Uncoupling of virus-induced inflammation and anti-viral immunity in the brain parenchyma.

Non-neuroadapted influenza virus confined to the brain parenchyma does not induce antigen-specific immunity. Nevertheless, infection in this site upregulated major histocompatibility complex (MHC) class I and MHC class II expression and recruited lymphocytes to a perivascular compartment. T cells recovered from the brain had an activated/memory phenotype but did not respond to viral antigens. In contrast, T cells recovered from the brain after infection in a lateral cerebral ventricle, which is immunogenic, showed virus-specific responses. As with infectious virus, influenza virus-infected dendritic cells elicited virus-specific immunity when inoculated into the cerebrospinal fluid but not when inoculated into the brain parenchyma. Thus, inflammation and dendritic cell function were both uncoupled from immune priming in the microenvironment of the brain parenchyma and neither was sufficient to overcome immunological privilege.

Antigen-presenting cells. Experimental and clinical studies of dendritic cells.

We present a summary of our experimental and clinical studies of dendritic cells (DCs) in five different areas. First, human DCs undergo chemotaxis and transendothelial migration in response to distinct chemokines, depending on their maturation state; inducible chemokines may recruit immature DCs to sites of inflammation and infection, whereas constitutive chemokines may regulate subsequent migration to lymphoid tissues. Second, M-tropic but not T-tropic strains of HIV-1 induce and modulate DC chemotaxis in culture; conceivably this may facilitate viral infection and contribute to pathogenesis in vivo. Third, malaria-infected erythrocytes modulate the maturation and inhibit the function of DCs; adhesion of infected erythrocytes to DCs may impart a selective advantage to the parasite by modulating host immunity. Fourth, a phase I clinical trial of DC-based immunotherapy of cancer has been initiated in which autologous DCs are administered directly into metastatic skin lesions; immune responses to a recall antigen, with which the cells were pulsed, were enhanced in several patients. Finally we describe the growth and characterization of a stably immature, maturation-resistant stage of DCs generated from mouse bone marrow that can induce transplantation tolerance in vivo, and outline studies of their capacity to induce responses of transgenic T cells in vivo.

Formation and kinetics of MHC class I-ovalbumin peptide complexes on immature and mature murine dendritic cells.

Dendritic cells (DC) are professional antigen-presenting cells that are able to induce primary T cell responses. Therefore, several strategies employ peptide-pulsed DC in tumor immunotherapy. For efficient antigen presentation and induction of an immune response by DC the number and stability of MHC I-peptide complexes is crucial. We studied this issue by using the antibody 25-D1.16 that specifically detects OVA peptide SIINFEKL in conjunction with H-2 Kb molecules, and determined its kinetics on mature and immature bone marrow-derived murine DC. Optimal peptide loading was reached after 8-16 h at 50 microM peptide pulse, and was comparable in serum-free versus serum-containing medium. Stimulation of DC with LPS or Poly I:C, and to a lesser extent TNF-alpha, upregulated the total number of surface MHC I molecules and thus improved peptide loading. Pulse-chase experiments revealed a constant half-life of peptide/Kb complexes independent of preceding DC stimulation or their maturation stage. The duration of peptide/Kb complex expression on mature DC, however, could be extended from 24 h to 72 h when the cultures were pretreated with LPS or Poly I:C, but not TNF-alpha. These data might have important implications for the clinical application of peptide-pulsed DC in tumor immunotherapy.

Macrophage-tropic HIV induces and exploits dendritic cell chemotaxis.

Immature dendritic cells (iDCs) express the CC chemokine receptor (CCR)5, which promotes chemotaxis toward the CC chemokines regulated on activation, normal T cell expressed and secreted (RANTES), macrophage inflammatory protein (MIP)-1alpha, and MIP-1beta. By contrast, mature DCs downregulate CCR5 but upregulate CXC chemokine receptor (CXCR)4, and as a result exhibit enhanced chemotaxis toward stromal cell-derived factor (SDF)-1alpha. CCR5 and CXCR4 also function as coreceptors for macrophage-tropic (M-tropic) and T cell-tropic (T-tropic) human immunodeficiency virus (HIV)-1, respectively. Here, we demonstrate chemotaxis of iDCs toward M-tropic (R5) but not T-tropic (X4) HIV-1. Furthermore, preexposure to M-tropic HIV-1 or its recombinant envelope protein prevents migration toward CCR5 ligands. The migration of iDCs toward M-tropic HIV-1 may enhance formation of DC-T cell syncytia, thus promoting viral production and destruction of both DC and T helper lymphocytes. Therefore, disturbance of DC chemotaxis by HIV-1 is likely to contribute to immunosuppression in primary infection and AIDS. In addition, migration of iDCs toward HIV-1 may aid the capture of R5 HIV-1 virions by the abundant DC cell surface protein DC-specific intercellular adhesion molecule (ICAM)3-grabbing nonintegrin (DC-SIGN). HIV-1 bound to DC cell-specific DC-SIGN retains the ability to infect replication-permissive T cells in trans for several days. Consequently, recruitment of DC by HIV-1 could combine with the ability of DC-SIGN to capture and transmit the virus to T cells, and so facilitate dissemination of virus within an infected individual.

