Eur. J. Immunol. 1992. 22: 1667-1669
Cord blood lymphocyte responses against heat-shock protcin
1667
Short paper Heinrich Peter Fischer.*, Claire E. M. Sharrock* and Gabriel S. Panayi. Rheumatology Unit, UMDS Guy's Hospital, Department of Immunology*, St. George's Hospital Medical School, London
High frequency of cord blood lymphocytes against mycobacterial 65-kDa heat-shock protein" A high frequency of nonadherent mononuclear cells in human cord blood prolifcrates in response to mycobacterial 65-kDa heat-shock protein. The frequency range in cord blood is not different from that in peripheral blood of Bacillus Calmctte-Gu6rin vaccinated adults. In comparison we found 10 to 100 times lower frequencies to purified protein derivative in nonadherent cord blood mononuclear cclls than in adult peripheral blood mononuclear cells. These findings may provide experimental support for Cohen's theory of the immunological homonculus.
1 Introduction
2.2 Cell culture
The immunological homonculus, a term coined by I . Cohen [ 1, 21 in analogy to the neurological homonculus, describes the immune system's picture of the self. In Cohcn's view, the T cell rcpertoire is dominated by a limited set of self antigens, each of which is encoded in a cellular network. One of the prime candidates for such a dominant self antigcn is human heat-shock protein (hsp) 60 [3]. Similar proteins occur in a large variety of bacteria, including mycobacterial hsp 65 [4, 51. All members of the hsp 60 family have a high dcgree of amino acid homology and immunological cross-reactivity has been shown [3, 4, 6, 71. Following our work with mycobacterial hsp 65 in patients with rhcumatoid arthritis (RA), [S] we have used limiting dilution analysis to cnumerate hsp 65-responsivc lymphocytes in human cord blood. The homonuculus theory predicts a relatively high lymphocyte frequency to hsp 65 in cord blood, i.e. prior to exposure of the immune system to cxternal antigens, compared with antigens that need priming such as tuberculin PPD.
Cells were cultured in 96-well round-bottom microtiter plates (Nunc) at 37 "C in humidified atmosphere with 5 % CO2/95 YO air. Culture medium consisted of RPMI 1640 (Gibco, Paisley, Scotland) supplemented with 7,s % heatinactivated human AB serum, 100 U/ml penicillin, 100 pg/ml streptomycin and 2 mM I,-glutamine.
2 Materials and methods 2.1 Cord blood and tissue typing Fresh cord blood (CB) was obtained by sterile puncture o f a major placental vein. Pcripheral blood (PB) was obtained from fivc healthy adult controls. In both cases heparinized tubes were used. After Ficoll-Hypaque separation from PB and CB, mononuclear cells (MNC) were deplctcd of adhcrcnt cells by adherence to plastic petri dishes (Nunc, Roskilde, Denmark).Tissue typing was performed with the standard NIH technique.
[I l0089]
* This work was gcncrously supported by the Deutsches RheumaForschungszcntrum, Bcrlin and partially by grant PS2 from the Arthritis and Rhcumatisrn Council of the UK. Correspondence: Gabricl S. Panayi, Rheumatology Unit, UMDS Guy's Hospital, London SEl 9RT, GB
0 VCH Vcrlagsgesellschaft mbH, D-6940 Wcinhcim, 1992
2.3 Antigens and mitogens Tuberculin PPD (batch 2%) was a gift from the Central Veterinary Laboratory in Weybridge, GB. Highly purified recombinant 65-kDa hsp of M . leprue was kindly provided by Dr. M. J. Colston, National Institute for Medical Research, London [9]. Optimal concentrations were determined for each batch and found to be 5 pg/ml for PPD and 12.5 pg/ml for hsp 65. The endotoxin concentration in the final cultures amounted to 130 pg/ml for hsp 65 and 8 pg/ml for PPD. Further reductions of the endotoxin content to 54 pg/ml and 1.3 pg/ml, respectively with a Detoxigen column (Pierce Warriner, GB) did not affect the ability of hsp 65 to stimulate nonadherent MNC from CB or adult PB (data not shown). The minimal stimulatory concentration of purified E. coli LPS (Sigma, Poole, GB) for human PBMC was 0.5-1 ng/ml (data not shown). PHA (Wellcome, Beckenham, GB) was used at 2 pg/ml.
