Scand. J. Immttnol. 9, 151-158, 1979
Lymphoblastoid Cell Lines are Polyclonal Activators of Human T Lymphocytes A. W. BOYLSTON & R. L. ANDERSON Department of Pathology, St Mary's Hospital Medical School, London, UK
Doylston. A.W, A Anderson. R,L, Lymphoblastoid Cell Lines are Potyclonal Aciivaiors of Human T Lymphocytes, Scanil. J. Immunol. 9, 151-158. 1979, Following stimulalion wiih aLitologoiis or allogeneic lyniphohlastoid cell lines (LCD human T lymphocytes acquire two properties which suggest ihat LCL act as polyclonal activators, Cytoloxic aciivtty. which has an antigen-specific component, is produced towards normal lympliocytes and LCL, and the celts become capable ormounling proliferative responses to antigens on human B lymphocytes which have accelerated secondary-type kinetics. Only weak responses to autologous cells occur. In addition, repeated restimulaiion wiih the original LCL leads lo a progressive increase in the number of cells in ihe culture for a period ofahout 4 weeks, This approach may prove a useful way to grow large numbers of human T lymphocytes for further study. A. ir. BorhtDii, Dcptirftiu'iil oJPathohigy. St Man's Hospital Mcilical School. Lomloti W2 IPG, UK.
Human lymphoblastoid cell lines (LCL) induce strong proliferative and cytotoxic responses in human peripheral blood T lymphocytes [15]. Both autologous and allogeneic LCL are efTective [2, 6] and the response is not due to Epstein-Barr (EB) viral antigens present in the stimulating LCL or to serum components adsorbed onto its surface [8, 18]. There are, however, conflicting reports as to the antigen specificity of the responses developed. In sonic reports the cytotoxicity has largely been limited to the initiating LCL, while in others it has been diflicult to show differences between specific and nonspecific targets [7, 10, 13-15]. The proliferative response which follows restimulation has been stated to be specific for antigens on the original LCL, and not to occur when autologous combinations of stimulator and donor are cultured together [I, 14]. The experiments reported here arose out of attempts to induce specific responses to haptens coupled to autologous LCL. Not only was no haptenspeeific immtinity detectable, but the response to unmodified LCL was also found to be nonspecific. Our results suggest that LCL act as polyelonal activators of human peripheral blood T lymphocytes and that cells with cytotoxic potential and with the ability to proliferate 0300-9475/79/0200-0151 $02.00
when exposed to aniigens expressed on B lymphocytes are activated. In addition, we have observed that growth occurs in LCLstimulated cultures, leading to a large increase in total cell number. This approach may be useful as a method of growing large numbers of human T lymphocytes for further characterization, and as a way of assessing an individual's immunological repertoire.
MATERIALS AND METHODS Lymphoblastoid cL'll lines. Bri 7, Bri 8, 1788 and BEC II were purchased from Searle Laboratories, High Wycombe. UK. Line ABI arose spontaneously from a peripheral blood culture of a normal human donor. Their HLA and B antigens were determined by ibe tissue typing laboratory of St Mary's Hospital and round to be Bri 7 A2,9,BI2.35; Bri 8-A L2.BI3.W2I ; 1788-A2,WI9.B5.W22; ABI-A1.2.BI3,W'38, which is also the tissue type of the cell donor, and BECIl At.B17. These cell lines were maintained in continuous stationary suspension culture in RPMI-1640 medium supplemented with 20 mggeniamycin, 50 mgcloxacillin, and 2 g Na HCO, per litre, plus 10"^ heat-inaciivated fetal calf serum (FCS) in a humidified atmosphere of 5",, COj in air. Peripheral blood cultures. Mononuclear cells (PBL) were separated from defibrinated venous blood over
r 1979 Blackwell Scientific Publications
151
152
A. W. Boylston & R. L. Anderson