Immature dendritic cells generated with low doses of GM-CSF in the absence of IL-4 are maturation resistant and prolong allograft survival in vivo.

Dendritic cells (DC) were cultured from mouse bone marrow (BM) progenitors in low concentrations of granulocyte-macrophage colony-stimulating factor (GM-CSF) (GM(lo) DC) by two different protocols. The phenotype and functional properties of these GM(lo) DC were compared to those of standard BM-DC cultures generated in high concentrations of GM-CSF (GM(hi) DC) or in low GM-CSF plus IL-4 (GM(lo)/IL-4 DC). An effect of IL-4 on maturation was observed only at low but not high doses of GM-CSF. Compared to mature DC, GM(lo) DC were phenotypically immature, weak stimulators of allogeneic and peptide-specific T cell responses, but substantially more potent in presentation of native protein. Immature GM(lo) DC were resistant to maturation by lipopolysaccharide, TNF-alpha or anti-CD40 monoclonal antibodies, as the expression of co-stimulatory molecules was not increased, and stimulatory activity in oxidative mitogenesis was not enhanced. These maturation-resistant immature GM(lo) DC induced T cell unresponsiveness in vitro and in vivo. GM(lo) DC also prolonged haplotype-specific cardiac allograft survival (from 8 days to >100 days median survival time) when they were administered 7 days (but not 3, 14 or 28 days) before transplantation. Our findings may have important implications for future studies in T cell tolerance induction in vivo.

Human cultured dendritic cells show differential sensitivity to chemotherapy agents as assessed by the MTS assay.

Assessment of the chemosensitivity of dendritic cells (DC) may allow more rational development of combined chemotherapy and immunotherapy protocols. Human monocyte-derived DC generated reproducible results in the MTS (Owen's reagent) assay, which was then used to study DC survival after treatment with four different chemotherapy agents. DC preparations from three different donors were used per drug. DC were sensitive to doxorubicin (concentration range 0.1-50 microM) with variation in sensitivity between donors (IC50 244-1100 nM). The most extreme variation was seen for vinblastine (concentration range 250-0.025 microM with IC50 0.15-17.25 microM). In contrast, there was relative resistance to etoposide (concentration range 0.2-200 microM) and 5-fluorouracil (concentration range 0.7-7700 microM) with no toxicity seen until 50 microM and 770 microM respectively. The function of DC in allogeneic mixed leucocyte reactions closely paralleled results from the MTS assays. The differential sensitivity to chemotherapy agents did not appear to be due to expression of P-glycoprotein. These results suggest that etoposide or 5-fluorouracil is less likely to reduce the immunotherapeutic potential of DC and may be valuable in the design of prodrug activation therapy.

Death, destruction, danger and dendritic cells.

Dendritic cells are known to be involved in recognition of foreign antigens and initiation of specific T-cell responses. The 'danger hypothesis' suggests that the immune system can also respond to endogenous signals of distress. New data indicate that dendritic cells are the first to respond to these signals, although the mechanisms involved in their activation are unclear (pages 1249-1255).

Potent immunostimulatory dendritic cells can be cultured in bulk from progenitors in normal infant and adult myasthenic human thymus.

