Aging, maturation and lymphokine production

Eur. J. Immunol. 1991. 21: 273-281

Lex NagelkerkenO, Anita Hertogh-HuijbregtsO, Ruud DobberO and Angelika DragernA Department of Immunology, Institute for Experimental Gerontology TNOO, Rijswijk and Holland Biotechnology BVO, Leiden

Age-related changes in lymphokine production related to a decreased number of CD45RBhi CD4+ T cells The ability of CD4+ T cells from CBA/Rij mice to produce interleukin (IL) 2 after stimulation with anti-CD3, concanavalin A, or the combination of phorbol 12-myristate 13-acetate and ionomycin declines during aging. This phenomenon was accompanied by an increased production of IL 4 and interferon-y. These age-related changes in lymphokine production correlated with the decrease in the percentage of CD45RBhiCD4+ Tcells from about 80% in 2-month-old to about 40% in 27-month-old mice. This phenotypic shift was responsible for the decline in IL 2 production, because in young and in old mice CD45RBhiCD4+ T cells were more potent IL 2 producers than CD45RBI0cells. Moreover, old CD45RBhi CD4+ T cells produced less IL 2 than their young counterparts. Proliferative responses by Tcells from old mice were lower than those of young mice, regardless whether the cultures were supplemented with IL2, I L 4 or both lymphokines. As far as CD4+ T cells were concerned, this hyporesponsiveness was found in the CD45RB'" as well as in the CD45RBhi CD4+ T cell population.

1 Introduction Aging is accompanied by a decline in the production of IL 2 in a variety of species [ 1-31, and this phenomenon has so far been regarded as a defect responsible for a decline in various immune functions during aging. With regard to antigen-specific responses this decline in IL 2 production is partly caused by a lower frequency of antigen-specific CD4+ Tcells [4, 51. As far as the low responsiveness to mitogenic stimuli is concerned, there is evidence that this may be due to an inadequate signal transduction [6-91. In these studies it has, however, not been taken into account that naive and memory cells may differ in their activation requirements and mode of response. As postulated by Mosmann and Coffmann [lo], murine CD4+ T cell clones can be divided on the basis of their function and lymphokine profiles into T H cells ~ and T H cells. In this concept, the production of IL 2, lymphotoxin and IFN-y is restricted to T H cells, ~ whereas T Hcells ~ are characterized by the production of IL 4, IL 5 and IL 6 (for review see [lo]). It may, therefore, be that several observations made in aged mice are related to a changed composition of the CD4+ Tcell compartment. In addition to a different profile of cytokines produced by T Hand ~ T H

[I 86311 A

273

Present address: Department of Haematology, Free University Hospital, PO. Box 7057, NL-1007 MB Amsterdam,TheNetherlands

Correspondence: Lex Nagelkerken, Department of Immunology, Institute for Experimental Gerontology TNO, PO. Box 5815, NL-2280 HV Rijswijk, The Netherlands Abbreviation: SCM: Standard culture medium 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991