Ficoll-Hypaque (Nyegaard, Oslo, Norway). Lymphocytes were purified by filtration on nylon wool columns [91. Sheep red blood cells (SRBC) were ireated with 2-amino-elhylisoihiouraiiium bromide hydrobromide tAET) and AET-SRBC binding lymphocytes were prepared as described [11], For primary cultures I -• 10'"' PBL/ml were cullured with 0,5 •-10" LCL/ml In the medium described above, supplemented as described [5]. Cultures were resiiniulated afler 8 days by diluting an aliquot to 0,5>:IO^ cells/ml and adding 0,25 > 10" LCL cells/ml. After primary stimulalion, restimulation wiib the original LCL was performed every 4—5 days. When used as target cells, PBL were separated, prepared as above, and cullured in 1640 10",, FCS for 4-6 days before use. Umbilical cord biood lymphocytes were obtained from the umbilical vein immediately after delivery and processed as for peripheral blood mononuclear cells. They were used lo stimulate cultures 48-96 h later, following incubation in RPMI 1640-10",, FCS al 1 - 10*/ml during tbe interval. Proliferation assay. The proliferative response of LCL-stimulated PBL was assessed by culturing 1 • lO'"' responder cells witb 5x lO-" LCL or I -• 10* PBL in 0.2 ml supplemented RPMI-1640 coniaining IO% FCS in flat-bottomed microtitre plaies (Linbro 76-201-05), LCL were exposed to 10.000 R and PBL to 3000 R before use as stimulator cells. Responses were measured by adding I [xCi of ^H-TdR in IO|il of medium to each well on day 2 and harvesting the cells onto filter paper with a Skatron cell harvester 3-4 h later. ^H-TdR incorporation was measured in a scintillaiion counter, Cytotoxicity. The cytotoxic activity of LCL-stimulated PBL was assessed by the *'Cr release method, I X 10* labelled target cells were incubated wiih various numbersof killer cells in RPMI 1640-10" „ FCS for 4 h at 37 C, The supernatant was harvested and counted in a Packard gamma counter. Maximum '^'Cr release was determined by incubating an aliquot of target cells in RPMI 1640 containing T\ Triton X-100. Percentage of specific *'Cr release was calculated from the formula; % cyfoioxicity -counts released in sample -counts spontaneously released/maximum counts releasedspontaneous release >• 100. In experiments designed to investigate the ability of unlabelled cells to inhibil cytotoxicity, 1 x 10^ cells were added lo the labelled target cells before the cytotoxic cells were added.
RESULTS PBL from several donors were sensitized by three to five cycles of stimulation with an LCL, and their ability to kill different LCL determined. The results are shown in Table I, where it can be seen that such cultures kill all LCL tested to approximately the same extent within any one experiment. Stimulated cells kill targets which difTer by no, one, three or four HLA antigens equally well, and the autologous combination tested also kills all three targets,
In Iwo experiments the ability of PBL which form rosettes with AET-SRBC to express cytotoxicity was compared to the cytotoxieity produced in cultures of nylon wool filtered lytiiphocytes. The AET-SRBC binding fraction of nylon wool filtered PBL is somewhat more cytotoxic than the population filtered through nylon wool alone. To further characterize the interaction between LCL-stimulated PBL and target cells the effect of adding unlabelled LCL to various combinations of killer and target was studied. The results are shown in Table IL Cytotoxicity, whether generated by autologous or allogeneic stimulation, was inhibited by ail LCL tested regardless of whether the labelled target was the one used to initiate the culture or another line. Some specificity for the labelled target cell is present since in these experiments maximum inhibition of lysis is produced by blocking with the same cell line used as the labelled target. Thus, cytotoxicity shows no relation to the LCL used to stimulate the culture but appears to contain two elements, one which is target cell line specific and one which is shared by all the LCL tested. Since the cytotoxicity towards LCL contained an element of target specificity, we examined the efTect of stimulated cells on normal peripheral blood lymphocytes. The results of a typical experiment are shown in Table IIL The particular responder/stimuiator combination used, donor A anti-Bri 8, could contain activity directed towards the HLA-B locus antigens 13 and W2I. None of the PBL used as target cells in this experiment possess these antigens. The data show that cytotoxic activity towards all three allogeneic donors is present and that only weak, if any. autologous cytotoxicity can be detected. The data also indicate that cytotoxicity can be inhibited by unlabelled cells from all four donors but that it is maximally inhibited by cells from the labelled target donor. The specificity of proliferative responses to various LCL by primed PBL was studied and the results are shown in Table IV. All three LCL tested restimulated equally well, and there was no evidence of specific restimulation by the initiating cell line. Since LCL induce nonspecific responses in PBL, the efTect of restimulation with normal PBL was investigated. Data from typical experiments are shown in
