INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Measles virus was the first virus reported to alter immune function (39). Since that time, various investigators have studied different aspects of immune dysfunction, including reduction of mitogen-induced proliferation of T cells obtained from patients following natural infection and from individuals following vaccination with measles virus (6, 13, 15, 20, 21, 23, 25, 37, 40-42) or by direct infection of lymphoid cells (36) as a means to study measles-induced immunosuppression (2, 4, 7, 16, 26-30, 43, 44). Similarly, B cells have been studied for functional alterations induced by measles virus infection. McChesney et al. found that infection of B cells led to decreased antibody production when B cells were stimulated by pokeweed mitogen (28). More recently, Ravanel et al. have shown that the nucleocapsid protein of measles virus had the ability to bind to B cells through the Fc receptor and inhibit immunoglobulin (Ig) synthesis (34). In contrast, measles virus-infected T cells still have the ability to produce cytokines required to help uninfected B cells differentiate into plasma cells and secrete Ig (31). Thus, measles virus infection can affect T- and B-cell function.

We have previously shown that infection of antigen-specific T-cell lines with measles virus can inhibit antigen and mitogen driven proliferation (1). We have also found that infected T cells produce a soluble factor which can also inhibit the proliferation of uninfected antigen-specific T cells in response to antigen, mitogen, and superantigen (38). By sizing experiments, the antiproliferative activity was shown to be larger than 60 kDa. Heating to 56°C for 30 min destroyed the activity. The activity was found in fractions smaller than 10 kDa after trypsin treatment (38). These data suggest that the factor may be a new cytokine with antiproliferative effects.

In this work, we examined the immunologic consequences of B-cell infection by measles virus. We demonstrate that measles virus infection of B cells at very low multiplicities of infection (MOIs) can inhibit the spontaneous proliferation of Epstein-Barr virus (EBV)-transformed B cells. This inhibitory effect was due to a soluble factor that was actively secreted by infected B cells. Further, the antiproliferative effect was not due to interleukin-10 (IL-10), transforming growth factor  (TGF-), or alpha/beta interferon (IFN-/). The soluble factor was larger than 50 kDa and inactivated at temperatures above 55°C. In addition, treatment of B cells with this soluble factor inhibited their ability to act as fully competent antigen-presenting cells (APCs) for antigen-specific T-cell proliferation. The inhibition of APC function was not due to the elaboration of the antiproliferative factor. The effects of direct infection and the production of the antiproliferative factor would limit T-cell expansion. However, effector function has not been studied. To determine whether effector functions were affected, cloned cytolytic CD4⁺ and CD8⁺ T-cell lines were tested for the ability to lyse target cells. Whether infected with measles virus or treated with high concentrations of soluble factor, these clonal T-cell lines retained the ability to perform cytolytic functions. This is the first demonstration that infection of B cells by measles virus can inhibit antigen presentation and that a soluble factor can mirror the inhibitory effects of infection. However, once activated T cells are generated, they retain the ability to kill. This suggests that while slow to be generated, effector T cells, once activated and present in sufficient numbers, maintain the ability to clear virus-infected cells.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

B cells. Peripheral blood mononuclear cells (PBMCs) were obtained from 15 ml of autologous blood. PBMCs were separated on Ficoll-Paque (Pharmacia Biotech, Uppsala, Sweden). Cells were collected and washed three times with RPMI 1640 (Gibco/BRL, Grand Island, N.Y.). One milliliter of a culture supernatant from IA3 cells producing EBV was added to 3 × 10⁶ PBMCs, and cells were cultured in the presence of cyclosporin (2 µg/ml) in RPMI 1640 supplemented with 10% fetal calf serum (FCS) and 1% antibiotics in 24-well plates. Transformed B cells were observed growing out of the PBMC population within 14 days after infection. B cells were subcultured two to four times prior to being used as APCs. By flow cytometry analyses, B-cell lines were greater than 99% surface Ig positive and less than 1% CD3 positive.

