CELLULAR IMMUNOLOGY 107,28 I-292 (1987)

Epstein-Barr Virus-Transformed B Cells Process and Present Mycobacterium tuberculosis Particulate Antigens to T-Cell Clones

GIOVANNALOMBARDI,*FLAVIADELGALLO,*DANIELAVISMARA,* ENZAPICCOLELLA,*C.DEMARTINO,-~C.GARZELLI,$

CARLA FUGLISI,#I AND V. COLIZZI$

*Department of Cellular and Developmental Biology, University of Rome, and tlnstitute “Regina Elena, ” Rome, and z#Institute of Microbialagy, University of Pisa, Piss, Italy

ReceivedAugust 251986; acceptedDecember IS.1986

We have analyzed the presentation of mycobacterial antigens by Epstein-Barr virus-trans- formed human B (EBV-B) cells to mycobacteria-specific T-cell clones and lines, and to purified resting T cells. EBV-B cells were able to process and present not only soluble forms of antigen, such as PPD and the expressate preparation ofM. tuberculosis strain H37Rv, but also particulate forms of antigen, such as whole mycobacterial H37Rv or M. bovis organisms. Electron micros- copy studies demonstrated the capacity of EBV-B cells to phagocytose mycobacterial cells in 18 hr and pulsing experiments confirmed that an 18-hr of incubation is required for an efficient processing and presentation of mycobacterial determinants to T cells. The processing of whole- H37Rv particulate antigen by EBV-B cells was inhibited by the lysosomotrophic compound chloroquine and by high doses of irradiation. Finally, the analysis of the presentation of soluble and particulate mycobacterial antigens by PPD-positive and PPD-negative EBV-B cell clones has shown a preferential presentation of both forms of antigen by PPD-positive EBV-B clones. 0 1987 Academic Press, Inc.

INTRODUCTION

It is well-known that intracellular pathogens, such as Mycobacterium, Listeria, and various protozoans, are able to evade and resist digestion within tissue macrophages. Nevertheless, macrophages present bacterial epitopes to specific T cells, resulting in the induction of the immune response with consequent macrophage hyperactivation and mycobacterial killing. In the last few years, several laboratories have reported that not only macrophages but also some cells classically defined as nonphagocytic, such as dendritic (1, 2), epidermal (3), endothelial (4), and B (4-7) cells, appear to efficiently present antigens to T cells. The several steps required for antigen presenta- tion including antigen uptake, internalization, degradation of the native antigen, and reexpression of epitopes together with class II determinants have been investi- gated (8-10).

Most of the studies on the antigen-presenting function of cells other than macro- phages have been performed using soluble forms of antigen. However, Malynn et al. (7) recently showed the presentation of sheep red blood cells by murine B lympho- cytes. However, the authors could not exclude the possibility that lysis of sheep red cells in the culture leads to the release of soluble antigens and that macrophages

281

0008-8749187 $3.00 Copyright Q 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

282 LOMBARD1 ET AL.

within the population process antigen, which is then taken up and presented by B cells. Using whole Mycobacterium tuberculosis organisms, Kaye et al. (2) showed the ability of mouse dendritic cells to efficiently process and present the particulate form of antigen to highly pm&d T cells. On the human side, Epstein-Barr virus (EBV)’ transformed B-cell lines have been shown to present M. leprae antigens to M. leprae- specific T-cell clones ( 11) and to have used as antigen-presenting cells ( 12, 13). How- ever, in such studies the authors do not investigate the nature of the antigen, as the question of the possible contamination of the bacilli preparation by soluble antigen was not formally addressed.

The question on the capacity of B cells to present particulate M. tuberculosis anti- gens has been approached in our laboratories using EBV-transformed B-cell clones and evaluating their ability to activate both T-cell clones and resting T cells. Func- tional activation of T cells by EBV-transformed B (EBV-B) cells and electron micros- copy studies have clearly demonstrated the capacity of B cells to internalize and to process the whole mycobacterial organisms, leading us to reconsider the role of B lymphocytes in antituberculous immunity.