2.4 Limiting dilution cultures A protocol similar to that previously described [8] was used. Briefly, graded numbers of nonadherent MNC were grown in the presence of lo4irradiated (5000 rad) plastic-adherent cells in 2OO-kl cultures. For antigen-specific responses 24 to 40 replicate wells per responder cell dilution were set up for each population. As a control, 16 to 40 replicate wells per responder cell dilution were set up without antigen. Sixteen replicates were used for irradiated adherent cells and antigen without nonadherent MNC. As a positive control, four replicates of PHA-pulsed cultures (15 000 nonadherent cells and lo4 adherent cells/well) were used for each subpopulation of nonadherent cells. Culture conditions and antigen concentrations were as stated above. Cells were pulsed with 0.5 pCi = 18.5 kBq ]H"[ dThd (Amersham, 00 14-2980/92/0606-1667$3.50+ ,2510
Eur. J. Immunol. 1992. 22: 1667-1669
H. P. Fischer, C. E. M. Sharrock and G. S. Panayi
1668
Int., Amersham, GB; sp. act. 5 Cilmmol) for the last 6 h of the 120-h culture period, and harvested onto glass fiber paper. Beta emission was measured with a multidetector direct beta counter (Packard, Downer’s Grove, IL).
ating to hsp 65 is virtually identical. By contrast, the frequencies against the second antigen, PPD, are 10 to 100 times lower in CB MNC than in adult PB MNC. Interestingly, we did not find a measurable frequency of responding CB MNC to human myelin basis protein (data not shown).
2.5 Statistical analysis
We have recently shown that there is no significant difference in the frequency of nonadherent PB MNC responding to hsp 65 between normal adults and patients with FL4 and between PB and synovial fluid (SF) of individual FL4 patients [S].
A culture well was scored positive when the cpm measured for that well exceeded the mean plus 3 SD of its corresponding controls (cultures without antigen). Results were calculated using maximum likelihood analysis based on the Poisson distribution [lo]. p values (n-1 degree of freedom) >0.05 are consistent with single-hit kinetics [lo]. Calculations were performed with a computer program based on GLIM 3.12 (Royal Statistical Society, London).
3 Results and discussion As shown in Table 1 and Fig. 1 the frequency range of nonadherent MNC from human CB and adult PB prolifer-
This kind of stability of the immune response in different stages of maturation, and in health and disease, is just what might be expected of a dominant self antigen to which a response is already anticipated by a preformed lymphocyte network [2]. A major role for such a regulatory network could be to control anti-self rectivity.
In three recently published reports the proliferation of Vy9/V8 T cells in response to exogenous and endogenous “superantigen-like’’ ligands [11-13] was described. In this
Table 1. Frequencies of nonadherent PB MNC of five healthy adult individuals and nonadherent CB MNC of seven human neonatals against tuberculin PPD and hsp 65 Subject Adult Adult Adult Adult Adult
1 2 3 4 5
CB 1 CB 2 CB 3 CB 4 CB 5 CB 6 CB 7
HLA-DR
314 7 114 4 114 314 317 113 317 415 31w8 512
Reciprocal frequency for: PPD hsp 65
300 (225-390)a) 370 (215-640) 1490 (1140-1935) 1520 (1130-2040) 3400 (2740-4325) 5920 (4380-8000) 7590 (5270-10 070) 26 410 (20 210-34 520) 26910 (20580-35 200) 28 530 (21 750-37 420) 180300 (107 000-304 00) 198200 (121 100-3244000)
5290 (4210-6640) 3380 (2675-4280) 9600 (7390-12500) 3310 (2570-4260) 5990 (4870-7370) 1690 (1310-2200) 2170 (1610-2920) 3360 (2650-4240) 11780 (9200-15 090) 3760 (2990-4730) 14500 (11 110-18860) €4300
a) Figure in parentheses are upper and lower 95 % confidence limits of the frequency estimates. The titration curves of all experiments shown were consistent with single-hit kinetics. All p-values >0.05.