Antigen-specific T-cell lines. PBMCs from healthy volunteers were obtained. Antigen-specific T-cell lines were isolated as previously described (1, 3). Briefly, T-cell lines were generated by culturing fresh PBMCs (3 × 10⁶ cells/well) with antigen (purified protein derivative [PPD; 5 µg/ml; Serum Institute, Copenhagen, Denmark] or myelin basic protein [MBP; 30 µg/ml]) in 24-well plates. At 1 week, these primary cultures were examined by inverted microscopy. Wells showing colonies of growing T cells were maintained with the addition of recombinant human IL-2 (20 U/ml; Chiron, Emeryville, Calif.) every 2 to 3 days and restimulated with either PPD or MBP presented by irradiated (4,000 R) autologous PBMCs as APCs at about 2-week intervals. Antigen-specific proliferation was assessed by [³H]thymidine incorporation (0.5 µCi/well) in response to stimulation with antigen. Cultured T-cell lines were CD4⁺ as determined by flow cytometry (Huntsman Cancer Institute Flow Cytometry Facility).

Infection, proliferation assays, and supernatant fluids. B cells were infected with the Edmonston strain of measles virus (American Type Culture Collection, Rockville, Md.) at various MOIs for 1 h at 37°C. Control B cells were mock infected and incubated at 37°C for 1 h. Cells were then washed three times with phosphate-buffered saline, resuspended in RPMI 1640 (Gibco/BRL), supplemented with 10% FCS (Gibco/BRL) and 1% antibiotics (MediaTech, Herndon, Va.), and incubated overnight at 37°C. B cells were washed three times, and viability was determined by trypan blue exclusion. Cells were then irradiated (6,000 R) and used as APCs. For proliferation assays, and 3 × 10⁴ to 5 × 10⁴ cells infected as described above were added to each well in 96-well plates, pulsed with [³H]thymidine on day 2, and harvested on day 3 postinfection.

Measles virus plaque assay and infectious center assay. Virus was diluted in phosphate-buffered saline containing 3% FCS and 1% antibiotics (MediaTech). Tenfold serial dilutions were made, and 200 µl was added to confluent Vero cell monolayers in six-well cluster plates (Costar, Pleasanton, Calif.). Virus was allowed to adsorb for 1 h at 37°C in a humidified incubator in a 5% CO₂ atmosphere. Cluster plates were rocked every 10 to 15 min, fluid was aspirated after 1 h, and 3 ml of overlay medium (medium 199 [Gibco] containing 1% antibiotics and 10% FCS) was added. Cells were incubated for 6 days at 37°C in a humidified incubator in a 5% CO₂ atmosphere, at which time cell monolayers were stained with neutral red and plaques were enumerated (11).

CTL assay. Target cell lines consisting of 2 × 10³ B-lymphoblastoid cells which differed at minor HLA antigens were labeled with ⁵¹Cr (DuPont NEN, Boston, Mass.), mixed with allo-specific CD4⁺ or CD8⁺ T-cell clones at various effector/target ratios, and cultured in 96-well plates. After 4 h, plates were centrifuged, and 1/2 volume of medium was removed from each well and counted in a Packard Cobra gamma counter. Maximum release and minimum release were determined by treating labeled cells with Triton X-100 (Sigma, St. Louis, Mo.) and medium, respectively (32).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

B cells become infected during the viremic stage of measles (8). When infected in vitro, B cells lose the ability to proliferate due to an inhibition in G₁ of the cell cycle (30). The studies by McChesney et al. (30) used high MOIs where all B cells would be expected to be infected. To extend these observations and assess the effect of measles virus infection on the proliferation of EBV-transformed B-cell lines, B cells were infected at MOIs of 0.05 to 0.000005 (28). We wanted to determine how few infectious measles virions could be used to inhibit proliferation and approximate levels that would be comparable to the viremic phase of infection. It is clear from Table 1 (experiment A) that B cells infected with very few infectious virions were inhibited from proliferation. UV-inactivated measles virus had no effect on B-cell proliferation (experiment B). Thus, we have confirmed the original observation by McChesney et al. (30) and have demonstrated that only a few infectious virions were needed to inhibit proliferation.

In these assays, it was important to determine whether MOI correlated with the actual number of infected cells. Therefore, we infected B cells at an MOI of 0.001 and performed infectious center assays. A portion of the solution used to infect the cells was also plaqued to confirm the actual number of input PFU. Three experiments were performed. The actual MOIs were 0.0014, 0.0017, and 0.0012 as determined by plaque assay. The numbers of infectious centers were 2.5/10,000, 4.3/10,000, and 4.2/10,000 cells, respectively. This corresponds to an input of 14.3 PFU/10,000 cells (average), resulting in 3 per 10,000 (average) cells actually being infected. Therefore, under the conditions used in our studies, the efficiency of infection was 21%.

