Immunosuppressive effects of human placental trophoblast interferon-β on lymphocytes in vitro

Placenta(1994), 15, 591-600

Immunosuppressive Effects of Human Placental Trophoblast Interferon-J3 on Lymphocytes In Vitro M. ZDRAVKOVIC a, G. ABOAGYEMATHIESEN a, V. ZACHAR ~, F. D. T O T H ~'b, A. M. D A L S G A R D ~, H. HAGER a & P. EBBESEN a'c a The Danish Cancer Sode01, Department of Virus and Cancer, Gustav Wiedsvej 10, DK-8OOOAarhus C. Denmark blnstitute of Microbiology, Medical UniversiOJ,H-4012 Debrecen, Hungary
SUMMARY

Human placental trophoblast cells produce predominantly interferon-beta-type (IFN-~) when stimulated with viral inducers. The aim of the present study was to determine the in vitro antiproliferative effect of the trophoblast interferon-~ (troIFN-~) on mitogen-stimulated and resting lymphocytes. The antiproliferative effect of the tro-IFN-~ was compared to human recombinant IFN-~. All activities of troIFN-~ and human recombinant IFN-~ ranging between l O-l O00 IU/ml showed suppression of proliferative responses on mitogen-stimulated and resting lymphocytes compared to cultures without IFN treatment. The inhibitory level of both tro-IFN-~ and recombinant IFN-~ was significantly higher on the stimulated than on the resting lymphocytes. Although there was a variation in the inhibition of lymphocyte proliferation by both IFNs with respect to time, there was no statistically significant difference in the antiproliferative effect of the IFNs on both resting and mitogenstimulated lymphocytes. Since IFNs are produced locally by the placenta during pregnancy, our data suggest that in addition to the antiviral activity, the human tro-IFN-~ may partidpate in the local control of the maternal immune response during pregnancy at the fetomaternal interface.

INTRODUCTION The human placenta serves as an immunological barrier between the maternal and fetal circulations, preventing the potentially destructive maternal immune response from damag0143-4004/94/060591 + 10 $08.00/0

9 1994 w. B. Saunders CompanyLtd

592

Placenta (1994), Vol. 15

ing the semi-allogeneic fetus. The manner in which the barrier functions has been the subject of current investigation. One hypothesis advanced to explain fetal survival is that local immunosuppression surrounding the placenta prevents the sensitization of the maternal immune system to paternal alloantigens and development of subsequent effector functions (Slapsys and Clark, 1983; Clark et al, 1984). Extracts and culture supernatants from hydatiform mole trophoblast, syncytiotrophoblast membranes and human choriocarcinoma cell lines have been demonstrated to inhibit lymphocyte proliferation and mixed lymphocyte reaction in vitro (Cowan et al, 1989; Thibault et al, 1990; Arkwright and Redmann, 1990; Krishnan et al, 1991). Furthermore, interferons (IFNs) have been reported to have antiproliferative effect on stimulated T- and B-lymphocytes (Blomgren, Strander & Cantell, 1974). We have recently reported that human placental trophoblast produce predominantly IFN-f3 when stimulated with viral (T6th et al, 1990a; Aboagye-Mathiesen et al, 1991) and non-viral inducers (AboagyeMathiesen et al, 1990; T6th et al, 1990b). Trophoblast inferons (tro-IFNs) have been purified and chemically characterized (Aboagye-Mathiesen et al, 1990, 1991). In this paper, we present the inhibition of resting and mitogen-stimulated T- and B-lymphocyte proliferation in vitro by human trophoblast IFN-I3 (tro-IFN-13). The possible significance of this property in vivo during human pregnancy is discussed.

