Primary proliferative and cytotoxic T-cell responses to HIV induced in vitro by human dendritic cells.

In earlier studies, primary proliferative and cytotoxic T-cell (CTL) responses to influenza virus were produced in vitro by using mouse dendritic cells (DC) pulsed with virus or viral peptide as the stimulus for syngeneic T cells in 20-microliters hanging-drop cultures. We have now adapted this system for producing primary responses with cells from non-immune donors to produce primary proliferative and CTL responses to human immunodeficiency virus I (HIV) and to HIV peptides in vitro using cells from normal human peripheral blood. All donors in this study were laboratory personnel with no history of HIV infection. DC enriched from peripheral blood were exposed to HIV in vitro and small numbers were added to T lymphocytes in 20-microliters hanging drops. Proliferative responses to virus-infected DC were obtained after 3 days in culture. After 6 days, CTL were obtained that killed virus-infected autologous--but not allogeneic--phytohaemagglutinin (PHA)-stimulated blast cells. Proliferative and CTL responses were obtained using cells from 14 random donors expressing a spectrum of major histocompatibility complex (MHC) types but the CTL, once produced, showed killing restricted by the MHC class I type. Treatment of cultures with monoclonal antibody (mAb) to CD4-positive cells at the beginning of culture blocked the development of both proliferative and CTL responses, but treatment after 5 days had no effect on the CTL activity. Treatment with MCA to CD8-positive cells at the beginning of culture did not block proliferation significantly, but treatment either before or after the 5-day culture period blocked CTL responses. Collaboration between proliferating CD4-positive cells and CD8-positive cells may thus be required to produce CTL of the CD8 phenotype. DC exposed to HIV also produced CTL that killed autologous blast cells pulsed with gp120 envelope glycoprotein. However, DC infected with whole virus did not produce CTL that lysed target cells pulsed with a synthetic peptide, which included a known T-cell epitope of gp120 (representing amino acids 111-126). DC pulsed with gp120 were a poor stimulus for the development of CTL. In contrast, DC pulsed with the peptide (111-126) stimulated both proliferative and CTL responses. The latter killed not only target cells pulsed with the peptide itself or with gp120 but also killed virus-infected autologous blast cells. CTL were again obtained reproducibly with this peptide using donors expressing a spectrum of MHC types.(ABSTRACT TRUNCATED AT 400 WORDS)     

High efficiency antigen presentation by thyroglobulin-primed murine splenic B cells

B cells primed in vivo with mouse or rat thyroglobulin present these antigens at very low concentrations to CH9, an Ly 1+2- T cell hybridoma specific for mouse and rat thyroglobulin. Presentation measured by interleukin 2 release from CH9 is sensitive to treatment with a monoclonal antibody eliminating splenic B cells but is unaffected by anti-Thy-1.2 or 33D1 (which destroy T cells and dendritic cells, respectively). Presentation is specific for the priming antigen and is blocked by preincubation of the B cells with sheep anti-mouse F(ab')2. We suggest that in this system, primed B cells present thyroglobulin and that this may represent a means by which an initial triggering event priming both B and T cells could allow maintenance of autoreactive responses in vivo in the presence of low concentrations of circulating antigen.     

Stimulatory and suppressive effects of infection of dendritic cells with HIV-1.

Two effects of HIV infection on human dendritic cells (DC) in vitro have been examined. The first was the stimulation of primary responses to HIV antigens in autologous lymphocytes from normal donors. When DC were exposed to HIV (10(4) TCID/10(5) cells) for up to 24 h before addition to autologous lymphocytes, a marked primary proliferative response to the virus was observed. No proliferative response was seen when the period of pre-exposure of DC to virus was extended. Cytotoxic T cells specific for HIV-infected target cells developed in stimulated cultures. The second effect of HIV infection of DC was to block responses to other antigens, such as alloantigens and the recall antigens tetanus toxoid and influenza virus. This inhibitory effect was only evident when the DC were exposed to HIV for longer than 24 h before being added to cultures. These in vitro studies suggest that infection of DC can produce both stimulatory and inhibitory responses in lymphocytes. Such effects operating through DC might underlie in vivo activity of HIV both in stimulating the proliferation of lymphocytes (e.g., in persistent generalised lymphadenopathy) and in the development of immunosuppression.     