MATERIALS AND METHODS

Microbial antigens. Purified protein derivative of tuberculin (PPD) from M. tuber- culosis (Statens Seruminstitut, Copenhagen, Denmark) and the expressate of M. tu- berculosis strain H37Rv (soluble-H37Rv), kindly provided by Dr. J. Ivanyi, were used as soluble forms of antigen. Nonviable M. tuberculosis H37Rv (wholsH37Rv) obtained from Difco and nonviable M. bovis, strain BCG (whole-BCG) obtained from Glaxo were washed three times by centrifugation and resuspended in culture medium and used as particulate forms of antigen. The concentration of all antigen preparations was previously determined and the dose of 10 &ml of all antigenic preparations was used throughout this study, unless otherwise stated.

Preparation of EBV-B lymphocytes. Peripheral blood mononuclear cells (PBMC) from healthy individuals with tuberculin-positive skin tests were isolated on a gradi- ent of Lymphoprep (Nyeegaard and Co. A/S, Oslo, Norway). T- and B-cell-enriched fractions were isolated from PBMC by E rosetting as previously described (8). The cells were suspended at 106/ml in RPM1 1640 medium (Flow Laboratories, Inc., McLean, VA) supplemented with 10% autologous heat-inactivated serum, 2 mM glutamine (Eurobio, Paris), and 40 &ml Gentalyn (Shering Corporation, Bloom- field, NJ) (culture medium). The E- cells were transformed with EBV obtained horn the marmoset lymphoblastoid line B95-8 as previously described (14). To detect EBV-B lines producing anti-PPD antibodies, an enzyme-linked immunosorbent as- say (ELISA) was used. Both nonproducing and producing anti-PPD antibody lines were cloned at least twice by limiting dilution (14). Both PPD+ and PPD- EBV-B clones expressed surface Ig receptors as shown by immunofluorescence.

Pulsing of EBV-B cells with mycobacterial antigens and treatment with chloro- quine. EBV-B cells (5 X lo5 cells/ml) were cultured in RPM1 1640 with 10% heat- inactivated fetal calf serum (FCS) in the presence of PPD ( 100 &ml) or with whole-

I Abbreviations used: EBV, Epstein-Barr virus; PBMC, peripheral blood mononuclear cells; PPD, puri- fied protein derivative; whole-H37Rv, intact Mycobucrerium tuberculosis cells, strain H37Rv; whole-BCG, intact Mycobacterium bovis cells, strain BCG, soluble-H37Rv, expressate of A4. tuberculosis, strain H37Rv; IL2, interleukin 2.

PROCESSING AND PRESENTATION OF MYCOBACT’ERIAL ANTIGENS 283

H37Rv ( 100 &ml). After 18 hr, antigen-pulsed cells were washed three times, resus- pended in complete medium, irradiated with 7500 rad, and then added to T-cell cultures. Chloroquine (Sigma) was prepared fresh in RPM1 1640 and the treatment was performed using two different protocols. ChIoroquine ( low4 or 10m5 M) was added to EBV-B cells (5 X 1 OS/ml) during the pulsing period of 18 hr in the presence of either PPD or whole-H37Rv. At the end of the pulsing time, EBV-B cells were washed three times, irradiated, and then added to cultures.

Electron microscopy analysis. EBV-B cells were pulsed with wholsH37Rv and after different time were harvested, washed three times, and then fixed in picric acid- formaldehyde solution with 2.5 mM glutaraldehyde acid in Sorensen buffer at pH 7.2-7.4 as previously reported (15). The postfixation was performed with unbuffered 1% 0~0~. After alcohol dehydration, the cells were centrifugated at 200g and embed- ded in Epon 8 12 (Serva). For electron microscopy, thin sections were stained with a saturated water solution of uranyl acetate and lead citrate and examinated under a Siemens Elmiskop 10 1 electron microscope.