Adult Controls 10 -2
!A
3 0
10 -3: 0
+O
A 0 O+
A 10 -4?
** 8
0
+
0
O+.
10 -5:
$
10 -6
PPD
HSP
Figure 1 . Scatter plot of the frequencies shown in Table 1. In each graph a pair of symbols represents one individual.
Eur. J. Immunol. 1992. 22: 1667-1669
Cord blood lymphocyte responses against heat-shock protein
context members of the hsp 60 family where named as possible candidate “superantigens”. The present findings argue against hsp 65 stimulation having elicited proliferation of this subpopulation of $6 T cells, as the frequency of V,9/V62 cells increases dramatically after birth [14], whereas the frequency of hsp 65-responsive nonadherent cells does not change between CB and adult PB. We thank the Tissue Typing Laboratory, St. George’s Hospital, London for the tissue typing and the Department of Obstetrics and Gynaecology at St. George’s Hospital for providing fresh placentas. We thank Professor N . A. Mitchison for inspiring discussions.
Received November 12, 1991; in revised form January 12, 1992.
4 References 1 Cohen, I. R., Atlan, H. and Cohen, I. R., (Eds.), Theories of Immune Newtworks, Springer Verlag, Berlin 1989, p.6. 2 Cohen, I. R. and Young, D. B., Immunol. Today 1991.12: 105. 3 Jindal, S., Dudani, A. K., Singh, B., Harley, C. B. and Gupta, R.S., Mol. Cell. Biol. 1989. 9: 2279.
1669
4 Young, D. B., Ivanyi, J., Cox, J. H. and Lamb, J. R., Immunol. Today 1987.8: 215. 5 Lindquist, S., Annu. Rev. Biochem. 1986. 55: 1151. 6 Koga, T., Wand-Wuertenberger. A , , DeBruyn, J . , Munk, M. E., Schoel, B. and Kaufmann, S. H. E., Science 1989.245: 1112. 7 Munk, M. E., Schoel, B., Modrow, S., Karr, R. W.,Young, R. A. and Kaufmann, S. H. E . , J. Immunol. 1989. 143: 2844. 8 Fischer, H. F!, Sharrock, C. E. M., Colston, M. J. and Panayi, G . S., Eur. .I. Immunol. 1991. 21: 2937. 9 Lamb, F. I . , Kingston, A. E., Estrada-Ci, I. and Colston, M. J., Infect. Immun. 1988. 56: 1237. 10 Fazekas de St. Groth, S., J. Immunol. Methods 1982. 49: R11. 11 Fisch, F!,Malkovsky, M., Kovats, S., Sturm, E., Braakman, E., Klein, B. S. ,Voss, S. D., Morrissey, L. W., deMars, R., Wclch, W. J., Bolhius, R. L. H. and Sondel, F! M., Science 1990. 250: 1269. 12 DeLibero, G ., Casorati, G., Giachino, C., Carbonara, C., Migone, N., Matzinger, P. and Lanzavecchia, A., J. Exp. Med. 1991. 173: 1311. 13 Kabelitz, D.. Bender, A., Prospero,T.,Wesselborg, S., Janssen, 0. and Pechhold, K . , J. Exp. Med. 1991. 173: 1331. 14 Parker, C. M., Groh,V., Band, H., Porcelli, S. A , , Morita, C., Fabbi, M., Glass, D., Strominger, J. L.. and Brenner, M. B., J. Exp. Med. 1990. 171: 1597.