MATERIALS AND METHODS

Preparation and mitogen stimulation oflymphocytes Human peripheral blood was obtained from Aarhus University Hospital (Denmark) and mononuclear lymphocytes were isolated by gradient centrifugation with FicoU-Paque (Pharmacia, Uppsala, Sweden) (WHO Workshop, 1974). The intermediate band containing the peripheral blood mononuclear cells (PBMC) was harvested and washed in phosphate buffer saline (PBS). The cells were either resuspended to a final concentration of 1 x 106 viable cells/ml in RPMI- 1640 supplemented with 10 per cent fetal calf serum (FCS) or used for further purification. B- and T-lymphocyte proliferation was obtained by stimulating PBMC with pokeweed mitogen (PWM) (5 p~g/ml) and phytohaemaglutinin (PHA) (1 p.g/ml), respectively. CD4 § and CD8 + lymphocytes were isolated from PBMC by adding mouse anti-human monoclonal antibodies against CD4 and CD8 (Becton-Dickinson) and afterwards coupling these with magnetic beads coated with sheep anti-mouse monoclonal antibodies (Dynal A/ S, Norway). The cells were positively isolated by adherence to a magnetic field. Unbound cells were removed by washing in PBS. CD4 § and CD8 § lymphocytes were incubated overnight in order to discharge the magnetic beads. The purity of the lymphocyte subpopulations was greater than 95 per cent as tested with FACStar Plus Flow Cytometer (BectonDickinson Immunocytometry Systems, Mountain View, CA) using FITC-conjugated monoclonal antibody to CD4 and CD8 (Becton-Dickinson). The cells were washed in RPMI-1640 and were resuspended to a final concentration of 1 x 106 viable cells/ml in RPMI-1640 supplemented with 10 per cent FCS (10 per cent RPMI) and 1 p~g/ml of PHA.

Production and purification of tro-IFN-f3 Tro-IFN-~3 was produced using human placental trophoblast cells infected with Sendai virus (T6th et al, 1991), and was purified by tandem high-performance dye-ligand affinity

Zdravkovic et ak Human Placental Trophoblast IFN-[3 on Lympho~ytes 1

593

2

9467-

43-

30- tro-IFN-

20.1 -

14.4 - I Figure 1. Silver-stained sodium dodecyl sulphate-polyacrylamide gel of tandem high-performance dye-ligand affinity and immunoaffinity-purified human tro-lFN-13. Lane 1 is standard molecular mass marker and lane 2 is tro-IFN-13.

(HP-DLAC) and immuno-affinity chromatography (Aboagye-Mathiesen et al, 1991, 1993a). In each experiment a different preparation of tro-IFN-13 was used. Human recombinant IFN-13 (Serotec, UK) used in the experiments had a specific activity of 3 x 106 IU/ ml. Effect o f tro-IFN-13 and recombinant IFN-13 on proliferation of lymphocyte cultures Resting lymphocytes, PHA-stimulated PBMC, CD4 § and CD8 § and PWM-stimulated PBMC at a concentration of 1 x 106 cells/ml were seeded in 96-well microtitre plates (NUNC-Denmark) with 50 x 103 cells/well. Triplicate wells were stimulated with various activities of tro-IFN-[3 or recombinant IFN-13 (0IU/ml, 10IU/ml, 100IU/ml and 1000IU/ml). Proliferation was measured by incorporation of [3H]thymidine (1 Ixg/ml, specific activity of 2 Ci/mmol, Amersham International). Briefly, cells were pulsed with [3H]thymidine for 4 h at 24, 48, 72, 96 and 128 h after seeding. Following incorporation of the radioactive precursor the cells were harvested and absorbed on filter discs. The filters were washed in 10 per cent trichloroacetic acid (TCA) (Sigma, USA) then 5 per cent TCA and finally dried in 99.9 per cent ethanol. The filters were transferred to plastic beta-vials (Sarstedt, Germany) and 5 ml of scintillation fluid (Filter-Count, Packard) was added to the vials. The radioactivity was assessed using a scintillation counter (LKB 1203, WaUac, Finland). The inhibition of proliferation in cells supplemented with IFN was calculated relative to cells not stimulated with IFN using the following equation: 1

ct/min cells without IFN - ct/min cells with IFN ct/min cells without IFN

Placenta (1994), Vol. 15

594 5O

r

40

~

30

r~

"~r

20

o 1

2

3 Time (days)

4

5

5O

@

~

40

30

~=o 20 ,~

10

1

3 '4 5 Time (days) Figure 2. Antiproliferative effect of tro-IFN-13 (a) and recombinant IFN-13 Co) on resting peripheral blood mononuclear cells. Values represent the mean of three assays + SEM (n = 3). ([::]), 10; ( i ) , 100; (O), 1000 IU/ml.