cells, there is growing evidence that in mouse and rat CD4+ T cells can be divided into two subsets on the basis of CD45RB expression.This has been demonstrated in the rat with antibody OX22 [ l l , 121, and in the mouse with antibodies 16A [13] and 2362 [14-161. These antibodies recognize CD45 isoforms which depend on the expression of the B exon of the CD45 gene. Evidence has been presented that the CD45RBhigh(CD45RBhi) cells correspond to naive, immature cells, whereas the CD45RB'O" (CD45RBI0) cells represent memory cells ([ 161 and reviewed in [17]). Although in rat and mouse CD45RBhi CD4+ T cells from the periphery displayTH1characteristics by the production of I L 2 , and CD45RB'" CD4+ Tcells display T H characteristics ~ by IL 4 secretion, the situation remains complex since T Hclones ~ tend to express CD45RB in high density in contrast toTHl clones [13,14].Therefore, probably, CD45FU3 expression may reflect the state of ~activation of a cell, rather than a separate lineage of cells. In humans, the naive CD4+ Tcells are characterized by the CD45RA isoform, which is dependent on the expression of the A exon and which is recognized by antibody 2H4, whereas the reciprocal CD45RO+ CD4+ T cells recognized ~by UCHLl represent the memory CD4+ Tcells (reviewed in [18]). By contrast, no antibody has been identified so far that selectively stains CD45RB- or CD45RBIO CD4+ T cells in the mouse, although Pgp-1 (the murine equivalent of CD44, which is regarded as a marker for memory cells [19,20]) may be an interesting candidate.Therefore, it is of interest that Lerner et al. [21] demonstrated that aging of the immune system is accompanied by an increase in the percentage of Pgp-l+ cells. In view of these findings we wanted to investigate further whether a relation exists between an age-related maturation of CD4+ T cells as measured by the expression of CD45RB and their functional capacities. Our results indicate that maturation of CD4+ Tcells which is accompanied by a shift in the functional capacity of the cells is a major cause of the age-related decrease in IL 2 production. 0014-2980/91/0202-0273$3.50+ .25/0

274

L. Nagelkerken, A. Hertogh-Huijbregts, R. Dobber and A. Drager

Eur. J. Immunol. 1991. 21: 273-281

2 Methods

2.3 Cell cultures

2.1 Animals

CD4+ T cells from individual mice were cultured at a density of 5 x 10'Vwell in flat-bottom microtiter plates (Greiner, Nurtingen, FRG) in standard culture medium (SCM). SCM consisted of RPMI 1640 (Gibco, Paisley, Scotland) supplemented with penicillin (100 IU/ml), streptomycin (100 pg/ml), L-glutamine (2 gA), 2-ME (5 x M), Hepes (20 mM) and 5% FCS (Seralab, Crawley Down, GB). Cells were stimulated with optimal concentrations of mitogen [2 pg/ml Con A (Pharmacia) or 2 pg/ml anti-CD3 E (hybridoma 145-2C11 [24], a kind gift Dr. J. Bluestone)] always in the presence of 105 2500 radirradiated syngeneic young spleen cells, which served as a source of accessory cells. Alternatively, CD4+ T cells were stimulated with the combination of PMA (2 ng/ml; Sigma, St. Louis, MO) and ionomycin (300 nM, Calbiochem, Lalolla, CA), without further addition of accessory cells. SN were collected after several culture periods and stored frozen at - 20 "C until assay.