Primary stimulation by dendritic cells induces antiviral proliferative and cytotoxic T cell responses in vitro

We used well-gassed hanging drop (20 microliters) cultures with high concentrations of purified T cells from normal BALB/c mice to examine whether dendritic cells (DC) can induce primary antiviral proliferative T cell responses and generate virus-specific CTL. We found that DC exposed to infectious influenza virus in vitro or in vivo in small numbers (0.1-1%) resulted in strong proliferation of responder T cells within 3 d, and this was strongly inhibited by antibodies to class II MHC molecules. In addition, in 5-d cultures, the influenza-treated DC generated CTL specifically able to lyse influenza-infected syngeneic target cells bearing MHC class I antigens. The most potent nucleoprotein (NP) epitope recognized by BALB/c CTL is peptide 147-158 (Arg156-) and influenza-infected DC in vitro stimulated CTL recognizing this peptide, thus mimicking the response in mice primed by intranasal influenza infection. We also induced T cell proliferation and virus-specific CTL in cultures of normal T cells by stimulating with DC pulsed with the natural NP sequence 147-158 or the potent peptide 147-158 (Arg156-). Small numbers of peritoneal exudate cells, after activation with Con A to produce class II MHC expression and after removal of DC with a specific mAb (33DI), did not lead to primary CTL generation but initiated secondary stimulation in vitro. Our results using the hanging drop culture method and DC as APC have implications for studying the T cell repertoire for viral components in humans without the necessity of previous immunization.     

Transfer of antigen between dendritic cells in the stimulation of primary T cell proliferation

Primary proliferative T cell responses require stimulation with antigen-pulsed dendritic cells (Ag-DC). Here we show that for optimal stimulation, dendritic cells (DC) not exposed directly to antigen are also required. Ag-DC added to DC-depleted T cells caused negligible primary stimulation; adding back DC resulted in stimulation. These effects were seen using the contact sensitizer fluorescein isothiocyanate (FITC), FITC conjugated to ovalbumin (FITC-OVA) or influenza virus as antigens. DC co-cultured with Ag-DC (using FITC or FITC-OVA) acquired antigen indicating that antigen was transferred between DC. DC that acquired antigen secondarily were separated by cell sorting and stimulated primary T cell proliferation directly. DC were also pulsed with FITC, washed thoroughly and incubated overnight. Super natants contained shed antigen since DC incubated in these supernatants acquired antigen as indicated by flow cytometry. DC acquiring the shed antigen also stimulated T cell proliferation although the stimulation was not as effective as that seen when cell contact between DC and antigen-bearing DC occurred. Thus, in primary stimulation, activation of T cells may occur when there is an antigen gradient between Ag-DC and DC and the mechanisms underlying these effects are now being sought. We propose that this unique interaction between antigen-presenting cells may be a paradigm for self/non-self discrimination.     

Dendritic cell infection, depletion and dysfunction in HIV-infected individuals.

Immune responses in resting T cells are initiated by the presentation of antigen by bone marrow-derived dendritic cells (DC). Normal DC are susceptible to infection with human immunodeficiency virus (HIV) in vitro (Patterson & Knight, 1987) and this blocks their capacity to stimulate T-cell responses to other antigens (Macatonia, Patterson & Knight, 1989a). To study the relationship between HIV and DC in patients and its relevance to the pathogenesis of disease, DC have been isolated from the blood of individuals in the different clinical categories, counted, examined for the presence of virus genome and their antigen-presenting capacity measured. Infection, depletion and impaired function of DC occur in early HIV infection. HIV seropositive patients who were asymptomatic and those with symptoms of disease had significantly reduced numbers of DC, but patients with persistent generalized lymphadenopathy had normal numbers. Between 3% and 21% of DC, identified as large low-density cells not bearing monocyte, lymphocyte or natural killer cell markers, were infected with HIV, as indicated by in situ hybridization. Less than 0.12% of the lymphocytes or monocytes were infected. The DC from infected individuals were poor at enhancing responses to the mitogen concanavalin A (Con A). They also caused low levels of stimulation in allogeneic lymphocytes in mixed leucocyte cultures. By contrast, T cells from asymptomatic patients gave normal T-cell responses to uninfected allogeneic DC, although those from acquired immunodeficiency syndrome (AIDS) patients did show reduced responsiveness. Defects in DC thus precede both the appearance of symptoms and changes in T cells and may be instrumental in the development of AIDS. Furthermore, since DC numbers and function differ at different stages of disease, monitoring these may contribute to clinical assessment and lead to new therapeutic approaches.     

