Age-related changes of mitogen responsiveness in different lymphoid organs from outbred NMRI mice

Immunobiol., vol. 158,

pp. 439-449 (1981)

Institute for Clinical Immunology, Inselspital, CH-3010 Berne, Switzerland

Age-Related Changes of Mitogen Responsiveness in Different Lymphoid Organs from Outbred NMRI Mice FRANZISKA JONCOURT, FLORENCE BETTENS, FLEMMING KRISTENSEN, and ALAIN L. DE WECK Received December 12, 1980 . Accepted February 20, 1981

Abstract Lymphocytes from spleen, thymus and lymph nodes from individual young adult (3-4 months) and aged (26-30 months) NMRI mice were stimulated with the mitogens Con A, PHA and LPS. 24 hours later, the number of cells with increased RNA-content (G J cells) was determined by cytofluorometry. In parallel the lH-thymidine incorporation after 48 hours was measured for the same cell samples. Aged animals in average produced less G J cells and incorporated less lH-thymidine as compared to young adults. By calculating the lH-thymidine incorporation per GJ cell, the proliferative capacity of mitogen-induced G J cells can be estimated. These ratios are lower in aged mice as compared to young adult, suggesting that in these animals not only less cells can be activated as measured by cytofluorometry, but also from these activated cells again fewer continue the cell cycle by initiating DNA-synthesis. In response to Con A and PHA, aged mice in average produce less G J cells in all of the three lymphoid organs tested. In response to LPS, however, the young adult produced only few G J cells in lymph nodes and practically none in thymus, whereas in aged animals a considerable number of G J cells was found in both organs. Corresponding results were found for the lH_ thymidine incorporation. These results indicate that in addition to the reduction of the mitogen-response an age­ related change in the distribution of mitogen-responsive cells in the different lymphoid organs takes place.

Introduction It is well documented that immune functions generally decline with age (1). Most studies on such age-related changes in the functional abilities of

the immune system in mice have been restricted to spleen cells. It is, however, not clear whether the activities found in spleen cells are represen­ tative for the whole immune system. Different lymphoid organs do not respond indentically to antigenic and/or mitogenic stimulation, which is partially explained by the different functions of these organs. Furthermore, in spleen, the age-related degenerative changes are less pronounced than in the other lymphoid organs. It has long been discovered that the morphol­ ogy of the thymus and lymph nodes change considerably with age (2). The thymus, which is believed to playa crucial role in ageing, involutes and its lymphocytes are partially replaced by macrophages. An infiltration by

440 . F. ]ONCOURT et a!.

plasma cells and mast cells can also be observed (3). Morphologic changes in lymph nodes are apparent mainly as hypertrophy of the reticular cells, a decreased number of plasma cells, macrophages and mature lymphocytes (4 ). We have compared the ability of young and old spleen-, lymph-node­ and thymus cells from individual outbred NMRI mice to respond to different mitogens. The purpose of this study was to investigate age-related changes at the level of the different lymphoid organs. In each of the three organs tested we found a characteristic pattern of responsiveness in terms of mitogen-induced lymphocyte activation and proliferation. This pattern, however, is apparent only in young animals. In old mice the mitogen­ response does not only decline in magnitude, but in addition the pattern of mitogen-responsiveness characteristic for each organ in young mice disap­ pears.

Materials and Methods 1. Animals NMRI (SPF : Tif) female mice were purchased from Siiddeutsche Versuchstierfarm AG, Tuttlingen, Germany, and kept at our institute until use. The 50 % survival time of these mice at our institute is 28 months, and maximal responses to Con A, PHA and LPS are reached by the age of 3 months. For the experiments 3-4-month-old mice were therefore used as young adults and 26-30 month-old as aged.