Preparation of T-cell lines and clones. PBMC from tuberculin-positive skin test donors were cultured in 24-well microtiter plates for 7 days in the presence of PPD. The cells were then harvested and passed through a gradient of Lymphoprep, and the number of lymphoblasts was determined. The cells were either expanded in the presence of irradiated (3000 rad) autologous PBMC, recombinant interleukin 2 (rIL-2) (20 units/ml, Biogen SA, Geneve), and PPD and used as lines, or cloned by limiting dilution in Terasaky plates (0.3 lymphoblasts per well) in the presence of irradiated autologous PBMC, lectin-free conventional IL-2 (Lymphocult-T-LF, Bio- test), or Biogen rIG2 and PPD. After 7 days, the growing clones were transferred first to 96-well microtiter plates and then to 24-well plates together with irradiated PBMC or EBV-B cells, IL-2, and PPD (16). T-cell lines and clones were screened for antigen specificity by a proliferation assay.

Proliferation assay. T-cell lines and clones were cultured ( lo4 cells/well) in tripli- cate in flat-bottom microtiter plates (No. 3040, Falcon) in the presence of irradiated, antigen-pulsed PBMC ( 10’ cells/well), or EBV-B cells ( 1 O4 cells/well) as antigen pre- senting cells. The plates were incubated for 4 days and [3H]thymidine ([3H]TdR, Amersham, England, sp act 5 Ci/mmol) was added to each well (0.5 PCi) 12 hr before harvesting. In some experiments, purified resting T cells ( lo5 E+ cells/well) were cul- tured for 6 days with pulsed EBV-B cells. The radioactivity was evaluated in a beta counter and expressed as counts per minute. Background proliferation values (no antigen) were subtracted.

Slope analysis of logarithmic plots. To determine the unique role of EBV-B cells as antigen-presenting cells in the experiment in which the purified resting T-cell pop- ulation was used (Fig. l), the log of the proliferation (cpm) was plotted against the log of the number of added EBV-B cells. The slope of the resulting straight line was calculated by the method of least squares (17).

RESULTS

Presentation of Mycobacterial Antigens by EB V-B cells

We have investigated the capacity of EBV-B cells to process and present different mycobacterial antigen preparations to mycobacterial-specific T-cell lines and clones and purified human resting T cells. In particular, two forms of antigen were used,

284 LOMBARD1 ET AL.

11 4 IO; EBV3B cells

5 678910

PIG. 1. Activation of purified resting T cells by EBV-B cells. E+ fraction (5 X lo4 T cells) of PBMC Born a PPD-positive skin test individual were cultured for 6 days with PPD (0) or whole-H37Rv (Cl) in the presence of different numbers of irradiated autologous EBV-B cells. The radioactive thymidine was added 20 hr before harvesting. The log of the response (cpm incorporated) was plotted against the log of the number of EBV-B cells added. The slopes of the resulting straight lines were 1 .O and 1.1 for respectively PPD- and whole-H37Rv-stimulated cultures.

namely, particulate and soluble. The particulate forms of antigen consisted of the whole M. tuberculosis (whole-H37Rv) and M. bovis strain BCG (whole-BCG) organ- isms, whereas soluble antigens consisted of PPD and of an expressate preparation containing soluble mycobacterial antigens from M. tuberculosis (soluble-H37Rv). Table 1 shows that EBV-B cells present not only the soluble but also the particulate forms of mycobacterial antigen when added to T-cell cultures, although less effi- ciently than PBMC. This antigen presentation is genetically restricted as shown by the failure of allogeneic EBV-B cells to activate mycobacterial-specific T-cell clones (data not reported).

TABLE 1

Presentation of Soluble and Particulate Mycobactetial Antigens by PBMC and EBV-B Cells

T cells” PPD Soluble-H37Rv Whole-H37Rv Whole-BCG

Autologous PBMC EW4/7 EW4/10 LLl

Autologous EBV-B cells DGAl LLl LB1

16,107b 57,024 35,656 14,522 5,647 4,373 4,769 1,264

16,212 ND 12,892 ND

12,541 10,944 1,475 ND 6,382 ND 12,648 ND 8,399 ND 5,077 3,497

’ PPD-specific T-cell lines and clones ( lo4 cells/well) were cultured for 4 days with either irradiated (3000 tad) autologous PBMC ( 1 O5 cells/well) or EBV-B cells ( lo4 cells/well). The antigen preparations were added at the beginning of the cultures, while radioactive thymidine (0.5 &/well) was added 12 hr before the end of cultures.

b The results are expressed as counts per minute and the standard deviation was always less than 10%. Background proliferation (no antigen) was subtracted: the net value in the presence of PBMC was always less than 200 cpm/culture, whereas in the presence of EBV-B cells it was between 500 and 2000 cpm/ culture. Bach T-cell clone or line has been tested at least three times with high reproducibility.