Male CBA/Rij mice were bred and maintained under specific pathogen-free (spf) conditions until an age of 10 weeks.Young mice were derived from this stock. Older mice were derived from our aging colonies of mice. These mice were derived from the spf stock and housed under clean conventional conditions behind a physical barrier from an age of 3 months. Mice that showed any signs of pathology were excluded from this study. 2.2 Isolation of CD4+ T cells After lysis of the erythrocytes, spleen cells from individual mice were centrifuged (20 min, 600 x g, 4°C) over discontinuous (e = 1.047/1.100 kg/l) Percoll (Pharmacia, Uppsala, Sweden) gradients. Cells from the interphase were washed and depleted of the majority of B cells and MQ, by one passage over nylon wool. CD4+ Tcells were further purified by a "panning" procedure. For this, nonadherent cells were simultaneously incubated with anti-Ly-2 antibody 55.6.7 [22], anti-Mac-la (American Type Culture Collection, ATCC, Rockville, MD; TIB 128) and rabbitanti-asialo-GM1 (WAKO Chemicals, Neuss, FRG), washed, and plated on petri dishes coated (16 h, 4 "C) with 10 ml 0.1 M NaHC03, pH 9.5 containing 25 pg/ml goat anti-rat Ig and 25 pg/ml swine anti-rabbit Ig to remove CD8+ Tcells, and to deplete M@ and NK cells further. Nonadherent cells were collected and examined for purity by FCM analysis, using FITC-labeled GK1.5 [23]. The enriched cell population from 2-month-old mice contained 95% CD4+ Tcells and that from 15-month or older mice contained 92% CD4+ T cells. In the CDCenriched fractions of young mice contaminating cells were equally distributed over B cells, MQ, and NK cells (1.5% each), whereas no CD8+ Tcells were detected. In aged mice the CD4+ Tcell populations contained an increased percentage of cells that expressed Mac-lu in low density. In later experiments. where CD4+ T cell subsets were needed,we shortened the isolation procedure by incubating spleen cells ( 108/ml; 15 min, room temperature) with 25% Magnisort-anti-mouse suspension, according to the instructions of the manufacturer (DuPont,Wilmington, DE).The adherent fraction was removed with a Dynal MPC-6 Magnetic Particle concentrator (Dynal AS, Oslo, Norway). The nonadherent cells were then depleted of CD8+ T cells, MQ,and NK cells as described above; anti-I-Ah.d.q(ATCC; TIB 120) was included in the antibody cocktail. By this approach the starting population of CD4+ Tcells had a purity of at least 95% in the case of young and old mice. These CD4+ Tcells were incubated (45 min on ice) with antibody 16A (which was kindly donated by Dr. K. Bottornly, Yale University, New Haven, CT,[ 13]), washed twice and incubated on petri-dishes coated with goat anti-rat IgG as described above. After incubation during 90 min, nonadherent cells were removed by gently swirling and washing the plates with PBS containing 1% FCS.These cells will be referred to as CD45RBI0 CD4+ Tcells. Plates were washed again twice, after which the adherent cells were collected by vigorously pipetting; these cells will be referred to as CD45RBhiCD4+ T cells.

For proliferative responses 50000 total T cells or CD4+ Tcells were stimulated with Con A or anti-CD3 in the presence of accessory cells as described above. Stimulation with PMA/ionomycin was performed in the absence of additional accessory cells. Maximal responses were found after 3 days of culture for total Tcells and after 4 days of culture for CD4+ T cells, and these were measured by labeling the cells for 6 h with 0.25 pCi (= 9.25 kBq) [3H]dThd (sp. act. 2 Ci/mmol; Radiochemical Center, Amersham, GB). Where indicated, 50 U/ml human rIL 2 (Cetus, Emeryville, CA) or 50 U/ml purified murine rIL 4 (a kind gift of Dr. H. Savelkoul, Department of Immunology, Erasmus University, Rotterdam, The Netherlands) were added to supplement the cultures. 2.4 Cytokine assays IL 2 and IL 4 were assayed with the use of CTLL-2 cells. For IL 2 , 5 x 103 cells were cultured in SCM in the presence of 5 pg/ml anti-IL 4 antibody ll.B. 11 [25] and serially diluted SN to be assayed. Cells were pulsed with 0.25 pCi ["HIdThd from 20-24 h of culture. For the measurement of IL 4, 10' cells were used, whereas cultures were performed in the presence of 5 yglml anti-IL 2 antibody S4B6 [26], and labeling of the cells was from 18 to 24 h of culture. IL 2 and I L 4 were expressed in U/ml, i.e. the reciprocal of the dilution which resulted in a half-maximal response. SN of transfectants secreting recombinant murine IL 2 or IL 4 [27] (kindly provided by Dr. F. Melchers, Basel Institute for Immunology, Basel, Switzerland) served as a standard to prepare the relevant calibration curves. IFN-y was measured by ELISA (Holland Biotechnology, Leiden, The Netherlands). Biological activity was verified by the use of a bioassay as described elsewhere [28] and correlated to the NIH standard Gg 02-901-533.