Effect o f tro-IFN-13 on the proliferation of non-lymphocyte cells

Human JAR cells (American Type Culture Collection HTB 144) were seeded out (50 x 103 cells/well) in 10 per cent RPMI-1640 with various activities of tro-IFN-13 (0, 10, 100 and 1000 IU/ml). Proliferation was determined according to the method described above, however pulsing with [3H]thymidine was performed after 24, 48 and 72 h of incubation. Cell viability In all experiments, the viability of the cells was determined using the trypan blue dyeexclusion test. Statistical analysis Analysis of variance was evaluated by the ANOVA tool in Microsoft-Exel (Microsoft Corp., USA). Triplicate determinations carried out within each experiment were averaged before analysis. The significance level (r was set at 0.05. Data are presented as the least square mean (averaged across experiments) + SEM.

Zdravkovic et al: Human Placental Trophoblast IFN-# on Lymphocytes

595

100 t a

o 80[F ~.~

60

~ o ~ 40 "~e~

20

2

3. Time (days)

100

o

8o

~

:~i 4~ ,.Q

.~

2o 0

i

2

3 4 5 Time (days) Figure3. Anfiproliferative effect oftro-IFN-13 (a) and recombinant IFN-13 (b) on phytohaemaglufinin-stimulated, T-lymphocytes. Values represent the mean of three assays + SEM (n = 3). (H), 10; (1), 100; ([]), 1000IU/ml.

RESULTS Figure 1 shows the silver-stained sodium dodecyl sulphate-polyacrylamide gel (SDSPAGE) of tandem HP-DLAC and immunoaffinity-purified tro-IFN-13 used in the experiments. The specific activity of the purified tro-IFN-J3 was 2.7 • 10 6 IU/mg of protein. The effect of tro-IFN-13 and recombinant IFN-J3 on resting PBMC is shown in Figure 2. The inhibition by both IFNs commenced on day 1. After further culture there was an increase in the antiproliferative effect oftro-IFN-J3 until day 5, where the effect decreased. In contrast to tro-IFN-J3 no clear dose-response relationship could be observed using recombinant IFN-13. The inhibitory level due to tro-IFN-J3 was not statistically different from that observed with recombinant IFN-J3 (P = 0.84). Furthermore, there was no difference in cell viability after 5 days of culture between IFN-treated and control cultureS. Figure 3 illustrates the inhibition by tro-IFN-J3 (a) and recombinant IFN-J3 (b) on the proliferation of PHA-activated PBMC. The inhibitory effect of both IFNs on activated Tlymphocytes increased with increasing IFN activity and time in culture. The highest level of

Pla*'enta (1994), VoL 15

596 100

.~

so

~ +

60

~

4o

o

~

20 0

I

t

i

i

~ Time (days)

4

5

5

4

s

100 9-~ ~u

80

~

40

9~

20

e

o

i

2

Time (days)

Figure 4. Antiproliferative effect of tro-IFN-~3 (a) and recombinant IFN-J3 (b) on phytohaemaglutinin-activated CD4+-lymphocytes. Values represent the mean of three assays + SEM (n = 3). (IN), 10; ( ' ) , 100; (D), 1000 I U / ml.

inhibition by tro-IFN-[3 was observed at day 4 and of recombinant IFN-I3 at day 5. Both IFNs inhibited the activated PBMC-lymphocyte proliferation in a dose-dependent manner, and furthermore the level of inhibition by both IFNs was significantly higher compared to the effect on resting PBMC (for both P < 0.001). However, no statistically significant difference in inhibition between tro-IFN-[3 and recombinant IFN-[3 (P = 0.15) was observed on the PHA-activated PBMC. The effect of tro-IFN-13 and recombinant IFN-I3 on the proliferation of T-lymphocyte subpopulations (CD4 + and CD8 +) was also studied (Figures 4 and 5). Both IFNs inhibited CD4 + and CD8 + lymphocyte proliferation, and the level of inhibition increased with increasing IFN activity between 10-1000IU/ml) and time in culture up to 4 days. The declining inhibition after day 4 was most pronounced for recombinant IFN-I3. There was no statistically significant difference between the inhibitory effect oftro-IFN-13 and recombinant IFN-I3 on the two T-lymphocyte subpopulations (CD4+: P = 0.58; CD8+: P = 0.13). The proliferation of PWM-stimulated B-lymphocytes was also suppressed by tro-IFN-[3 and recombinant IFN-[3 (Figure 6). The suppression by tro-IFN-[3 was dose-dependent