PROCESSING AND PRESENTATION OF MYCOBACTERIAL ANTIGENS 285

TABLE 2

Time of Pulsing Required for Antigen Presentation by EBV-B Cells

Antigen T resting’

2 hr

LB1

Pulsing of EBV-B cells’

LLl T resting

18hr

LB1 LLl

PPD 64,763 3881 4579 99,687 10,663 9,103 VW (36) (50)

Whole-H37Rv 28,280 4849 72,623 8,725 15,369 (39) (32)

’ EBV-B cells (5 X 10’ cells/ml) were incubated for respectively 2 and 18 hr with 100 &ml of PPD or whole-H37Rv. After pulsing, the EBV-B cells were washed three times, irradiated with 7500 rad, and added ( lo4 cells/well) to cultures consisting of either purified resting T cells (5 X lo4 cells/well) or T-cell lines ( lo4 cells/well). Radioactive thymidine was added for 12 hr after 3 days of culture. The results are expressed as cpm. A representative experiment of three replications is reported.

b The purity of the resting T-cell population is supported by their low thymidine incorporation in the absence of EBV-B cells (2384 cpm).

’ The percentage of antigen presentation by EBV-B cells after a 2-hr pulsing is shown in parentheses, considering 100% the presentation obtained after an 18-hr pulsing.

The presentation of mycobacterial determinants by EBV-B cells was also analyzed in pulsing experiments. EBV-B cells were pulsed either for 2 or 18 hr with PPD, soluble-H37Rv, whole-H37Rv, and whole-BCG. At the end of the pulsing procedure, EBV-B cells were washed, irradiated, and then added to cell cultures containing either purified resting T cells or T-cell lines. Table 2 shows that a 2-hr pulsing of EBV-B cells with both forms of antigen, although sufficient to induce proliferation of resting T cells and T-cell lines, is less efficient than an 18-hr pulsing. The percentage of anti- gen presentation after a 2-hr pulsing with PPD was always greater than that observed with whole-H37Rv. This is particularly evident considering the presentation of whole-H37Rv to the line LB1 which failed to be activated by EBV-B cells pulsed 2 hr only. The direct relationship between the number of EBV-B cells and the degree of T-cell proliferation was analyzed in Fig. 1 using a fixed number of resting T cells as the responder population. The presence of only one interacting cell population which is limiting in the culture, i.e., the functional absence of macrophages in the T- cell population, is shown by plotting the log cell number versus log response (mean slope value with PPD and whole-H37Rv is respectively 1 .O and 1.1).

The physical integrity of the particulate preparations used was then investigated to exclude the possibility that soluble antigens derived from the bacterial cell prepara- tion were present and were responsible for T-cell activation. The following protocol was used. Washed M. tuberculosis organisms were incubated at 1 mg/ml in culture medium for 48 hr at 37°C. At the end of incubation, mycobacterial organisms were spun down. The supernatant was removed, concentrated 10 times by negative pres- sure, sterilely filtered, and then tested for associated antigenicity in a T-cell prolifera- tion assay using PBMC from a tuberculin-positive skin test donor. Figure 2 shows that the whole-H37Rv cells induced a strong proliferative response whereas only a slight, statistically insignificant, degree of proliferation was observed when lo-fold concentrated supernatant was used as antigen.

286 LOMBARD1 ET AL.

10x lx whole -

H37Rv SUP

FIG. 2. Lack of significant soluble antigen release from bacterial preparations. PBMC ( lo5 cells/well) from a healthy PPD-positive skin test individual were cultured for 6 days with whole-H37Rv preparation or with the supernatants of the particulate preparation concentrated 10 times or unconcentrated. The radioactive thymidine was added 20 hr before harvesting.