2. Mitogens For the assessment of the response to mitogens, Concanavalin A (Pharmacia Fine Chemicals AG, Uppsala, Sweden), Phytohemagglutinin P (Difco Laboratories, Detroit, Mich.) and Lipopolysaccharide E. coli 055 : B5 (Difco Laboratories) were used in the following final concentrations: Con A 2 !J.g/ml, PHA 20 !J.g/ml and LPS 50 !J.g/m!. These concentrations were found to give maximal stimulation in the system used. 3. Preparation of the cultures

The animals were sacrificed, and spleen, thymus, brachial and submandibular lymph nodes were removed. Cell suspensions were prepared by gently teasing the organs in Hanks' BSS. The cells were then washed and resuspended in RPM I 1640 medium containing 3 % FCS (Microbiological Associates, Bethesda, Maryland) and 100 IU penicillin and 100 !J.g/ml strep­ tomycin (Difco Laboratories). The cell concentrations were adjusted to 4 X 106 nucleated cells/m!. 100 !J.l aliquots of cell suspension were distributed into Micro test tissue culture plates (Falcon Plastics, Los Angeles, Calif.) containing 50 !J.l mitogen in appropriate dilution. In parallel macrocultures of 600 !J.lItube (no. 25200, Corning, New York) were prepared with the same relative concentrations of cells and mitogens. The cultures were incubated at 37°C in a humidified atmosphere containing 5 % CO2 ,

4. Determination of 3H-thymidine incorporation After 24 hrs of incubation, 0.3 !J.Ci tritiated thymidine (spec. activity 5 CilmM, Radiochem­ ical centre, Amersham, England) were added to each well of the microcultures. After 18-24 hrs of further incubation, the cultures were harvested with a Skatron Multiple Cell Culture Harvester (Lierbyen, Norway) and counted in a Beckmann Scintillation Counter (LS 3150 T). Results are expressed as CPM(stimulated cultures) - CPM(cootrol cultures)'

Age-related Changes in Different Lymphoid Organs· 441

5. Cytofluorometric measurements The method described by DARZYNKIEWICZ et al. (5) slightly modified by STADLER and DE WECK (6) and STADLER et al. (7) was used. Briefly, after 24 hrs of incubation, 0.2 ml cell suspension was mixed with 0.3 ml of a solution containing 0.1 % Triton x-l00 (Calbiochem, San Diego, Calif.), 0.05 n HCl and 0.15 M NaC!. Then 1.0 ml acridine orange (Serva Feinbiochemica, Heidelberg, Germany) (27.6 I-tg/ml in 5 X 10-3 M EDTA, 0.15 M NaCl, 0.1 M phosphate-citrate buffer, pH 0.6) was added. Green (DNA) and red (RNA) fluores­ cence of individual cells were measured using a cytofluorograph 4802 (Bio/Physics System Inc. Mahopac, New York). Dead cells, cell duplets and cells having an elevated DNA-content were excluded by a selective adjustment of the cytofluorograph. The counted cells were assumed to be Go and G 1 cells. The G 1 cells, defined as cells with elevated RNA-content but normal DNA­ content, were calculated on the basis of a selective setting of the multichannel analyzer, based on the histrogram of non-stimulated lymphocytes and a normal distribution curve of the Go cells, as described elsewhere (7). The setting was kept constant throughout a whole experiment and thus systematic errors are negligible, because the data are expressed as number of G 1 cells(stimulated) - number of G 1 cells(control).

6. Immunofluorescence A rabbit-anti-mouse-IgG antiserum was added to the cell suspension to yield a final concentration of 1 : 20. The cells were incubated for 30 minutes at 4 °C, then washed twice with Hanks' BSS. The cells were further incubated with FITC-coupled Swine-anti-rabbit Ig (N ordic Pharmaceuticals and Diagnostics, Tilburg, Netherlands) at a concentration of 1 : 6 for 30 minutes at 4 dc. The cells were washed twice in Hanks' BSS and examined under a fluorescence microscope using 0.05 % Evans blue as a counterstain.