We have recently demonstrated that infection of antigen-specific T cells with measles virus results in an inability to proliferate in response to antigen presented by APCs (1) and that supernatants from infected T cells contain a factor that could inhibit antigen-, mitogen- and IL-2-induced T-cell proliferation (38). To ascertain whether measles virus infection of B cells (APCs) could affect antigen-specific T-cell proliferation, B cells were infected with measles virus at various MOIs or mock infected, washed three times, and cultured for 18 h at 37°C. B cells were then collected, washed extensively, irradiated, and used as APCs. PPD-specific T cells, B cells (mock or infected), and PPD were cultured for 3 days and pulsed with [³H]thymidine. Cultures were harvested the next day (day 4, and thymidine incorporation was determined. As shown in Table 2, infection of B cells in these assays suppressed antigen-specific T-cell proliferation. Addition of as few as five infectious virions to 10³ B cells resulted in a greater than 90% decrease in antigen-specific T-cell proliferation. Infectious center assays of four experiments revealed that among B cells infected at an MOI of 0.005, 3 to 7.4% of the total cells in the cultures scored positive.

A potential explanation for the above observations is that the few infected cells secrete a cytokine or synthesize a viral product which has antiproliferative effects and/or affect the ability of B cells to present antigen. To further explore this issue, B cells were infected at an MOI of 0.005 or mock infected, cultured for 18 h, washed extensively, and further cultured for 72 h, after which supernatants were collected. The supernatants from infected and mock-infected B cells were exposed to UV irradiation and then added to uninfected cultures of antigen-specific T cells, APCs, and antigen. The data are shown in Table 3, experiment A. The supernatant from infected B cells had the ability to inhibit the proliferation of stimulated uninfected T-cell cultures. Similar inhibitory effects were observed for a different T-cell line specific for MBP generated from a different individual. In contrast, infected supernatant fluids were obtained from measles virus-infected HeLa and Vero cells. When supernatants were UV treated, filtered, and added to T-cell proliferation assays, no suppression of T-cell proliferation was observed (Table 3, experiment B).

To determine whether the supernatant from infected B cells could affect B-cell function, three sets of experiments were conducted. First, B cells were cultured in the presence of various dilutions of UV-treated supernatants from infected or mock-infected B-cell cultures (Table 4). Cultures were pulsed with [³H]thymidine and harvested the next day, and thymidine incorporation was measured. Proliferation was inhibited greater than 90% at a 1/25 or 1/50 dilution of supernatant from infected B cells (Table 4). Thus, a factor in the supernatant fluid inhibited proliferation of the EBV-transformed B cells.

Second, since B cells were inhibited from proliferating by the supernatant factor, we wanted to determine how long the antiproliferative effect lasted. To accomplish this, B cells were treated for 3 days (equivalent to the time of our standard B-cell proliferation assay) with supernatant from infected or mock-infected B cells. After 3 days, B cells were washed and placed into culture without supernatant supplementation. On days 2, 6, and 8, equivalent numbers of B cells from each group were aliquoted into 96-well plates. Supernatant (mock infected and infected)-treated or nontreated B cells were pulsed with [³H]thymidine and harvested 6 to 8 h later, and incorporation was measured. As shown in Fig. 1, B-cell proliferation was suppressed on days 2 and 6 following treatment with supernatant from infected B cells compared to B cells not treated or treated with supernatant from uninfected B cells. By day 8, proliferation levels of the cells were comparable. These data show that even though equivalent numbers of cells were added to each well (5 × 10⁴ cells/well), treated cells did not incorporate thymidine until day 8. Thus, B cells were able to recover and proliferate after supernatant treatment.

View larger version (18K): [in this window] [in a new window]   FIG. 1.   Proliferation of B cells after supernatant treatment. B cells were cultured with medium alone (), UV supernatant from measles virus-infected B cells (), or supernatant from mock-infected B cells () for 3 days. B cells were then washed three times and cultured in medium not containing supernatant. On days 2, 6, and 8 of culture, 5 × 104 B cells were added to 96-well plates and pulsed for 6 to 18 h with [3H]thymidine.