Evidence for Internalization and Degradation of Whole-H37Rv Mycobacteria by EBV-B Cells

The capacity of EBV-B cells to phagocytose whole-H37Rv organisms was analyzed by electron microscopy after different times of antigen pulsing. After 2 hr of incuba- tion, several EBV-B cells showed cytoplasmic elongated projections within clusters of mycobacteria organisms (Fig. 3a). After 18 hr of incubation, l-3% of EBV-B cells are seen to contain numerous mycobacteria organisms which are engulfed in enlarged phagosomelike organelles (Fig. 3b). The mycobacterial organisms appeared morpho- logically intact with the characteristic double membrane (Fig. 3~).

The requirement of the processing step for antigen presentation of the whole- H37Rv cells has been analyzed by using chloroquine, an agent which interferes with

TABLE 3

Effect of Chloroquine Treatment on the Antigen Presentation by EBV-B Cells

Treatment of EBV-B cells” I

PPD

II III

Whole-H37Rv

I II III

None 55156 51,042 11,384 8174 18,961 ND Chloroquine 6252 41,753 1,081 1282 3,918 ND

(0) (If-9 (90) (84) (79)

’ Chloroquine ( 1 Om5 M in Experiments I and II, and lO+ M in Experiment III) was added to EBV-B cells (5 X 105/ml) during the pulsing period of 18 hr in the presence of either PPD or whole-H37Rv. At the end of pulsing, EBV-B cells were washed three times, irradiated using a low dose (3000 rad), and then added to cultures. In Experiments I and III, the responder cell population consisted of T-cell lines, whereas in the Experiment II purified resting T cells were used. All cultures were harvested after 4 days and radioactive thymidine was added in the last 12 hr of culture.

’ The results are expressed as cpm. The percentage of inhibition of antigen presentation by chloroquine treatment is shown in parentheses.

PROCESSING AND PRESENTATION OF MYCOBACTERIAL ANTIGENS 287

normal lysosome functions (18, 19). Table 3 reports three experiments showing that the treatment of EBV-B cells with chloroquine inhibits the activation of both T-cell lines (Experiment I) and purified resting T celIs (Experiment III) when EBV-B cells were pulsed with whole-H37Rv. The treatment of EBV-B cells with 10m4 M chloro- quine interferes with PPD presentation in Experiment III but did not inhibit PPD- driven T-cell activation in Experiments I and II when chloroquine was used at lop5 M. The possibility that monocytes contaminating the purified T-cell population are responsible for whole-H37Rv presentation was excluded by plotting analysis, as re- ported above, and showed that T cells were not activated by whole bacteria in the absence of EBV-B cells (data not shown).

Comparison of the Presenting Capacity of PPD-Specijk and NonspeciJic EB V-B Cell Clones

PPD-specific and nonspecific EBV-B cell clones were selected by their capacity to release anti-PPD antibodies ( 14) and express Ig surface molecules (unpublished re- sults). Figure 4 shows the difference between PPD-specific and nonspecific EBV-B cell clones in the presentation of soluble (A and B) and particulate (C) antigens to T- cell clones in 1%hr pulsing experiments. As can be seen, the level of T-cell prolifera- tion induced by PPD-specific EBV-B cells is higher than that observed with nonspe- cific EBV-B cells at all ratios used and both with soluble and particulate forms of antigen. In contrast, Figs. 4d and 4e show no differences in T-cell activation between PPD-specific and nonspecific EBV-B cells when EBV-B cells were heavily irradiated before the addition of antigen in culture.

The effect of irradiation on the capacity of EBV-B cells to process and then present antigens was then analyzed. Table 4 shows that irradiation (7500 rad) before the addi- tion of antigen diminishes, when compared to the effect of radiation after an 18-hr pulsing, the EBV-B cell capacity of presenting both soluble and particulate forms of antigen.

DISCUSSION

It has been recently reported that cells classically defined as nonphagocytic might be able to process particulate forms of antigen and to present relevant determinants to antigen-committed T cells. In a murine model, Kaye et al. (2) showed that whole M tuberculosis organisms can be phagocytosed and processed to an immunogenic form by nonphagocytic dendritic cells which then become able to activate T cells. Furthermore, Malynn et al. (7) using murine B cells showed a genetically restricted presentation of sheep red blood cells to T-cell lines. These two papers have given rise to the question whether nonphagocytic cells are able to process particulate forms of antigen and to present relevant epitopes to T cells. Whole M leprae bacilli have been reported to be presented by EBV-B cells to specific T-cell clones, but the questions of the nature of antigen and of the contamination of the bacilli preparation by soluble antigens were not addressed ( 1 1 - 13).