Results 1. Mitogen-induced activation (Go-Gj shift)

Spleen-, lymphnode- and thymus cells from 12 young adult (3-4 months old) and 12 aged (26-30 months old) outbred NMRI mice were stimulated individually with Con A, PHA and LPS. 24 hrs after the onset of the cultures, the number of Gj-cells as a parameter for activation of the cells was determined using cytofluorometry. An incubation time of 24 hrs was chosen because the number of Gj-cells reaches plateau values at that time (data will be published elsewhere). The results shown in Figure 1 can be summarized as follows: on the average, spleen cells from aged animals produced less Gj-cells upon stimulation with either one of the three mitogens. The same was true for the Con A- and PHA-stimulated lymph node cells. Upon stimulation with LPS, however, the lymph nodes from young adult animals produced only relatively few Gj-cells, whereas in those from old animals a considerable number of Gj-cells could be found. In thymus very similar numbers of Gj-cells were produced in Con A- and PH A-stimulated cultures of both young and old NMRI mice. Practically no Gj-cells were formed as expected in cultures of young thymocytes stimulated with LPS. Thymocytes from more than half of the aged mice, on the other hand, produced considerable numbers of Gj-cells in response to LPS stimulation.

442 . F.

]ONCOURT

et al.

:1 WOD WOw LPS

.!!1 Qj

",

c5

'0

u·

:1 .y~S . M·· U . '.'

o

z

PHA

Con A

wu·u .'.

;;~:.:: ,--,--

,--

9 6

3

r-:-t=-:' ..... .

..

~

S

LN

T

.,

"

~ . ... r:+.. .~ ..... S

LN

T

Fig. 1. Activation of mitogen-stimulated lymphocytes from spleen, lymph nodes and thymus from young adult and aged NMRI mice measured by cytofluorometry after 24 hrs of incubation. The results are expressed as G1(stimulated) - G1(control)' Each dot represents the response of one animal. The horizontal bars indicate the mean values. S: spleen, LN: lymph nodes, T: thymus.

2. Mitogen-induced proliferation In parallel with the cytofluorometric measurements, the 3H-thymidine incorporation as a parameter for cell proliferation was determined. Accord­ ing to kinetic studies, 3H-thymidine incorporation is maximal after 48 hrs of incubation (data not shown). All cultures were therefore harvested after 48 hrs. The results are shown in Figure 2. In principle the same pattern appeared for thymidine incorporation as for the Gl-cells. In the spleen, thymidine incorporation was lower for cells from old animals stimulated with either one of the three mitogens. In lymph nodes, too, Con A- and PHA-stimulation resulted in lower thymidine incorporation in old cells than in young. For the LPS response in lymph nodes again the opposite was observed. On the average, however, the age-related decrease of the thy­ midine incorporation was more pronounced than the decrease in produc­ tion of Gl-cells. Thymocytes from young and old NMRI mice incorporated similar amounts of thymidine upon Con A-stimulation, whereas after PHA-stimulation only old thymocytes responded. Thymocytes from young animals, although they were activated by PHA (see Fig. 1) did not incorporate 3H-thymidine, which is consistent with the well-known fact,

Fig. 2. Proliferative response of mitogen-stimulated lymphocytes from spleen. lymph nodes and thymus from young adult and aged NMRI mice measured by 3H-thymidine incorpora­ tion. The cultures were harvested after 48 hrs of incubation. The results are expressed as CPM(stimulated) - CPM(control)· Each dot represents the response of one animal. The horizontal bars indicate the mean values. S: spleen. LN: lymph nodes. T: thymus.

that upon PHA -stimulation T -cells are not able to proliferate in the absence of a sufficient number of macrophages or certain macrophage-derived products (8). Finally, upon LPS-stimulation, again young thymocytes did not respond with thymidine incorporation, whereas most of the old did.

3. Immunofluorescence As a control, in order to see whether any relation existed between the distribution of LPS-reactive cells and IgG-bearing cells, stained cells from the different lymphoid organs of young and old mice were examined Table 1. Comparison of the percentage of IgG-bearing cells in spleen. lymph nodes and thymus from young adult and aged NMRI mice as determined by fluorescence microscopy % IgG-bearing cells

Source of cells

Spleen

young old

26 ± 32

') mean ± standard deviation

±

Lymph node 6' ) 10

10 20

± ±

4 6

Thymus

o± 10

1

± 5

444 . F. JONCOURT et al.

c: ..