Zdravkovic et al: Human Placental Trophoblast IFN-f3 on Lymphooites

597

100 o

~

so

+ :~

60

i

~'~ 40 0 ,.Q

20 0 2

3

Time (days) 100 o) o

.=

80

60

40

~

2o | 0

i

1

2

3

4

5

Time (days) Figure 5. Antiproliferative effect of tro-IFN-13 (a) and recombinant IFN-13 (b) on phytohaemaglutinin-activated CD8+-lymphocytes. Values represent the mean of three assays _ SEM (n = 3). (D), 10; (11), 100; (D), 1000IU/ ml.

and started at day 2 and peaked at day 4. However, the inhibitory effect of recombinant IFN-13 started at day 1 and was withheld at day 5. There was no statistically significant difference in inhibitory effect of the IFNs on the B-lymphocyte proliferation (P = 0.45). In all the mitogen-stimulated lymphocyte populations, there was no difference in cell viability after 5 days of culture between the IFN-treated culture and the control. Finally tro-IFN-13 was able to inhibit the proliferation of non-lymphocyte cells in vitro. Tro-IFN-(3 inhibited the proliferation of trophoblast-derived malignant (choriocarcinoma) cell line JAR in a dose-dependent manner (Table 1). Furthermore there was no difference in cell viability after 72 h of culture in the IFN-treated and control cells.

DISCUSSION

In several species including ruminants and human, IFN production is known to be a characteristic of the placenta (Chard et al, 1986; Howatson et al, 1988; Aboagye-

Placenta (1994), Vot 15

598 100

80 O

eL 60

no

o

.~

.

40

9.0

1

2

3 Time (days)

4

5

100

80 e~ 60

40 ,.Q

0

I 1

I

2

' 3 Time (days)

4

Figure 6. Antiproliferative effect ofu'o-IFN-13 (a) and recombinant IFN-J3 (b) on pokeweed mitogen-activated Blymphocytes. Values represent the mean of three assays + SEM (n = 3). (l-I), 10; (11), 100; ([2), 1000 IU/ml.

Table 1. Antiproliferative effect oftro-IFN-J3 on choriocarcinoma cell line JAR Time of culture Tro-IFN-J3 activity (IU/ml) 0 10 100 1000

24 h 41392 35 472 37 498 23076

__ 3970 4- 13 100 _+ 1098 _+ 8150

48 h (9.6%) (37%) (2.9%) (35%)

17 243 16 999 14138 8394

4- 1188 + 2064 + 2384 + 1174

72 h (6.9%) (12%) (17%) (14%)

6232 5829 4308 2958

+ 633 _+ 586 _+ 372 + 193

(10%) (10%) (8.6%) (6.5%)

Values are ct/min 4- s.d. (n = 2). The values in parentheses are the coefficient of variation.

Mathiesen, 1993b). In non-pregnant women, no IFN can be detected in the blood in contrast to during pregnancy where the activity rises to between 1-10 IU/ml (Chard et al, 1986). We have earlier demonstrated that human placental trophoblast cultures stimulated with viruses produce high amounts of IFNs, and the predominant class of IFN produced is [3 (Aboagye-Mathiesen et al, 1992, 1993a). We have also suggested that the tro-IFNs