Lymphoblastoid Cell Lines Activate Hunutn Lymphocytes
1 53
TABLF I. Cylolo\icity of lymphoblastoid cell lino (LCL)-siiiiuilatcO liiiman lymphocyies towards LCL target cells Culture, re^pondcr vs stiiiiulatiiig LCL
Killer/targel ratio
'„ specific lysis of labelled LCL* Alii
A^ABl
40 10
HLA differences BxABIS
40 10
HLA ilifTerentes D-/I788
HLA dilferences
21 12 (1
36 23 0
61 39 A2.BI2.W38
52 28 A2
92 65 A:
29 20 AW19,B7.I4
27 20 0
ND
30
28 19 A1.2.B12.W3R
43 33 Al
21 17 A2
25
Bri7
Bri8
1788
79 21 0 86 19
91 31 0 91
Bri7
BriS
BECII
47 20 0 53
373 9 0 51 "•O
41 15 B17 68 35
to 30 10
HLA diU'erences
A-/Bri7*
20
HLA ditVerence.'i AxBri7t
20
20 HLA dilTertnces AxBECHt
27 19 Autologous
30 10
HLA differences E:timulation of lymphoblastoid cell line (LCLl-primed huniiin Ivinphocytcs bv LCL is nonspecilic Responder
.Slinnilating LCL
Restimiilating LCL
A
ABI (aulologous)
ABi
Cpm HI-TilR incurpLirated' 1-4.331 13.44tt 22;f.245 16.141 107.212 • 16.923
1788 Hri8
66
211
1.056 3.931 7.246 80 ( < 1.9(X))
B
ABI
\BI 1788 DriS
52.075 55.984 49.720
D
178S
AH!
92.554 1 13.207 X2.71O 12.215 3,944 (
Lymphoblastoid Cell Lines are Polyclonal Activators of Human T Lymphocytes A. W. BOYLSTON & R. L. ANDERSON Department of Pathology, St Mary's Hospital Medical School, London, UK
Doylston. A.W, A Anderson. R,L, Lymphoblastoid Cell Lines are Potyclonal Aciivaiors of Human T Lymphocytes, Scanil. J. Immunol. 9, 151-158. 1979, Following stimulalion wiih aLitologoiis or allogeneic lyniphohlastoid cell lines (LCD human T lymphocytes acquire two properties which suggest ihat LCL act as polyclonal activators, Cytoloxic aciivtty. which has an antigen-specific component, is produced towards normal lympliocytes and LCL, and the celts become capable ormounling proliferative responses to antigens on human B lymphocytes which have accelerated secondary-type kinetics. Only weak responses to autologous cells occur. In addition, repeated restimulaiion wiih the original LCL leads lo a progressive increase in the number of cells in ihe culture for a period ofahout 4 weeks, This approach may prove a useful way to grow large numbers of human T lymphocytes for further study. A. ir. BorhtDii, Dcptirftiu'iil oJPathohigy. St Man's Hospital Mcilical School. Lomloti W2 IPG, UK.
Human lymphoblastoid cell lines (LCL) induce strong proliferative and cytotoxic responses in human peripheral blood T lymphocytes [15]. Both autologous and allogeneic LCL are efTective [2, 6] and the response is not due to Epstein-Barr (EB) viral antigens present in the stimulating LCL or to serum components adsorbed onto its surface [8, 18]. There are, however, conflicting reports as to the antigen specificity of the responses developed. In sonic reports the cytotoxicity has largely been limited to the initiating LCL, while in others it has been diflicult to show differences between specific and nonspecific targets [7, 10, 13-15]. The proliferative response which follows restimulation has been stated to be specific for antigens on the original LCL, and not to occur when autologous combinations of stimulator and donor are cultured together [I, 14]. The experiments reported here arose out of attempts to induce specific responses to haptens coupled to autologous LCL. Not only was no haptenspeeific immtinity detectable, but the response to unmodified LCL was also found to be nonspecific. Our results suggest that LCL act as polyelonal activators of human peripheral blood T lymphocytes and that cells with cytotoxic potential and with the ability to proliferate 0300-9475/79/0200-0151 $02.00
when exposed to aniigens expressed on B lymphocytes are activated. In addition, we have observed that growth occurs in LCLstimulated cultures, leading to a large increase in total cell number. This approach may be useful as a method of growing large numbers of human T lymphocytes for further characterization, and as a way of assessing an individual's immunological repertoire.