The present paper provides that (i) EBV-B cells are able to process whole-H37Rv microorganisms and to present mycobacterial determinants to T cells, (ii) whole- H37Rv can be seen inside phagosomelike organelles after 18 hr of incubation, (iii) chloroquine treatment inhibits the processing and the presentation of whole-MTB,

288 LOMBARD1 ET AL.

FIG 3. Electron micrograph of EBV-B cells after different times ofwhokH37Rv incubation. (a) Several mycobacterial cells can be Seen outside the cells which often show pseudopodia (arrows) after 2 hr of incubation; (b) several mycobacterial cells are visible inside membrane-bound intracytoplasmic vacuole (V) after 18 hr of incubation; (c) magnification of two mycobacterial cells with a visible double membrane. N, nucleus; M, mitochondtia. Bars = 1 pm.

PROCESSING AND PRESENTATION OF MYCOBACTERIAL ANTIGENS 289

FIG. 3-Continued.

and (iv) PPD-specific B-cell clones preferentially present both forms of the mycobac- terial antigen to T cells when compared to PPD nonspecific B cell clones.

EBV-B cells have been shown to possess lysosomelike structures but with a lower frequency than monocytes (20). However, the capacity of EBV-B cells to phagocytose intact M. tuberculosis organisms has clearly shown by electron microscopy studies where mycobacterial organisms can be seen in close relationship to the EBV-B plasma membrane at the beginning of incubation, and, after 18 hr of pulsing, M. tuberculosis organisms were present inside enlarged phagosomelike organelles (Figs. 3b and 3~). The role of lysosomal enzymes in the processing of intact mycobacterial organisms is clearly indicated in our model by the inhibition of the whole-H37Rv antigen presentation after chloroquine treatment of EBV-B cells (Table 3). The pre- cise mechanisms by which cloroquine interferes with antigen degradation is unclear, although it is thought that this compound exerts its effect by inhibiting the fusion of phagosomes and lysosomes (18, 2 1) and by depressing acid hydrolase activity (19). Both these mechanisms may be involved in the inhibition of processing of the whole- H37Rv mycobacterial organisms. A quantitative difference in the sensitivity to chlor- oquine treatment between soluble (PPD) and particulate (whole-H37Rv) forms of mycobacterial antigen was observed when chloroquine at 1 O-’ Mwas used. However, an increase of chloroquine molarity completely abrogated PPD presentation. To- gether these results suggest that particulate antigens require more complex metabolic events than soluble forms of antigen.

Although intact mycobacteria organisms are phagocytosed by EBV-B cells, it might be possible that the T-cell activation is carried out by soluble forms of antigen con- taminating the whole-H37Rv antigen preparation. However, we are against this view for two lines of evidence. First, the IO-fold-concentrated supernatant from the whole- H37Rv preparation fails to stimulate T cells (Fig. 2). Second, both the different sensi- tivity to chloroquine treatment and the different requirement in the time of pulsing between PPD and whole-H37Rv antigen preparations in T-cell activation support the view that the particulate antigen preparation is not contaminated by soluble mol- ecules.

Both resting and activated T cells can be activated by the whole-H37Rv-pulsed EBV-B cells. This result seems to be in contrast to the results reported by Chu et al.

290 LOMBARD1 ET AL.

ooeeo of PPU (rgha

FIG. 4. Presentation of soluble and particulate forms of mycobacterial antigen by PPI&pecific and nonspecific EBV-B cell clones. PPD-specific (0) and nonspecific @) EBV-B cells were pulsed for 18 hr in the presence of PPD (a and b) or wholeH37Rv (c) antigens. After pulsing, the cells were washed, irradiated (7500 rad), and added at different numbers to T-cell clone DGAl (a and c) and clone DGA6 (b). Two different experiments (d and e) the illustrate presentation of PPD by PPD-specific and nonspecific EBV-B cells irradiated (7500 rad) in the absence of antigen and then added to T-cell (clone DGA 1) cultures con- taining different doses of PPD. Cells were cultured for 4 days and radioactive thymidine was added 20 hr before harvesting.