2.5 Phenotypic analysis Purity of the cells was determined after incubation of the cells with FITC-labeled antibodies by FCM analysis with the use of a FACScan (Becton Dickinson, Sunneyvale, CA). CD45RB expression was studied in an indirect

Eur. J. Immunol. 1991. 21: 273-281

Aging, maturation and lymphokine production

275

immunofluorescence technique, by incubating the cells with antibody 16A and by using FITC-labeled mouse anti-rat Ig as a second antibody.

2.6 Statistical analysis Results were subjected to statistical analysis by using the Mann-Whitney U-test.

3 Results 3.1 Decreased expression of CD45RB on CD4+ T cells from old mice

-

.r

2

15

27

Age (months)

To investigate to what extent age-related changes in immune regulation are due to a changed composition of the CD4+ T cell compartment, we first studied the expression of CD45RB on CD4+ T cells. Since we combined our phenotypic analysis with functional studies, we first purified the CD4+ Tcells from individual mice of different ages. It has been demonstrated previously that CD45RB is present on CD4+ Tcells either in low or in high density. For the calculation of the contribution of each of these subsets we,

1

I I I

2 rno

15 rno

iy

Figure 2. The percentage of CD45RBhiCD4+ T cells decreases during aging. Purified CD4+ Tcells from mice of different ages were stained with antibody 16A. Each symbol represents one individual mouse.

therefore, preferred an arbitrary marker setting (determined using cells from 15-month-old mice in which both subsets are about equally represented) rather than the marker setting for the negative control (second antibody only). As can be seen in Fig. 1, FCM analysis of cells from representative mice of different ages resulted in the discrimination of two cell populations, a population that expressed this determinant in low density in addition to a strongly positive subset. In young mice, there was a predominance of CD4+ Tcells that express CD45FU3 in high density; in old mice the fraction of CD45RBhiCD4+ T cells had decreased in favor of the CD45REi'" CD4+ Tcells. This phenotypic shift within the CD4+ Tcell compartment is further illustrated in Fig. 2 for three groups of five mice of different ages. In2-month-old mice about 80% of the CD4+ Tcells expressed CD45RB in high density, in 15-month-old mice this fraction had already decreased to 60%, whereas a further decline to 40% was noted for 27-month-old mice. In separate experiments we found that this decline in CD45RB expression was accompanied by an increased expression of Pgp-1 on the CD4+ T cells (data not shown).

3.2 The age-related decline in IL 2 production is accompanied by an increase in IL 4 production

.-

, I I

27 rno

I

4

1

Fluorescence intensity

Figure 1. Expression of CD45FU3on CD4+Tcells from 2-, 15-and 27-month-old mice. Cells were incubated with rat antibody 16A, after which bound antibodies were detected with FITC-labeled mouse anti-rat Ig.

To examine whether the age-related change in the composition of the CD4+ Tcell population as described above was accompanied by a shift in the profile of lymphokines produced, we studied the production of IL 2 and IL 4 after activation of the cells by different stimuli. Kinetic studies showed that under the present culture conditions, maximal concentrations of IL 2 were detected after 3 days of culture of the purified CD4+ Tcells, whereas highest levels of IL 4 were detected after 4 days of culture (data not shown). As shown in Table 1, for groups of five mice each, stimulation with anti-CD3 resulted in a mean IL 2 production of 9.1 U/ml in young mice, which declined to 5.5 U/ml in the intermediate group and to 1 U/ml in the old mice (Expt. 2, p < 0.00.5). Con A and PMA/ionomycin were more potent in the induction of IL 2 production by CD4+T