MATERIALS AND METHODS Lymphoblastoid cL'll lines. Bri 7, Bri 8, 1788 and BEC II were purchased from Searle Laboratories, High Wycombe. UK. Line ABI arose spontaneously from a peripheral blood culture of a normal human donor. Their HLA and B antigens were determined by ibe tissue typing laboratory of St Mary's Hospital and round to be Bri 7 A2,9,BI2.35; Bri 8-A L2.BI3.W2I ; 1788-A2,WI9.B5.W22; ABI-A1.2.BI3,W'38, which is also the tissue type of the cell donor, and BECIl At.B17. These cell lines were maintained in continuous stationary suspension culture in RPMI-1640 medium supplemented with 20 mggeniamycin, 50 mgcloxacillin, and 2 g Na HCO, per litre, plus 10"^ heat-inaciivated fetal calf serum (FCS) in a humidified atmosphere of 5",, COj in air. Peripheral blood cultures. Mononuclear cells (PBL) were separated from defibrinated venous blood over
r 1979 Blackwell Scientific Publications
151
152
A. W. Boylston & R. L. Anderson
Ficoll-Hypaque (Nyegaard, Oslo, Norway). Lymphocytes were purified by filtration on nylon wool columns [91. Sheep red blood cells (SRBC) were ireated with 2-amino-elhylisoihiouraiiium bromide hydrobromide tAET) and AET-SRBC binding lymphocytes were prepared as described [11], For primary cultures I -• 10'"' PBL/ml were cullured with 0,5 •-10" LCL/ml In the medium described above, supplemented as described [5]. Cultures were resiiniulated afler 8 days by diluting an aliquot to 0,5>:IO^ cells/ml and adding 0,25 > 10" LCL cells/ml. After primary stimulalion, restimulation wiib the original LCL was performed every 4—5 days. When used as target cells, PBL were separated, prepared as above, and cullured in 1640 10",, FCS for 4-6 days before use. Umbilical cord biood lymphocytes were obtained from the umbilical vein immediately after delivery and processed as for peripheral blood mononuclear cells. They were used lo stimulate cultures 48-96 h later, following incubation in RPMI 1640-10",, FCS al 1 - 10*/ml during tbe interval. Proliferation assay. The proliferative response of LCL-stimulated PBL was assessed by culturing 1 • lO'"' responder cells witb 5x lO-" LCL or I -• 10* PBL in 0.2 ml supplemented RPMI-1640 coniaining IO% FCS in flat-bottomed microtitre plaies (Linbro 76-201-05), LCL were exposed to 10.000 R and PBL to 3000 R before use as stimulator cells. Responses were measured by adding I [xCi of ^H-TdR in IO|il of medium to each well on day 2 and harvesting the cells onto filter paper with a Skatron cell harvester 3-4 h later. ^H-TdR incorporation was measured in a scintillaiion counter, Cytotoxicity. The cytotoxic activity of LCL-stimulated PBL was assessed by the *'Cr release method, I X 10* labelled target cells were incubated wiih various numbersof killer cells in RPMI 1640-10" „ FCS for 4 h at 37 C, The supernatant was harvested and counted in a Packard gamma counter. Maximum '^'Cr release was determined by incubating an aliquot of target cells in RPMI 1640 containing T\ Triton X-100. Percentage of specific *'Cr release was calculated from the formula; % cyfoioxicity -counts released in sample -counts spontaneously released/maximum counts releasedspontaneous release >• 100. In experiments designed to investigate the ability of unlabelled cells to inhibil cytotoxicity, 1 x 10^ cells were added lo the labelled target cells before the cytotoxic cells were added.