(5) showing the inability of EBV-B cells to trigger tetanus toxoid-specific proliferation in highly purified resting T-cell cultures. However, the inability of EBV-B cells has been reported to be caused not by defective antigen processing but by the failure of IL1 secretion. In our system, the activation of resting T cells by EBV-B cells might be due to monocyte contamination and hence to the production of monocytederived IL-l. We are against this interpretation for two reasons. First, the analysis of the number of interacting cells in the cultures containing EBV-B and resting T cells by plotting the log cell number versus log response suggests the presence of only one interacting cell which is limiting in the culture (mean slope value = 1 .O for PPD and 1.1 for whole-H37Rv). Second, the EBV-B cell preparation we used is fully able to release IL l-like factor in the culture supematant (Garzelli et al., unpublished results) as reported with other EBV-B lines (22,23).

The mechanism by which B cells take up antigens and present the resulting fmg- ments to T cells has only recently been addressed experimentally. Rock et al. (24) in the murine system and Lanzavecchia in the human model (25) have recently given clear evidence that the Ig receptor serves to concentrate antigen for subsequent inter- nalization, degradation, and presentation, but in both studies only soluble forms of antigen were tested. Since particulate antigen may have different requirements for presentation as compared to soluble antigens, Malynn et al. (7) analyzed the presenta- tion of particulate (sheep red blood cells) and soluble (L-glutamic acids-alaninez-

PROCESSING AND PRESENTATION OF MYCOBACTERIAL ANTIGENS 291

TABLE 4

Effect of the Time of Irradiation on Antigen Presentation by EBV-B Cells

Irradiation of EBV-B cells”

Before the addition of antigen After 1%hr pulsing

PPD Whole-H37Rv PPD Whole-H37Rv

DGAl 9203* 808 20784 5979 (56) (86)

DGA6 1142 487 5595 1139 (79) (57)

LB2 223 ND 2683 ND (92)

’ EBV-B cells were either irradiated (7500 rad) before their addition to the cultures containing PPD or whole-H37Rv, or irradiated after an 18-hr pulsing with both forms of antigen.

* Results are expressed as thymidine incorporation of T-cell lines and clones. The percentage of inhibi- tion when EBV-B cells were irradiated before the addition of antigen is reported in parentheses, considering 100% the presentation activity of EBV-B cells irradiated after an 18-hr antigen pulsing. A representative experiment of three replications is reported.

tyrosine copolymer) forms of antigen by primed murine B lymphocytes and showed that 100 to 1000 times more antigen is required when nonspecific B cells were used. However, the authors cannot exclude the possibility that lysis of sheep red cells in the culture leads to release of soluble antigens and that macrophages within the popula- tion process antigen, which is then taken up and presented by B cells. These two possibilities are clearly excluded in our experimental model as discussed above. Com- parison of antigen-specific versus nonspecific B-cell presentation of PPD and whole- H37Rv on a per celI basis showed a preferential activation by PPD-specific EBV-B cell clones (Figs. 4a, 4b, and 4~). This preferential presentation by PPD-specific B- cell clones is eliminated after EBV-B cell irradiation (Figs. 4d and 4e). The effect of irradiation on the antigen-presenting function of B cells has been shown to be related to the cellular activation state. In fact, resting B cells were reported to be unable to present antigen due to their sensitivity to the radiation procedure, whereas activated or neoplastic B cells are reported to be more radioresistant (26,27). However, in our experimental model we have established that irradiation (7500 rad) of EBV-B cells, independently of whether they are specific or nonspecific, impairs the processing and presentation of both forms of mycobacterial antigen (Table 4).

The meaning of specific B cells in tuberculosis is still unclear and, although specific antibodies arise in the course of infection and are used for diagnostic and prognostic purposes, there is no evidence on their role in the protective immunity. The present paper showing the ability of human B cells to phagocytose and to process whole M. tuberculosis cells and to present relevant determinants to resting and activated T cells opens the question to consider the role of B cells in the antituberculous immunity, at least as antigen-presenting cells. Work is in progress to analyze the role of B lym- phocytes from tuberculosis patients in handling viable Ad. tuberculosis organisms.