Ql!

0.",

0 .64

II) II) "0.

!I)~ II) C

"C §

W !!..H

05' 0 .63 0.54

0 .6'

o u c: ..

Ql!

E

0 .62

.s::." 0.0.

Q.11

>'"

..

....I~ c

\/I

::J

0.58 0 .52

0 .6'

0 .63

~

0

050

u

E ..

-G1

>~

0 .62

050

ill

0 .59

0 .72

0 .53

Con A

PHA

.s::.

1-",

0 .56

~

Con A

PHA

lPS

Spleen

c: ;:

0 .55

0.",

0 .52

II) II)

Con A

PHA

lPS

lymphnode Thymus

0.

!I)~ II)

"C

0 .56

C

c

--

CPM

0

0"

c: ..

!LIll

.s::. " 0.0.

!U§

E",

0 .5'

>0.

..

....I'-

- 0 55 0 .53

c

059

~ 8

E ;:

W

054

>0.

~

.s::.

1-", ~

-0 .59 -0.52 Con A

PHA

Spleen

lPS

Con A

PHA

LPS

Con A

PHA

Lymphnode Thymus

Fig. 3. Correlation coefficients. In the upper half of the squares the values for young mice are listed and in the lower half the values for the aged mice. Only values falling into the 10 % limit in the probability test are listed. The highly significant correlation coefficients (2 % limit) are underlined.

additionally by fluorescence microscopy. In all three organs of old mice more IgG-bearing cells were present than in young, but the increase was less pronounced in spleen than in lymph nodes and thymus. Thymus from old mice on the average contained 10 % IgG-bearing cells, which is consistent with the finding that most thymi from old mice could be stimulated by LPS (Table 1). 4. Statistics

In a statistical analysis the numbers of G]-cells produced in the different organs in response to each mitogen were compared in all possible comb ina-

Age-related Changes in Different Lymphoid Organs· 445 Table 2. }H-thymidine incorporation per G, cell CPM / G 1

Organ

Mitogen

young NMRI

aged NMRI

p values

Spleen

Con A PHA LPS

14.78 ± 1.64') 15.57 ± 1.20 14.40 ± 2.03

10.65 ± 0.32 8.84 ± 1.57 7.85 ± 0.89

P < 0.025 P < 0.005 p < 0.005

Lymph nodes

Con A PHA LPS

14.28 ± 1.58 13.08 ± 1.80 8.77 ± 1.52

9.34 ± 2.50 4.96 ± 0.70 4.91 ± 0.87

P < 0.1 P < 0.0005 P < 0.025

Thymus

Con A PHA LPS

11.95 ± 1.39 0.98 ± 0.42 1.70 ± 0.54

9.73 ± 1.74 9.64 ± 6.22 12.91 ± 2.92

P < 0.2 P < 0.1 P < 0.0025

' ) standard error of the mean

tions in order to see whether any of the responses were correlated either positively or negatively. The same analysis was made also for the prolifera­ tive responses. The correlation coefficients are shown in Figure 3. For the G(-cell production, positive correlations were found between Con A and PHA responses within the same organ in all cases except the thymus of old mice. Positive correlations were found also between the Con A and LPS responses in spleen and thymus of old mice and between PHA responses in spleen and thymus of young mice. For the proliferative responses, positive correlations were found again between Con A and PHA responses in lymph node as well as thymus but not in spleen.

5. CPMI Gj The proliferative capacity of the activated cells can be estimated by calculating the thymidine incorporation per G(-cell. The ratios obtained for the different mitogen-responses are listed in Table 2. With the obvious exception of the thymus, these ratios are lower in old mice compared to young, which means that old mice incorporate less thymidine per G(-cell than young ones.