RESULTS PBL from several donors were sensitized by three to five cycles of stimulation with an LCL, and their ability to kill different LCL determined. The results are shown in Table I, where it can be seen that such cultures kill all LCL tested to approximately the same extent within any one experiment. Stimulated cells kill targets which difTer by no, one, three or four HLA antigens equally well, and the autologous combination tested also kills all three targets,
In Iwo experiments the ability of PBL which form rosettes with AET-SRBC to express cytotoxicity was compared to the cytotoxieity produced in cultures of nylon wool filtered lytiiphocytes. The AET-SRBC binding fraction of nylon wool filtered PBL is somewhat more cytotoxic than the population filtered through nylon wool alone. To further characterize the interaction between LCL-stimulated PBL and target cells the effect of adding unlabelled LCL to various combinations of killer and target was studied. The results are shown in Table IL Cytotoxicity, whether generated by autologous or allogeneic stimulation, was inhibited by ail LCL tested regardless of whether the labelled target was the one used to initiate the culture or another line. Some specificity for the labelled target cell is present since in these experiments maximum inhibition of lysis is produced by blocking with the same cell line used as the labelled target. Thus, cytotoxicity shows no relation to the LCL used to stimulate the culture but appears to contain two elements, one which is target cell line specific and one which is shared by all the LCL tested. Since the cytotoxicity towards LCL contained an element of target specificity, we examined the efTect of stimulated cells on normal peripheral blood lymphocytes. The results of a typical experiment are shown in Table IIL The particular responder/stimuiator combination used, donor A anti-Bri 8, could contain activity directed towards the HLA-B locus antigens 13 and W2I. None of the PBL used as target cells in this experiment possess these antigens. The data show that cytotoxic activity towards all three allogeneic donors is present and that only weak, if any. autologous cytotoxicity can be detected. The data also indicate that cytotoxicity can be inhibited by unlabelled cells from all four donors but that it is maximally inhibited by cells from the labelled target donor. The specificity of proliferative responses to various LCL by primed PBL was studied and the results are shown in Table IV. All three LCL tested restimulated equally well, and there was no evidence of specific restimulation by the initiating cell line. Since LCL induce nonspecific responses in PBL, the efTect of restimulation with normal PBL was investigated. Data from typical experiments are shown in
Lymphoblastoid Cell Lines Activate Hunutn Lymphocytes
1 53
TABLF I. Cylolo\icity of lymphoblastoid cell lino (LCL)-siiiiuilatcO liiiman lymphocyies towards LCL target cells Culture, re^pondcr vs stiiiiulatiiig LCL
Killer/targel ratio
'„ specific lysis of labelled LCL* Alii
A^ABl
40 10
HLA differences BxABIS
40 10
HLA ilifTerentes D-/I788
HLA dilferences
21 12 (1
36 23 0
61 39 A2.BI2.W38
52 28 A2
92 65 A:
29 20 AW19,B7.I4
27 20 0
ND
30
28 19 A1.2.B12.W3R
43 33 Al
21 17 A2
25
Bri7
Bri8
1788
79 21 0 86 19
91 31 0 91
Bri7
BriS
BECII
47 20 0 53
373 9 0 51 "•O
41 15 B17 68 35
to 30 10
HLA diU'erences
A-/Bri7*
20
HLA ditVerence.'i AxBri7t
20
20 HLA dilTertnces AxBECHt
27 19 Autologous
30 10
HLA differences E:timulation of lymphoblastoid cell line (LCLl-primed huniiin Ivinphocytcs bv LCL is nonspecilic Responder
.Slinnilating LCL
Restimiilating LCL
A
ABI (aulologous)
ABi
Cpm HI-TilR incurpLirated' 1-4.331 13.44tt 22;f.245 16.141 107.212 • 16.923
1788 Hri8
66
211
1.056 3.931 7.246 80 ( < 1.9(X))
B
ABI
\BI 1788 DriS
52.075 55.984 49.720
D
178S
AH!
92.554 1 13.207 X2.71O 12.215 3,944 (