ACKNOWLEDGMENTS The authors thank D. Guerritore for his continuous encouragement. This work was supported by timds

from the WHO Vaccine Development Programme (IMMjTUB) and from the Italian Research Council

292 LOMBARD1 ET AL.

(CNR) Grants 85.00438, 85.00875.52, and 85.02199.44. G.L. was supported for part of this study by a fellowship of the AIRC (Associazione Italiana Ricerca sul Cancro).

REFERENCES

1. Steinman, R. M., and Cohn, Z. A., Immunol. Rev. 53,127, 1980. 2. Kaye, P. M., Chain, B. M., and Feldmann, M., J. Zmmunol. 134,1930, 1985. 3. Stingl, G., Gasse-Sting& L. A., Aberer, W., and Wolff, K., J. Immunol. 127,1707, 1981. 4. Hirschberg, H., Braathen, L. R., and Thorsby, E., Immunol. Rev. 66,57,1982. 5. Chu, E. T., Lareau, M., Rosenwasser, L. J., Dinarello, C. A., and Geha, R. S., J. Zmmunol. 134, 1976,

1985. 6. Krieger, J. I., Grammer, S. F., Grey, H. M., and Chesnut, R. W., J. Zmmunol. 135,2937, 1985. 7. Malynn, B. A., Romeo, D. T., and Wortis, H. H., J. Zmmunol. 135,980,1985. 8. Piccolella, E., Lombardi, G., and Morclli, R., J. Immunol. 125,2082, 1980. 9. Bergholtz, B., and Thorsby, E., Stand. J. Immunol. 10,267,1979.

10. Grey, H. M., Colon, S. M., and Chestnut, R. W., J. Immunol. 129,2382, 1982. 11. Elferink, B. G., Gttenhoff, T. H. M., and de Vries, R. R. P., Stand. J. Zmmunol. 22,585,1985. 12. Ottenhoff, T. H. M., Klatser, P. R., Ivanyi, J., Elferink, D. G., de Wit, M. Y. L., and de Vries, R. R.

P., J. Immunol. 319,66, 1986. 13. Gttenhoff, T. H. M., Elferink, D. G., Klatser, P. R., and de Vries, R. R. P., J. Zmmunol. 322,462,

1986. 14. Garzelli, C., Pug&, C., and Falcone, G., Eur. J. Immunol. 16,584, 1986. 15. Zamboni, L., and De Martino. C., J. Cell. Biol. 35, 148, 1967. 16. Matthews, R., Scoging, A., and Rees, A. D. M., ZmmunoZogy54,17,1985. 17. Tse, H. Y., Schwartz, R. H., and Paul, W. E., J. Zmmunol. 125,491,1980. 18. Gordon, A. H., D’Arcy Hart, P., and Young, M. R., Nature (London) 286,79,1980. 19. Gersow, M. J., D’Arcy Hart, P., and Young, M. R., J. Cell. Biol. 89,645, 1981. 20. Chu, E., Umetsu, D., Lareau, M., Schneeberger, E., and Geha, R. S., Eur. J. Zmmunol. 14,29 1, 1984. 2 1. Ziegler, H. K., and Unanue, E. R., Proc. Nat/. Acad. Sci. USA 79, 175, 1982. 22. Scala, G., Kuang, Y. D., Hall, R. E., Muchmore, A. V., and Gppenheim, G. G., J. Exp. Med. 159,

1637,1984. 23. Getrard, T. L., andvolkman, D. J., J. Zmmunol. 135,3217, 1985. 24. Rock, K. L., Benacerraf, B., and Abbas, A., J. Exp. Med. 160,1102, 1984. 25. Lanzavecchia, A., Nature (London) 314,537,1985. 26. Ashwell, J., DeFranco, A., Paul, W., and Schwartz, R. H., J. Exp. Med. 159,88 I, 1984. 27. Krieger, J. I., Grammer, S. F., Grey, H. M., and Chesnut, R. W., J. Zmmunol. 135,2937, 1985.