Zdravkovic et al: Human Placental Trophoblast IFN-B on Lymphocytes

599

produced in the fetoplacental unit may play a crucial role in placental growth, differentiation, protection and immunoregulation. The present results demonstrate the antiproliferative effect oftro-IFN-[3 on resting PBMC, mitogen-stimulated T-, B-, CD4 + and CD8 + lymphocytes and non-lymphocytic cells (notably choriocarcinoma). Stimulation of PBMC with PHA and PWM predominantly activates T- and B-lymphocytes respectively. However, there is a requirement of the presence of T-cells to optimize the stimulatory effect of PWM on B-lymphocytes (Hume and Weidemann, 1980). The inhibition of tro-IFN-13 and recombinant IFN-13 on resting lymphocytes was lower at IFN activities between 10-1000 IU/ml, compared with the inhibition of activated lymphocytes. Since T- and B-lymphocytes constitute the major effector mechanisms in the immunological rejection of an allograft (Ratner et al, 1991), the inhibition of both activated Tand B-lymphocyte proliferation by tro-IFN-[3, suggests that tro-IFN-[3 may regulate the maternal immune response against alloantigens presented by the placenta during pregnancy. Furthermore tro-IFN-13 could be involved in regulation of cytokines released by lymphocytes present in the endometrium during pregnancy. These cytokines may be involved in the maintenance of the placenta (Wegmann 1988). The overall inhibitory effect of tro-IFN-[3 on both resting and activated lymphocytes was similar to that of recombinant IFN-[3. No statistically significant difference in the inhibitory level could be observed both for unstimulated (P = 0.84) and pooled rnitogen-stimulated lymphocytes (P = 0.60). The aetiology of recurrent spontaneous abortion remains unexplained in 40-60 per cent of couples (Stray-Pedersen and Stray-Pedersen, 1984; Tho, Byrd and McDonough, 1979). It is speculated that impaired maternal tolerance to the semi-allogeneic fetus might be an explanation (Hill, 1990; Hill et al, 1992). This could be due to a decreased placental secretion of immunosuppressive factors among those tro-IFN. There is increasing evidence that immunosuppressive factors secreted by trophoblast of the pre-embryo may play a key role for successful implantation (Pinkas et al, 1992; Jones et al, 1992; Sheth et al, 1991). The tro-IFN-[3, based on the present results may be one of the immunosuppressive factors produced locally in the fetoplacental unit, and thus be of importance for successful implantation.

REFERENCES Aboagye-Mathiesen, G., T6th, F. D., Juhl, C., Norskov-Lauritsen, N., Petersen, P. M. & Ebbesen, P.

(1990) Purificationand initialcharacterizationof human placentaltrophoblastinterferoninducedby polyriboinosinic-polyribocytidylicacid.Journal of General Vfrolog?, 71, 3061-3066. Aboagye-Mathiesen, G., T6th, F. D., Juhl, C., Norskov-Lauritsen, N., Petersen, P. M., Zacher, V. & Ebbesen, P. (1991) Characterizationof Sendaivirus-inducedhumanplacentaltrophoblastinterferons.Journal of General Virology, 72, 1871-1876. Aboagye-Mathiesen,G., T6th, F. D., Dalsgaard, A. M., Petersen, P. M., Zacher, V. & Ebbesen, P. (1992) Isolation, purification and biochemical characterizationof human placental interferons by tandem highperformanceaffinitychromatography.PreparativeBiochemistry, 22, 105-121. Aboagye-Mathiesen,G., T6th, F. D., Petersen, P. M., Gildberg, A., Norskov-Lauritsen, N., Zachar, V. & Ebbesen, P. (1993a) Differentialinterferonproductionin humanfirstand third trophoblastculturesstimulated with viruses.Placenta, 14, 225-234. Aboagye-Mathiesen, G., T6th, F. D. Petersen, P. M., Hager, H., Norskov-Laurtisen, N., Zdravkovic,M. & Ebbesen, P. (1993b) Humantrophoblastinterferons:productionand biochemicalcharacterization.Trophoblast Research (in press). Arkwright, P. D. & Redmann, C. W. G. (1990) Syncytiotrophoblastmembraneglycoproteincomponentsblock lymphocyteproliferationby interferingwithan earlyeventin their activation.In Cellular and Molecular Biology of the Materno-Fetal Relationship (Eds) Chaouat,G. & Mowbray,J. pp. 227-236. Blomgren, H., Strander, H. & Cantell, K. (1974) Effectof human leukocyteinterferonon the response of lymphocytesto mitogenicstimuliin vitro. Scandinavian Journal of Immunology, 3, 697-705.