Third, we examined whether treatment of uninfected B cells with supernatant from infected B cells could inhibit the ability of these B cells to act as APCs in antigen-specific T-cell proliferation assays. B cells were incubated with UV-treated supernatant from infected or uninfected B cells. Cells were then washed three times, counted, and irradiated with 6,000 R. These B cells were then used as APCs in antigen-specific T-cell proliferation assays. A T-cell line reactive with MBP was used. As shown in Table 5, in assays using B cells treated with supernatant from measles virus-infected B cells, no proliferation was observed. In contrast, when B cells were treated with supernatant from uninfected B cells, T-cell proliferation in response to MBP was observed.

Two explanations could account for the data presented in Table 5: (i) the treated B cells elaborated additional antiproliferative factor, which could inhibit T-cell proliferation; and (ii) the treated B cells were not effective in presenting antigen to T cells. To differentiate between these two possibilities, the following experiment was performed. Uninfected B cells were treated for 24 h with supernatant from measles virus-infected or uninfected B cells. Cells were then washed three times and cultured in the absence of supernatant. The supernatants from these treated B cells were harvested after 3 days and assayed for the ability to inhibit B-cell proliferation. No inhibition of B-cell proliferation or IL-2-driven T-cell proliferation was observed in supernatants from treated B cells (data not shown). This finding indicates that treated B cells do not elaborate an antiproliferative factor, which suggests that treated B cells do not inhibit T-cell proliferation by the production of the antiproliferative factor but are unable to present antigen and do not directly inhibit T-cell proliferation.

To determine whether B cells treated with UV-irradiated supernatant from measles virus-infected B cells resulted in an abortive infection, we incubated uninfected B cells with undiluted supernatant for 24 h and then stained them for surface expression of measles virus HA. Two reagents were used to detect measles virus HA and/or fusion protein. The first was a monoclonal antibody to HA (9) and human SSPE serum containing antibody to HA and fusion protein (12). These were used at very low dilutions (1/25 and 1/100, respectively) to detect low levels of measles virus protein expression. No differences in mean fluorescence intensity were observed between B cells treated with supernatant derived from measles virus-infected B cells and B cells treated with supernatant derived from mock-infected B cells (data not shown). There was a dramatic increase in fluorescent staining in measles virus-infected B cells treated with the same reagents (positive control).

Several cytokines, including TGF-, IL-10, IFN-, and IFN- (5, 22, 33, 35), have been shown to inhibit T- and B-cell proliferation. To determine whether these cytokines play a role in the inhibition of proliferation observed by supernatants from infected B cells, the supernatants were supplemented with neutralizing antibodies to the cytokines and then tested for the ability to inhibit B-cell proliferation. As shown in Table 6, anti-TGF-, IL-10, IFN-, and IFN- had no effect on the supernatant's ability to inhibit proliferation, which indicates that the antiproliferative effect mediated by the supernatant was not due to these cytokines.

View this table: [in this window] [in a new window]   TABLE 6.   Antiproliferative effects are not due to TGF-, IL-10, or IFN- or -a

The supernatant from infected B cells was subjected to sizing membranes. B cells were then cultured with supernatant fractions larger than 100, 50 to 100, and 10 to 50 kDa. As shown in Table 7, antiproliferative activity was seen in fractions larger than 50 kDa but not in equivalent fractions from supernatant fluid from mock-infected B cells.

To determine whether the antiproliferative activity was stable to heating, supernatants from infected B cells were incubated at 50, 55, and 62°C. B cells were then cultured in the presence of the treated supernatant. As shown in Table 8, the antiproliferative activity was stable for 2 h at 50°C and was destroyed at 55 and 62°C.