Discussion Most studies on age-related changes in the immune functions are per­ formed in spleen cells. These studies in general reveal a decline of the immune responses with age. The decline of the mitogen responses is an Abbreviations:

Can A

=

Lipopolysaccharide; Hanks' BSS

Concanavalin A; PHA = Phytohemagglutinin; IPS = = Hanks' balanced salt solution; FCS = Fetal calf serum.

446 . F. ]ONCOURT et al.

often described phenomenon of ageing mice as well as other ageing species (9, 10, 11, 12). It is still controversial, however, whether this decline is due to qualitative or quantitative changes of lymphocytes, namely a lower response of an identical cell population or a decrease in number of normally responding cells. Reports supporting either of the two hypotheses have been published in recent years (13,14). According to our results, qualitative and quantitative changes appear to playa role in the age-related decline of the immune-responses. The cytofluorometric measurements show that in cell populations of various lymphoid organs of old mice generally less cells can be activated to RNA-synthesis by T-cell mitogens as compared to young mice. This probably reflects a quantitative change in the number of cells capable of being activated, due to a decrease in the percentage of responding cells. It is possible, however, that this decreased percentage of responding cells is compensated for by an increase in the total number of lymphocytes present in lymphoid tissue of old mice (15). In this case, the quantitative change would correspond to an increase in number of non­ responding cells rather than a decrease in total number of responding cells. In addition to this, as shown in Table 2, we find lower ratios of CPM/G 1 in old mice than in young. If we assume that every cell synthesizing DNA incorporates the same amount of 3H-thymidine, it is clear from these results that in old mice the proportion of activated cells that subsequently undergo DNA-synthesis decreases. This means that in old mice fewer of the activated cells are able to complete the cell cycle, which indicates a qualitative defect. Either the cells are lacking a signal they need in order to make the transition from G 1 to the S-phase or they are actually unable to complete the cell cycle due to a defect within the cells. A more detailed study on these findings will be published elsewhere. Furthermore, our results, where spleen, lymph nodes and thymus have been compared, show that there is also a change in the distribution of the reactive cells in the different lymphoid organs. Comparing the mean values in Figures 1 and 2, it is evident that in young mice we find a characteristic distribution of responsiveness to Con A, PHA and LPS in the three organs, as measured by activation as well as proliferation. In the old animals, however, the responses to a given mitogen tend to become equalized in all three organs. As all the responses measured are produced by the same number of cells (400,000/culture), this means that in the old animals in contrast to the young the proportion of cells responding to Con A, PHA and LPS, respectively, tends to be equal in spleen, lymph nodes and thymus. The results obtained by immunofluorescence showed that lymph nodes from young mice contain on the average 10 % IgG-bearing cells, in most cases, however, these lymph node cells could not be activated nor did they proliferate upon stimulation with LPS. A similar observation has been reported by DUMONT (16). This author speculates that the B-cells in the spleen and those in lymph nodes might belong to different subpopulations.