Cord blood lymphocyte responses against heat-shock protcin
1667
Short paper Heinrich Peter Fischer.*, Claire E. M. Sharrock* and Gabriel S. Panayi. Rheumatology Unit, UMDS Guy's Hospital, Department of Immunology*, St. George's Hospital Medical School, London
High frequency of cord blood lymphocytes against mycobacterial 65-kDa heat-shock protein" A high frequency of nonadherent mononuclear cells in human cord blood prolifcrates in response to mycobacterial 65-kDa heat-shock protein. The frequency range in cord blood is not different from that in peripheral blood of Bacillus Calmctte-Gu6rin vaccinated adults. In comparison we found 10 to 100 times lower frequencies to purified protein derivative in nonadherent cord blood mononuclear cclls than in adult peripheral blood mononuclear cells. These findings may provide experimental support for Cohen's theory of the immunological homonculus.
1 Introduction
2.2 Cell culture
The immunological homonculus, a term coined by I . Cohen [ 1, 21 in analogy to the neurological homonculus, describes the immune system's picture of the self. In Cohcn's view, the T cell rcpertoire is dominated by a limited set of self antigens, each of which is encoded in a cellular network. One of the prime candidates for such a dominant self antigcn is human heat-shock protein (hsp) 60 [3]. Similar proteins occur in a large variety of bacteria, including mycobacterial hsp 65 [4, 51. All members of the hsp 60 family have a high dcgree of amino acid homology and immunological cross-reactivity has been shown [3, 4, 6, 71. Following our work with mycobacterial hsp 65 in patients with rhcumatoid arthritis (RA), [S] we have used limiting dilution analysis to cnumerate hsp 65-responsivc lymphocytes in human cord blood. The homonuculus theory predicts a relatively high lymphocyte frequency to hsp 65 in cord blood, i.e. prior to exposure of the immune system to cxternal antigens, compared with antigens that need priming such as tuberculin PPD.
Cells were cultured in 96-well round-bottom microtiter plates (Nunc) at 37 "C in humidified atmosphere with 5 % CO2/95 YO air. Culture medium consisted of RPMI 1640 (Gibco, Paisley, Scotland) supplemented with 7,s % heatinactivated human AB serum, 100 U/ml penicillin, 100 pg/ml streptomycin and 2 mM I,-glutamine.
2 Materials and methods 2.1 Cord blood and tissue typing Fresh cord blood (CB) was obtained by sterile puncture o f a major placental vein. Pcripheral blood (PB) was obtained from fivc healthy adult controls. In both cases heparinized tubes were used. After Ficoll-Hypaque separation from PB and CB, mononuclear cells (MNC) were deplctcd of adhcrcnt cells by adherence to plastic petri dishes (Nunc, Roskilde, Denmark).Tissue typing was performed with the standard NIH technique.
[I l0089]
* This work was gcncrously supported by the Deutsches RheumaForschungszcntrum, Bcrlin and partially by grant PS2 from the Arthritis and Rhcumatisrn Council of the UK. Correspondence: Gabricl S. Panayi, Rheumatology Unit, UMDS Guy's Hospital, London SEl 9RT, GB
0 VCH Vcrlagsgesellschaft mbH, D-6940 Wcinhcim, 1992
2.3 Antigens and mitogens Tuberculin PPD (batch 2%) was a gift from the Central Veterinary Laboratory in Weybridge, GB. Highly purified recombinant 65-kDa hsp of M . leprue was kindly provided by Dr. M. J. Colston, National Institute for Medical Research, London [9]. Optimal concentrations were determined for each batch and found to be 5 pg/ml for PPD and 12.5 pg/ml for hsp 65. The endotoxin concentration in the final cultures amounted to 130 pg/ml for hsp 65 and 8 pg/ml for PPD. Further reductions of the endotoxin content to 54 pg/ml and 1.3 pg/ml, respectively with a Detoxigen column (Pierce Warriner, GB) did not affect the ability of hsp 65 to stimulate nonadherent MNC from CB or adult PB (data not shown). The minimal stimulatory concentration of purified E. coli LPS (Sigma, Poole, GB) for human PBMC was 0.5-1 ng/ml (data not shown). PHA (Wellcome, Beckenham, GB) was used at 2 pg/ml.