We have shown that T cells infected with measles virus (1) or uninfected T cells treated with supernatant from infected T cells do not proliferate in response to antigen and APCs (38). In addition, we have reported that cytokine production in infected T cells is relatively intact, particularly for IL-2 production (1). Here, we provide data that B-cell APC function is inhibited. However, effector functions of T cells have not been studied. Therefore, we wanted to determine whether killing by CD4⁺ or CD8⁺ T cells was altered by measles virus infection or treatment with soluble factor from lymphoid cells. An allo-specific CD4⁺ CTL clone (C-7) was infected with measles virus at an MOI of 0.05 (18 h) or treated with supernatant from measles virus-infected or mock-infected T cells. Cells were cultured overnight, washed, and counted. A standard ⁵¹Cr release assay was performed on allotarget cells. Shown in Table 9 are results of two experiments examining the specific killing by the CD4⁺ T cells. T cells treated with 1/2 dilution of supernatant from infected T cells killed at high effector/target ratios, but killing was somewhat reduced when effector cells were in limited numbers (Table 9, experiment 1). This reduction in killing was not observed when T cells were incubated overnight with a 1/4 dilution of supernatant (Table 9, experiment 2). Allo-specific CD8⁺ T cells were also tested for the ability to kill allotarget cells. CD8⁺ T cells were infected with measles virus at an MOI of 0.05 or treated with a 1/2 or 1/4 dilution of supernatant from mock-infected B cells or measles virus-infected B cells. In experiments 3 to 5, two independent T-cell lines (F5 and F6) were tested. Whether the T cells were infected with measles virus or not, the cytotoxic activity remained intact. The first cell line (Table 9, experiment 3) was treated for 18 h with a 1/2 dilution of supernatant and then tested for cytolytic activity. At the various effector/target ratios, there was a slight decrease in the cells' ability to lyse allotype-specific targets. However, when the same CD8⁺ T-cell clone was treated with a 1/4 dilution of supernatant from B cells, there was no difference between the T cells treated with infected B-cell supernatant and those treated with uninfected B-cell supernatant. Similar data were obtained with the second clone incubated at a 1/4 dilution of measles virus-infected supernatant (Table 9, experiment 5). These data corroborate the CD4⁺ data (experiments 1 and 2). Thus, T cells, whether infected with measles virus or supernatant from infected B cells, retain cytolytic activity.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Here we demonstrate that infection of B cells can have profound effects on proliferation and ability to present antigen to T cells. We show that infection with as few as five infectious viral particles added to 10,000 EBV-transformed B cells can inhibit their proliferation. Similarly, five infectious viral particles used to infect 1,000 B cells used as APCs can reduce antigen-specific T-cell proliferation, where by infectious center assays less than 8% of total cells become infected. Therefore, only a minority of cells become infected at the time of the assays. In addition, supernatants from infected B cells (containing no detectable virus) had the ability to inhibit antigen-specific T-cell proliferation as well as the proliferation of EBV-transformed B cells. Supernatant-treated B cells did not produce additional antiproliferative factor yet were not able to act as APCs and support antigen-specific T-cell proliferation. This finding suggests that B cells that have come in contact with the factor lack the ability to present antigen to T cells.

Viruses have evolved many strategies to evade the host immune response. In the case of measles virus, direct infection of lymphoid cells leads to the inhibition of proliferation. The effect of this is to limit the number of expanding virus-specific T cells and antibody-producing B cells. This limited expansion allows the virus to replicate to sufficient numbers for spread to new hosts before the immune response can eliminate virus-infected cells and neutralize virus. We have shown that T cells infected with measles virus produce an antiproliferative cytokine that can limit clonal expansion of the antigen-specific T cells (38). Here we demonstrate that B cells can also produce an inhibitory factor, most likely the same entity since its physical properties are similar. The factor has the ability to limit the proliferation of lymphoid cells but does not affect the growth of HeLa cells (data not shown). Besides inhibiting proliferation of transformed B cells, the factor affects the ability of B cells to present antigen. This is the first example of a virus causing a cell to produce a factor that inhibits APC function.

Our data on presentation of antigen are consistent with those of Leopardi et al. (24), who studied the ability of measles virus-infected monocytes to present antigen. In experiments using an MOI of 5, they found that major histocompatibility complex class II was up-regulated by infection and that monocytes retained the ability to present measles virus antigens but were unable to present other exogenous antigens to T lymphocytes. A difference between our study and theirs is that we used 10⁴-fold less virus. In addition, we studied the effects on proliferation and presentation of determined antigens by B cells and treatment with soluble factor.

Unlike human immunodeficiency virus, measles virus is ultimately cleared by the host's immune response. Jacobson et al. found that much of the CTL response in humans to measles virus was major histocompatibility complex class II restricted (18), and they identified the restricting element as HLA-DR2 (17). They suggested that the CTL response was skewed to the internal viral proteins such as the nucleocapsid protein (19). Our data indicate that while proliferative responses are reduced and APC function is hindered, the ability to kill is not abrogated by infection or treatment with soluble factor. We find that CD4⁺ and CD8⁺ T-cell killing is intact (Table 9) and have data indicating that measles virus-infected MBP-specific T cells can still kill MBP-pulsed B cells (data not shown). We speculate that while slow to be generated, virus-specific CTLs once activated following infection, can still kill virus-infected cells. This ability would eventually contribute to viral clearance, but only after the virus had spread to other susceptible individuals.