Age-related Changes in Different Lymphoid Organs· 447

More recent studies show that in fact various B-cell mitogens including LPS each stimulate only a subpopulation of B-cells, probably representing populations of cells at different stages of differentiation (17). Alternatively it is possible that B-cells from young lymph nodes lack a population of adherent cells, which has been shown together with T -cells to facilitate the LPS-response of B-cells in Peyer's patches (18). Furthermore in this system as well as many others, 2-mercaptoethanol has been shown to replace adherent cells, most likely by reducing glutathione which can scavenge oxygen-derived radicals deleterious to macrophages and lymphocytes (19). In the present study, no 2-mercaptoethanol or other reducing agent has been added to the culture medium. Therefore it cannot be excluded that the changes in LPS-responsiveness of lymph node cells are due to a change in the population of adherent cells. In either case it is clear from our results that in old mice the composition of the lymph node cells is altered in such a way that more cells are capable of responding to LPS. In thymus of young animals we find practically no LPS-responding cells. At the same time there are practically no IgG-bearing cells as judged by the immunofluorescence technique. Thymi from young animals therefore do not seem to contain any significant number of B-cells. In contrast to this, we find LPS-reactive cells as well as IgG-bearing cells in thymus of aged mice. It could be argued that these cells are contaminating cells from parathymic lymph nodes. This possibility cannot be ruled out completely, as in fact in many cases it is difficult to distinguish in old mice the thymus from surrounding tissue. As can be seen in Table 1, old lymph nodes on the average contain 20 % IgG­ bearing cells. If the 10 % IgG-bearing cells found in old thymi were in fact contaminating cells, 50 % of the cells measured as thymic would be contaminating cells, which is certainly not the case. We therefore conclude that the thymus with age is indeed infiltrated by B-cells. The statistical analysis of individual responses show that the three organs tested apparently are independent of each other in terms of mitogen responsiveness (Table 2). This means that e.g. an animal with a high Con A response in the lymph node does not necessarily have a high Con A response in spleen nor in the thymus. According to our results and with the only exception of the PHA response between thymus and spleen, correla­ tions exist only between Con A and PHA response within the same organ. A similar lack of correlation between the magnitude of mitogen-responses in different lymphoid organs has been demonstrated in 3 aged C57BL mice (20). This illustrates that general conclusions on the immune response based on analysis of single organ cell populations may not be warranted and that much more attention should be paid to the compartmentalisation of re­ sponses. This study is based on results obtained in an outbred strain of mice representing a heterogeneous population from a genetic point of view. Due to this heterogeneity a big variability is found between the responses of individual young as well as aged mice. It can be argued that in such a

448 . F. JONCOURT et al .

situation it becomes impossible to differentiate changes in immune responses due to the ageing process from those due to different genetic background. The fact that even under such conditions some changes, such as the modification in LPS-responsive populations and the efficiency in proliferation of activated cells (ratio CPM/G t) are highly significant, suggest strongly that these are age-related changes. It is also possible that the variations found between aged individuals are not always an expression of inherent differences between animals: the changes in cell composition and functional abilities in various organs may be indications of a pathologic condition. Similar changes have been observed in thymus of ageing S}1I} mice, which develop leukemia follow­ ing treatment with chemical carcinogens (21). In fact, approximately 5-10 % of the 26-30 months old NMRI show leukemic changes as evalu­ ated by macro- and microscopic examination. Although such apparently sick animals have been excluded from the present study, it must be assumed that some of the apparently healthy animals are nevertheless in a pre­ leukemic stage, which might influence the cell composition of their lym­ phoid organs. Our results show that in ageing mice changes take place not only in the overall apparent lymphocyte function, but also at the level of the lymphoid organs. It is possible that these functional and morphological changes in the lymphoid organs are a direct consequence of changes in the lymphocytes themselves, leading to altered homing properties of these cells. It is also possible, however, that they are due to more general physiological changes within the ageing organism or the lymphoid organs, such as for example the age-related decline of epithelial cell activity within the thymus and the serum thymic factor (22). Most likely, however, the changes are due to a combination of the two possibilities, which would explain the complexity and variability of the ageing patterns seen in outbred NMRI mice.

References 1. KAY, M. M. B. 1979. An overview of immune ageing. Mech. Age. D . 9: 39. 2. ANDREW, W. 1952. Problems of ageing. 3rd edn. Ed. Lansing, A. 1., Williams and Wilkins, Baltimore. p. 527. 3. HIROKAWA, K. 1977. Immunity and ageing. Ed. Makinodan, T ., Yunis, E. Plenum Press, New York. p . 51. 4. H ANNA, M . G., P . N OTfESHEIM, L. O GDEN, and T. MAKINODAN. 1967. Reduced immune potential of aged mice: significance of morphologic changes in lymphatic tissue . Proc. Soc. Exp. BioI. Med. 125: 882. S. D ARZYNKIEWICZ, Z ., F. T RAGANOS, T . SHARPLESS, and M . R . M ELAMED. 1976. Lympho­ cyte stimulation : a r apid multiparameter analy sis. Proc. N atl. Acad. Sci. USA 73: 2881. 6. STADLER, B. M ., and A . L. DE W ECK. 1978. Cytofluorometri c a nalysis o f macrophages activated in vivo or in vitro. Eu r. ]. Immunol. 8: 243 . 7. STADLER, B. M., F . KRISTENSEN, a nd A. L. DE WECK. 1980. Thymocy te activation by Iymphokines: Direct assessment of Go-G J transition by Flow-Cytometry. Immunol. 55: 436.