2.4 Limiting dilution cultures A protocol similar to that previously described [8] was used. Briefly, graded numbers of nonadherent MNC were grown in the presence of lo4irradiated (5000 rad) plastic-adherent cells in 2OO-kl cultures. For antigen-specific responses 24 to 40 replicate wells per responder cell dilution were set up for each population. As a control, 16 to 40 replicate wells per responder cell dilution were set up without antigen. Sixteen replicates were used for irradiated adherent cells and antigen without nonadherent MNC. As a positive control, four replicates of PHA-pulsed cultures (15 000 nonadherent cells and lo4 adherent cells/well) were used for each subpopulation of nonadherent cells. Culture conditions and antigen concentrations were as stated above. Cells were pulsed with 0.5 pCi = 18.5 kBq ]H"[ dThd (Amersham, 00 14-2980/92/0606-1667$3.50+ ,2510
Eur. J. Immunol. 1992. 22: 1667-1669
H. P. Fischer, C. E. M. Sharrock and G. S. Panayi
1668
Int., Amersham, GB; sp. act. 5 Cilmmol) for the last 6 h of the 120-h culture period, and harvested onto glass fiber paper. Beta emission was measured with a multidetector direct beta counter (Packard, Downer’s Grove, IL).
ating to hsp 65 is virtually identical. By contrast, the frequencies against the second antigen, PPD, are 10 to 100 times lower in CB MNC than in adult PB MNC. Interestingly, we did not find a measurable frequency of responding CB MNC to human myelin basis protein (data not shown).
2.5 Statistical analysis
We have recently shown that there is no significant difference in the frequency of nonadherent PB MNC responding to hsp 65 between normal adults and patients with FL4 and between PB and synovial fluid (SF) of individual FL4 patients [S].
A culture well was scored positive when the cpm measured for that well exceeded the mean plus 3 SD of its corresponding controls (cultures without antigen). Results were calculated using maximum likelihood analysis based on the Poisson distribution [lo]. p values (n-1 degree of freedom) >0.05 are consistent with single-hit kinetics [lo]. Calculations were performed with a computer program based on GLIM 3.12 (Royal Statistical Society, London).
3 Results and discussion As shown in Table 1 and Fig. 1 the frequency range of nonadherent MNC from human CB and adult PB prolifer-
This kind of stability of the immune response in different stages of maturation, and in health and disease, is just what might be expected of a dominant self antigen to which a response is already anticipated by a preformed lymphocyte network [2]. A major role for such a regulatory network could be to control anti-self rectivity.
In three recently published reports the proliferation of Vy9/V8 T cells in response to exogenous and endogenous “superantigen-like’’ ligands [11-13] was described. In this
Table 1. Frequencies of nonadherent PB MNC of five healthy adult individuals and nonadherent CB MNC of seven human neonatals against tuberculin PPD and hsp 65 Subject Adult Adult Adult Adult Adult
1 2 3 4 5
CB 1 CB 2 CB 3 CB 4 CB 5 CB 6 CB 7
HLA-DR
314 7 114 4 114 314 317 113 317 415 31w8 512
Reciprocal frequency for: PPD hsp 65
300 (225-390)a) 370 (215-640) 1490 (1140-1935) 1520 (1130-2040) 3400 (2740-4325) 5920 (4380-8000) 7590 (5270-10 070) 26 410 (20 210-34 520) 26910 (20580-35 200) 28 530 (21 750-37 420) 180300 (107 000-304 00) 198200 (121 100-3244000)
5290 (4210-6640) 3380 (2675-4280) 9600 (7390-12500) 3310 (2570-4260) 5990 (4870-7370) 1690 (1310-2200) 2170 (1610-2920) 3360 (2650-4240) 11780 (9200-15 090) 3760 (2990-4730) 14500 (11 110-18860) €4300
a) Figure in parentheses are upper and lower 95 % confidence limits of the frequency estimates. The titration curves of all experiments shown were consistent with single-hit kinetics. All p-values >0.05.
