLEDGEMENT We are indebted to Dr. Giovanni Forza for the invaluable help in statistical analysis.
REFERENCES 1. Weksler M.E.: Senesc ence of the immune syste m. Med . Clin. North Am. 67: 263-272,1983. 2. Bruley-Rosset M., Vergnon I.: Interleukin-1 synthesis and activity in aged mice. Mech. Ageing Dev. 24: 247264,1984. 3. Thoman M.L., Weigle W.O.: Cell-mediated immunity in aged mice: an underlying lesion in IL-2 synthesis. J. Immunol. 128: 2358-2361, 1982. 4. Thoman M.L., Weigle W.O.: Reconstitution of in vivo cell-mediated Iympholysis responses in aged mice with interleukin-2. J. Immunol. 134: 949-952, 1985. 5. GillisS., Kozak R., Durante M., Weksler M.E.: Immunological studies of aging. Decreased production of and response to T cell growth factor by lymphocytes from aged humans. J. Clin. Invest. 67: 937-942, 1981. 6. Gutowsky J.K., Innes J.8., Weksler M.E., Cohen S.: Impaired nuclear responsiveness to cytoplasmic signals in lymphocytes from elderly humans with depressed proliferative response . J. Clin. Invest. 78: 4043, 1986.
Aging, Vol. 2, N. 4 353
U. Fagiolo, A. Amadori, F. Borghesan, et al.
7. Amador i A , Zano vello P., Co zzi E., Ciminale V., Borghesan F., Fagiolo U., Crepaldi G.: Study of some ea rly immunolgical parameters in aging humans. Gerontology 34: 277-283, 1988. 8. Zat z M.M., Goldstein A.L.: Th ymosins, lymphok ines and the immunolog y of aging. Gerontology 31: 263277, 1985. 9. Weksler ME : The thymus gland and aging. A nn. Intern. Med. 98: 105-107, 1983. 10. Wek sler M.E., Innes J .B., Goldstein A L.: Immuno logical studies of aging. IV. The contribution of thymic involution to the immune deficien cies of aging mice, and reve rsa l with thymopoietin. J. Exp . Med. 148: 996-1006, 1978. II. Frasca D., Garavin i M., Dor ia G.: Recovery of T-cell functions in aged mice inject ed with synthetic thymosin alpha one . Cell. Imm uno/. 72: 384-391, 1982. 12. Ershler W.B., He bert J.e., Blow AJ ., Granter SF , Lynch J.: Effect of thyrnosin alpha one on specific antibody response and susceptib ility to infection in young and aged mice. Int. J. Immunopharmacol. 7: 465-471, 1985. 13. Frasca D., Ador ini L., Mancini C., Doria G.: Reconstitution of T-cell function s in aging mice by thymosin alpha one. Immunopharmacology 11: 155-163, 1986. 14. Ershler W.B., Co e e.L. , Laugh lin N., Klopp RG., Gravenst ein S., Roecker E.8 ., Schult z KT.: Aging and immunity in non-human primates. II. Lymphocyte response in thymosin treated middle-aged monkeys . J. Geronto/. 43: B142-146, 1988. 15. Gravenst ein S., Miller BA , Dut hir E.H., Drinka P., Prathipati K., Odiet F.M., Schicker J .M., Siewart M.E., Ershler W.B.: Enhancement of the anti-influenza antibody resp onse rate in elde rly men by thymosin alpha o ne. Clin. Res. 35: 346, 1987 (Abstract) . 16. Meroni P.L., Barcellini W., Frasca D., Sguotti G., Borghi M.a., De Bartolo G., Dor ia G. , Zanussi e. : In vivo immunopotentiating ac tivity of thymopoietin in aging humans: increase of IL-2 production. Clin. Immun ol. Imm unopathol. 42: 151-159, 1987. 17. Lin C.Y., Hsu e.H., Liu K'C; Ch en e.L., Hsu H.C: Serial immunologic and histologic studies in the treatment of necrotizing fasciitis with combined immunodeficiency by a bovine thimic ex trac t (thymost imulin). J. Pediatr. Surg . 21: 1000·1004, 1986. 18. Lin c.v., Kuo v.c., Lin e.e. , au B.R: Enhancement of interleukin-2 and gamm a-interferon production in vitro on co rd blood lymphocytes and in vivo on primary ce llular immunodeficiency pat ients with thymic extrac ts (thymostimulin). J. Clin. Immunol. 8: 103107,1988. 19. C hisesi T., Capnist G., Ranc an L., Pellizzari G., Vespignani M.: The effect of thymic substances on circulating T cells of patients treated for Hodgkin 's disease. J. Bioi. Regu/. Hom eost . Agents 2: 193-198, 1988. 20. Lighthart G.J ., Co urberaud J.X, Fourni er G., Galanaud P., Hijmans W., Kenn es B., Muller-HermeIink
354
Aging, Vol. 2, N. 4
2I.
'22.
23.