Age-related Changes in Different Lymphoid Organs· 449 8. WAKSMAN, B. H., and A. B. WAGSHAL. 1978. Lymphocyte functions acted on by immunoregulatory cytokines. Significance of the cell cycle. Cell. Immunol. 36: 180. 9. HORI, Y., E. H . PERKINS, and M . K. HALSALL. 1973. Decline in phytohemagglutinin responsiveness of spleen cells from ageing mice. Proc. Soc. Exp. BioI. Med. 144: 48. 10. RODEY, G. E., and P. O. TEAGUE. 1970. Ageing and in vitro lymphocyte transformation in different mouse strains. Fed. Proc. 29: 502. 11. MATHIES, M., L. LIpps, G. S. SMITH, and R. L. WALFORD. 1973. Age-related decline in response to phytohemagglutinin and pokeweed mitogen by spleen cells from hamsters and a long lived mouse strain. J. Geront. 28: 425. 12. HALLGREN, H . M., J. H. KERSEY, D. P. DUBEY, and E. J . YUNIS. 1978. Lymphocyte subsets and integrated immune functions in ageing humans. Clin. Immun. 10: 65. 13. ABRAHAM, c., Y. TAL, and H. GERSHON . 1977. Reduced in vitro response to Con A and LPS in senescent mice: a function of reduced number of responding cells. Eur. J. Immunol. 7: 301. 14. CALLARD, R. E., A. BASTEN, and R. V. BLANDEN. 1979. Loss of immune competence with age may be due to a qualitative abnormality in lymphocyte membranes. Nature 281: 218. 15. CALLARD, R. E., and A. BASTEN. 1977. Immune functions in aged mice. 1. T-cell responsiveness using PHA as a functional probe. Cell. Immunol. 31: 13. 16. DUMONT, F. 1973. Absence de stimulation par la lipopolysaccharide des lymphocytes B du ganglion lymphatique peripherique de la souris. C. R. Acad. Sci. Paris 276: 2605. 17. GRONOWICZ, E ., and A. COUTINHO. 1974. Selective triggering of B-cell subpopulations by mitogens. Eur. J. Immunol. 4: 771. 18. KA GNOFF , M. F., P. BILLINGS, and M. COHN. 1974. Functional characteristics of Peyer's Patch lymphoid cells. II. Lipopolysaccharide is thymus dependent. J. Exp. Med. 139: 407. 19. H OFFELD, J. T., and J. J. OPPENHEIM. 1980. Enhancement of the primary antibody response by 2-mercaptoethanol is mediated by its action on glutathione in the serum. Eur. J. Immunol. 10: 391. 20. ADLER, W . H ., K. H. JONES, and H. N ARIUCHI . 1977. Recent advances in clinical immunology. 1st edn. Churchill Livingstone, London. p. 85. 21. BEN-YAAKov, M., and N. HARAN-GHERA. 1975. T and B lymphocytes in thymus of SJLI J mice. Nature 255: 64. 22. BACH, J. F., M. DARDENNE, J. M. PLEAU, and A. M. BACH. 1975. Isolation, biochemical characterization and biological activity of circulating thymic hormone in the mouse and in the human. Ann. N .Y. Acad. Sci. 249: 186. FRANZISKA JON COURT, Institute for Clinical Immunology, Inselspital, CH-3010 Berne, Switzerland