Antigen-presentation by macrophages but not by dendritic cells in human immunodeficiency virus (HIV) infection.

Dendritic cells (DC) have a potent antigen-presenting capacity for recruiting resting T cells into immune responses. They also promote expansion of already activated memory T cells. By contrast, macrophages (M phi) are only effective in stimulating memory responses. Infection and depletion of DC occur in human immunodeficiency virus (HIV)-infected individuals and recruitment of T cells into primary responses is blocked. Here comparisons between DC and M phi in stimulating secondary T-cell responses in HIV infection were made. Adherent M phi, and DC isolated by a new method, were separated from peripheral blood of patients in different stages of HIV infection and from uninfected controls and added to allogeneic lymphocytes in mixed leucocyte reactions (MLR). Some were pulsed with influenza virus or tetanus toxoid and used to stimulate autologous T cells. Responses were measured from uptake of [3H]thymidine in 20 microliters hanging drop cultures. DC, but not M phi, from normal individuals stimulated MLR but both populations stimulated secondary responses to recall antigens. DC from all HIV seropositive individuals caused little or no stimulation of any lymphocyte responses. However, M phi from HIV seropositive asymptomatic individuals and those with persistent generalized lymphadenopathy stimulated responses to recall antigens. There was no stimulation using cells from acquired immune deficiency syndrome (AIDS) patients. Blocked DC but not M phi function may underlie progressive immunological non-responsiveness in HIV infection. Without recruitment of resting T cells, loss of memory T cells may be cumulative; failure of secondary activation (e.g. by M phi) would lead to lost T-cell activity. Identification and circumvention of the defect in DC could offer new therapeutic approaches.     

HIV I infection of dendritic cells

Dendritic cells (DC) from human peripheral blood are susceptible to productive and probably to latent infection with HIV-I. Infection of DC also occurs in vivo since in HIV-seropositive individuals Langerhans' cells of the skin and DC from peripheral blood, (in preparation) are infected. In peripheral blood 3-25% of DC, identified as large, low-density cells lacking monocyte markers, are infected as judged by in situ hybridization with an HIV probe. This contrasts with the lower proportion (< 0.2%) of other cells infected. DC exposed to HIV in vitro or in vivo fail to present other antigens or mitogens to stimulate T cells. This functional defect in infected DC is not blocked by the presence of soluble CD4 antigen and occurs in the absence of T cell infection suggesting a block at the level of the antigen-presenting cell itself. Infection, depletion and dysfunction of DC in HIV seropositive patients is already present in asymptomatic individuals and this precedes the appearance of T cell defects. We speculate that loss of functional DC may be a fundamental defect leading to a block in recruitment of resting T cells into immune responses. In contrast to the HIV-induced impairment of antigen presentation by DC, these cells were potent stimulators of responses to the HIV antigens themselves. Normal DC infected with HIV in vitro stimulated primary proliferative and cytotoxic T cell responses (in preparation). These were produced in cells from individuals expressing a range of different MHC types but the cytotoxic cells, once produced, killed autologous but not allogeneic, infected T cell blasts. Primary response to viral peptides can also be produced suggesting that this system may be useful for identifying immunogenic epitopes of HIV using cells from sero-negative, non-immunocompromised individuals.     

Dendritic cells and the initiation of contact sensitivity to fluorescein isothiocyanate.

Lymph node cells taken 24 hr after skin-painting mice with the contact sensitizer fluorescein isothiocyanate (FITC) induce delayed-type hypersensitivity in recipient mice. Skin-painting increased the number of dendritic cells (DC) in the draining lymph nodes without significantly changing the number of lymphocytes at 24 hr. The antigen was preferentially located on the DC. Raising the dose of FITC increased both the number of DC and the amount per cell. The addition of these DC to syngeneic lymph node cells at a ratio as low as 1:300 initiated proliferative responses in vitro. The level of proliferation was related to the amount of antigen on the DC. Mice given 50,000 of these fluorescent DC developed specific contact sensitivity reactions. DC exposed in vitro to FITC also acquired antigen and were able to initiate proliferative responses in vitro and to sensitize recipient mice. The DC may therefore be the prime cell involved in the induction of delayed-type hypersensitivity.