24.
25.
26.
27.
28.
29.
30.
3I.
32.
33.
H.K, St einmann G.G.: Admission criteria for immunogerontological studies in man : the SENIEUR pro tocol. Mech . Ageing Dev . 28: 47-55, 1984. Falchetti R , Bergesi A, Eshkol A , Caliero e., Adorini L., Caprino L.: Pharmacological and biological properties of a calf thymus extr act (TP 1). Drugs Exp. Clin. Res. 3: 39-47, 1977. Amadori A , De Rossi A, Faulkner-Valle G.P., ChiecoBianchi L.: Spontaneous in vitro production of virusspecific antibody by lymphocytes from HIV-infected subjects. Clin. Immuno/. Immunopatho/. 46: 342-351, 1988. Amadori A , Faulkner-Valle G.P., De Rossi A., Zano vella P., Collavo D., Chieco-Bianch i L.: HIV-mediated immunodepression: in vitro inhibition of T-Iymphocyte pro liferative response by UV-irradiated virus . Clin. Immunol. Immunopathol. 46: 37·54, 1988. Romagnani S., Maggi E., Amadori A , Giudizi M.G., Ricci M.: Cooperation bet ween T and B lymphocytes from human tonsil in the response to mitogens and antigens. Clin. Exp. Im muno/. 28: 332-338, 1977. Qu arant a V., Tanigaki N., Ferron e S .: Distribution of antigenic determinants recognized by thr ee monocl onal antibodies (Q 2/70, Q 5/ 6 and Q 5/ 13) on human Ia-like alloantigens and on their subunits . Immunogenetics 12: 175-182, 198I. Pross HF, Baines M.G.: St udies of human natural killer cells. I. In vivo parameters affecting norm al cytotox ic function. Int. J. Cancer 29: 383-390, 1982. Mysliwska J ., Mysliwsk i A , Witkowski J .: Age-depen dent decline of natural killer and 'antibody-dependent cell mediated cytotoxicity activity of hum an lymphocytes is connecte d with decrease of their acid phosp hatase activity. Mech . Ageing Dev. 31: 1-11, 1985: Wek sler ME , Hutt eroth T.H.: Impaired lymphocyte function in aged humans . J. Clin. Invest. 53: 99-104, 1974. Hefton J.M ., Darlington G.J ., Casaz za B.A, Weksler ME : Immuno logic studies of aging. V. Impaired proliferation of PHA responsive human hymphocytes in culture. J. Immuno/. 125: 1007·1010, 1980. De Paoli B., Battistin S., Santini G.F.: Age-related changes in human lymphocyt e subsets: progressive reduction of the CD4 CD4 5R (suppressor inducer) population. Clin. Immunol. Immu nop athol. 48: 290296,1988. Akbar AN. , Terry L., Timms A , Beverley Pol ., Janossy G .: Loss of CD45R and gain of UCHLl reac tivity is a feature of pr imed T ce lls. J. Immuno/. 140: 2171·2178,1988. Pilar ski L.M., Deans J .P. : Selective expression of CD45 isoforms and of maturation antigens during hum an thymocyte differentiation : observation and hypothesis .lmmuno/. Lett . 21: 187-198, 1989. Sobel RA , Hafler D.A, Castro E.E., Morimoto e.,
TPl effect on immune function in aging
Wein er H.L.: Th e 2H4 (C D45R) antige n is se lec tively decreased in multiple scle rosis lesions. J. Immunol. 140: 2210·22 14, 1988. 34. Cant or H., Kasai M., Sh en F.W. , Leclerc J .e. , Glimch er L.: Immunogen etic analysis of "natural killer " activity in the mou se. Immunol. Rev. 44: 3-12, 1979. 35. Brooks e.G., Urdal D.L. , Hen ney e.S.: Lyrnphokinedriven differentiation of cyto toxic T-cell clones into ce lls with NK-like specificity: co rre lations with display of mac rom olecules . lmmunol. Reu. 72: 43-72, 1983. 36. Morim oto G., Letv in N.L. , Dist aso J.A., Aldrich W.R., Schlossman S.F. : Th e isolation a nd cha rac te riza tion
of the human suppressor inducer T cell s ubse t. J. Immunol. 134: 1508-1515, 1985. 37. Shimi zu A., Kondo S., Sa be H. , Ishid a N., Honjo T. : Structure and function of the interleukin-2 receptor: affi nity co nve rsion mod els. Immun ol. Rev. 92: 103120, 1986. 38. Laing T.J. , Weiss A.: Eviden ce for IL-2 inde pende nt pr oliferation in human T ce lls. J. lmmunol. 140: 10561062, 1988. 39. Ern st D.N. , Weigle W.O., Th oman M.L.: Retention of T cell react ivity to mitogen s a nd alloantigen s by Peyer pat ch ce lls of aged mice. J. lmmunol. 138: 26-31, 1987.
Aging, Vol. 2, N. 4 355