Adult Controls 10 -2
!A
3 0
10 -3: 0
+O
A 0 O+
A 10 -4?
** 8
0
+
0
O+.
10 -5:
$
10 -6
PPD
HSP
Figure 1 . Scatter plot of the frequencies shown in Table 1. In each graph a pair of symbols represents one individual.
Eur. J. Immunol. 1992. 22: 1667-1669
Cord blood lymphocyte responses against heat-shock protein
context members of the hsp 60 family where named as possible candidate “superantigens”. The present findings argue against hsp 65 stimulation having elicited proliferation of this subpopulation of $6 T cells, as the frequency of V,9/V62 cells increases dramatically after birth [14], whereas the frequency of hsp 65-responsive nonadherent cells does not change between CB and adult PB. We thank the Tissue Typing Laboratory, St. George’s Hospital, London for the tissue typing and the Department of Obstetrics and Gynaecology at St. George’s Hospital for providing fresh placentas. We thank Professor N . A. Mitchison for inspiring discussions.
Received November 12, 1991; in revised form January 12, 1992.
4 References 1 Cohen, I. R., Atlan, H. and Cohen, I. R., (Eds.), Theories of Immune Newtworks, Springer Verlag, Berlin 1989, p.6. 2 Cohen, I. R. and Young, D. B., Immunol. Today 1991.12: 105. 3 Jindal, S., Dudani, A. K., Singh, B., Harley, C. B. and Gupta, R.S., Mol. Cell. Biol. 1989. 9: 2279.
1669
4 Young, D. B., Ivanyi, J., Cox, J. H. and Lamb, J. R., Immunol. Today 1987.8: 215. 5 Lindquist, S., Annu. Rev. Biochem. 1986. 55: 1151. 6 Koga, T., Wand-Wuertenberger. A , , DeBruyn, J . , Munk, M. E., Schoel, B. and Kaufmann, S. H. E., Science 1989.245: 1112. 7 Munk, M. E., Schoel, B., Modrow, S., Karr, R. W.,Young, R. A. and Kaufmann, S. H. E . , J. Immunol. 1989. 143: 2844. 8 Fischer, H. F!, Sharrock, C. E. M., Colston, M. J. and Panayi, G . S., Eur. .I. Immunol. 1991. 21: 2937. 9 Lamb, F. I . , Kingston, A. E., Estrada-Ci, I. and Colston, M. J., Infect. Immun. 1988. 56: 1237. 10 Fazekas de St. Groth, S., J. Immunol. Methods 1982. 49: R11. 11 Fisch, F!,Malkovsky, M., Kovats, S., Sturm, E., Braakman, E., Klein, B. S. ,Voss, S. D., Morrissey, L. W., deMars, R., Wclch, W. J., Bolhius, R. L. H. and Sondel, F! M., Science 1990. 250: 1269. 12 DeLibero, G ., Casorati, G., Giachino, C., Carbonara, C., Migone, N., Matzinger, P. and Lanzavecchia, A., J. Exp. Med. 1991. 173: 1311. 13 Kabelitz, D.. Bender, A., Prospero,T.,Wesselborg, S., Janssen, 0. and Pechhold, K . , J. Exp. Med. 1991. 173: 1331. 14 Parker, C. M., Groh,V., Band, H., Porcelli, S. A , , Morita, C., Fabbi, M., Glass, D., Strominger, J. L.. and Brenner, M. B., J. Exp. Med. 1990. 171: 1597.