600

Placenta (1994), VoL 15

Chard, T., Craig, P. H., Menebawey, M. & Lee, C. (1986) Alpha interferon in human pregnancy. British Journal of Obstetrics and Gynaecology,93, 1145-1149. Clark, D. A., Slapsys, R. M., Croy, B. A. & Rossant J. (1984) Immunoregulation of graft-versus-host responses in the uterus. Immunology Today, 5, 111. Cowan, B. D., Bennett, W. A., Brackin, M. N. & McGehee, R. P. (1989) Suppression of lymphocyte proliferation in vitro by macromolecules in the vesicle fluid and tissue extracts of hydatiform mole. Journal of Reproduaive Immunology, 15, 39-49. Hill, J. A. (1990) Immunological mechanisms of pregnancy maintenance and failure: a critique of theories and therapy. AmericanJournal of Reproduaive Immunology, 22, 33-42. Hill, J. A., Polgar, K., Harlow, B. L. & Anderson, D.J. (1992) Evidence of embryo- and trophoblast-toxic cellular immune response(s) in women with recurrent spontaneous abortion. AmericanJournal of Obstetrics and Gynecology, 166, 1044-1052. Howatson, A. G., Farquharson, M., Maeger, A., McNicol, A. M. & Foulis, A. K. (1988) Localization of ainterferon in the human feto-ptacental unit.Journal of Endocrinology, 119, 531-534. Hume, D. A. & Weidemann, M.J. (1980) MitogenicLymphoo:te Transformation. Amsterdam: Elsevier. Jones, K. P., Warnock, S. H., Urry, R. L., Edwin, S. S. & Mitchell, M. D. (1992) Immunosuppressive activity and alpha interferon concentrations in human embryo culture media as an index of potential for successful implantation. Fertili~ and Sterili~, 57, 637-640. Krishnan, L., Menu, E., Chaouat, G., Talwar, G. P. & Raghupathy, R. (1991) In vitro and in vivo immunosuppressive effects of supernatants from human choriocarcinoma cell lines. CellularImmunology, 138, 313-325. Pinkas, H., Fisch, B., Tadir, Y., Ovadia, J., Amit, S. & Shohat, B. (1992) Immunosuppressive activity in culture media containing human oocytes fertilized in vitro. Archives ofAndrology, 28, 53-59. Rather, F. E., Hadley, G. A., Hanto, D. W. & Mahanakumar, T. (1991) Immunology of renal allograft rejection. Archives of Pathology and LaboratoryMedicine, 115, 283-287. Sheth, K. V., Roca, G. L., al-Sedairy, S. T., Parhar, R. S., Hamilton, C.J. & al-Abdul-Jabbar, F. (1991) Prediction of successful embryo implantation by measuring interleukin-l-alpha and immunosuppressive factor(s) in pre-implantation embryo culture fluid. Fertility and Sterility, 55,952-957. Slapsys, R. M. & Clark, D. A. (1983) Active suppression of host-versus-graft reaction in pregnant mice. V. Kinetics, specificity and in vivo activity of non-T suppressor cells localized to the genital track of mice during first pregnancy. AmericanJournal of Reproduaive Immunology, 3, 65. Stray-Pedersen, B. & Stray-Pedersen, S. (1984) Etiologic factors and subsequent reproductive performance in 195 couples with a prior history of habitual abortion. American Journal of Obstetrics and Gynecology, 148, 140146. Thibault, G., Girard, A. C., Degenne, D., GuHlaumin, J. M., Lacord, M., de Russe, J., Lanson, M. & Barrios, P. (1990) Inhibitory effect of human syncytiotrophoblast plasma membranes (STPM) on in vitro lymphocyte proliferation: mechanism of action and membrane components involved. In Cellular and Molecular Biology of the Materno-Fetal Relationship. (Eds.) Chaouat, G. & Mowbray, J. Tho, P. T., Byrd, J. R. & McDonough, P. G. (1979) Etiologic and subsequent reproductive reform of 100 couples with recent abortion. FertiliO:and Sterili~, 32, 389-395. Tfth, F. D., Norskov-Lauritsen, N.,Juhl, C. B., Aboagye-Mathiesen, G. & Ebbesen, P. (1990a) Interferon production by cultured human trophoblasts and choriocarcinoma cell lines induced by Sendal virus. Journal of General Virology, 71, 3067-3069. Ttth, F. D., Juhl, C., Norskov-Lauritsen, N., Mosborg-Petersen, P. & Ebbesen, P. (1990b) Interferon production by cultured human trophoblast induced with double stranded polyribonucleotide. Journal of Reproductive Immunology, 17, 217-227. Ttth, F. D., Norskov-Lauritsen, N., Juhl, C. & Ebbesen, P. (1991) Human trophoblast interferon: pattern of response to priming and superinduction of purified term trophoblast and choriocarcinoma cells. Journal of Reproduaive Immunology, 19, 55-67. Wegmann, T. G. (1988) Maternal T cells promote placental growth and prevent spontaneous abortion. Immunology Letters, 17, 297-303. Workshop, W. H. O. (1974). Identification, enumeration, and isolation of B- and T-lymphocytes from human peripheral blood. ScandinavianJournal oflmmunology, 3, 521.