276

L. Nagelkerken, A. Hertogh-Huijbregts, R. Dobber and A. Drager

Eur. J. Immunol. 1991. 21: 273-281

cells from young mice, in that about 65 U/ml and 100 U/ml were produced, respectively. An age-related decline in IL 2 production was observed after stimulation with Con A (Expt. 2, p < 0.005). Differences were less evident after costimulation with PMA and ionomycin (Expt. 2, p < O.Ol).These data confirm previous observations made by several investigators using unpurified lymphocytes, i. e. an age-related decline in IL 2 production is found with a variety of mitogens, whereas PMA/ionomycin is capable of inducing a substantial IL 2 production in old cells. Different results were obtained for the production of IL 4. IL 4 production was only occasionally detected after stimulation of the cells with anti-CD3. However, IL 4 could be detected after stimulation with Con A or PMAhonomycin. This was most obvious for cells derived from aged mice. As indicated inTable 1, the production of IL 4 by CD4+ T cells increased during aging. In response to Con A, the mean IL 4 production by CD4+ T cells from the 2-month-old mice was 5.7 U/ml. In the group of 15-month-old mice, the mean IL 4 production was 10.7 U/ml (p < 0.01) and in the group of 27-month-old mice 17.0 U/ml (p < 0.005). Stimulation with PMA/ionomycin resulted in lower levels of IL 4, but again IL 4 production was found to increase during aging. The same conclusion could be drawn from three additional experiments comparing pools of cells from young or old mice (data not shown).

CD45RB" /CD45RBh'

Figure 3. The age-related maturation of CD4+ T cells is accompanied by a shift from IL 2 to IL 4 production.Theratio IL 4/IL 2 in Con A SN of CD4+ T cells from individual mice of different ages (m, 2 months; A,15 months; 0,27 months) was plotted against the CD45FU3Io/CD45RBhiratio of the corresponding cells. See also Table 1.

er, under these conditions we still detected only low levels of IL 4. As will be shown later, this might have been due to the fact that Tcells from young mice - in contrast toTcells To determine whether the decrease in IL 2 and the increase from old mice - responded vigorously to IL 4 and hence in IL 4 production were related to the shift from CD45RBhi may have consumed all the I L 4 that had been proto CD45RJ3I0 CD4+ Tcells, we plotted the ratio IL 4/IL 2 duced. against the ratio CD45RB'o/CD45RJ3hifor each cell suspension and its corresponding Con A SN (results of Expt. 2 were used). As shown in Fig. 3, the shift from CD45RBhito 3.3 Age-related increase in the production of IFN-y CD45RB'O CD4+ T cells correlated (linear regression, IFN-y production could be detected after stimulation with r = 0.96) with the shift from IL 2 to IL 4 production. anti-CD3, Con A and PMA/ionomycin.With either of these Only low levels of IL 4 were detected when CD4+ Tcells stimuli, the highest amounts were found after 3 days of from young mice were investigated. Since IL 4 production culture. This conclusion was based on the results obtained has been suggested to occur predominantly by CD45RBl" with cells from 15- or 27-month-old mice. CD4+ Tcells CD4+ Tcells [13, 161 and these form a minority of the from 2-month-old mice, however, did not secrete detectayoung CD4+ T cells, we also studied young CD4+ Tcells at ble amounts of IFN-y, i.e. less than 0.3 ng/ml after higher cell numbers (l00OOO to 400000 cells/well). Howev- stimulation with PMAhonomycin or with anti-CD3 antiTable 1. The age-related decrease in IL 2 production is accompanied by an increase in IL 4 production

IL 2" (Ulml)

ApPJ (months)

Exp. I

3

Anti-CD3

31

XI

Ant i-CD3 Con A Con A

3 30

PMA/ionomycin PM A lionomycin

2

Ant i-CD3 Anti-CD3 Ant i-CD3

3

Exp. 2

1.5

27 2 15

27

-1

IS 17

Con A

Con A Con A PM Aliimomycin PM A /ionomycin PMA lionomycin

i4.n f 4.6 f 42.4 f 27.4 f 111.4 5H.4 f

2.3 1.5 6.9 7.4 13.0 16.3

Y . l f 1.7 s.5 f 2.0 1.0 f 1.3 64.8 f 9.1 19.2 f 6.5 11.8 f 4.2 100.1 f 16.0 win f 26.0 76.0 f Y.I

IL 4'' (Ulml)

< 2.5 f 0.6