Low density cells can readily be enriched from thymus tissue both of children undergoing cardiac surgery and of older patients with myasthenia gravis, and can be cryostored in bulk. When fresh or thawed cells are cultured with granulocyte-macrophage colony-stimulating factor and stem cell factor with or without tumour necrosis factor-alpha (TNF-alpha), they generate numerous cells with the characteristic ultrastructural, phenotypic and functional properties of dendritic cells. These proved to be very potent, both as stimulators of primary mixed leucocyte responses and as costimulators in oxidative mitogenesis. Especially after exposure to TNF-alpha, these dendritic cells also processed a natural epitope from a 437-residue polypeptide and presented it efficiently to an autoimmune T-cell clone (of T helper type 0 phenotype). Thus, immunostimulatory dendritic cells can be cultured in relative abundance from progenitors in infant and adult human thymus. Both are convenient sources of potent antigen-presenting cells of identifiable origins, e.g. for use in selecting human T-cell lines.

Plasmodium falciparum-infected erythrocytes modulate the maturation of dendritic cells.

The malaria parasite Plasmodium falciparum is one of the most successful human pathogens. Specific virulence factors remain poorly defined, although the adhesion of infected erythrocytes to the venular endothelium has been associated with some of the syndromes of severe disease. Immune responses cannot prevent the development of symptomatic infections throughout life, and clinical immunity to the disease develops only slowly during childhood. An understanding of the obstacles to the development of protective immunity is crucial for developing rational approaches to prevent the disease. Here we show that intact malaria-infected erythrocytes adhere to dendritic cells, inhibit the maturation of dendritic cells and subsequently reduce their capacity to stimulate T cells. These data demonstrate both a novel mechanism by which malaria parasites induce immune dysregulation and a functional role beyond endothelial adhesion for the adhesive phenotypes expressed at the surface of infected erythrocytes.

Dendritic cell chemotaxis and transendothelial migration are induced by distinct chemokines and are regulated on maturation.

The capacity of dendritic cells (DC) to initiate immune responses is dependent on their specialized migratory and tissue homing properties. Chemotaxis and transendothelial migration (TEM) of DC were studied in vitro. Immature DC were generated by culture of human monocytes in granulocyte-macrophage colony-stimulating factor and IL-4. These cells exhibited potent chemotaxis and TEM responses to the CC chemokines macrophage inflammatory protein (MIP)-1alpha, MIP-1beta, RANTES, and monocyte chemotactic protein-3, and weak responses to the CC chemokine MIP-3beta and the CXC chemokine stromal cell-derived factor (SDF)-1alpha. Maturation of DC induced by culture in lipopolysaccharide, TNF-alpha or IL-1beta reduced or abolished responses to the former CC chemokines but markedly enhanced responses to MIP-3beta and SDF-1alpha. This correlated with changes in chemokine receptor expression: CCR5 expression was reduced while CXCR4 expression was enhanced. These findings suggest two stages for regulation of DC migration in which one set of chemokines may regulate recruitment into or within tissues, and another egress from the tissues.

Bacterial lipopolysaccharide contamination of commercial collagen preparations may mediate dendritic cell maturation in culture.

Dendritic cells (DC) are potent antigen presenting cells, which are responsible for the initiation of naive T and T-dependent immune responses. The present studies were based upon recent reports that commercial collagen I preparations induce the maturation of human DC in vitro. We show that human blood monocyte-derived (GM-CSF and IL-4 cultured) DC pulsed on collagen I-coated plates undergo a dose-dependent increase in stimulatory capacity in oxidative mitogenesis assays. This is accompanied by the upregulation of costimulatory molecules (CD40, CD80, CD86), CD25, ICAM-1 and the DC-specific marker CD83. The maturation effect is more potent than TNF-alpha, which is a known mediator of DC function. However, bacterial lipopolysaccharide (LPS), a powerful inducer of DC maturation, was found to be present at very high levels in one commercial collagen solution that was tested. The effect of LPS upon DC maturation was similar to culture with collagen. Furthermore, a different collagen I preparation with low levels of LPS contamination was less effective at inducing DC maturation, while spiking the collagen solution with LPS prior to plastic coating equalised these effects. Finally, human monocyte-derived DC were found not to express typical collagen receptors VLA-1, 2 and 3. We therefore propose that LPS contamination may at least partially explain reported collagen I induced DC maturation.

Dendritic cells.

Dendritic cells are potent stimulators of immune responses against foreign antigens. Recent advances in this area include the delineation of distinct developmental pathways for different dendritic cell subsets; the emerging concept that one dendritic cell subset has regulatory functions that may contribute to induction of tolerance to self antigens; increased understanding of the interaction of dendritic cells with microbial products and viruses, such as HIV; and the application of dendritic cells for immunotherapy in certain cancers and possibly for the induction of transplantation tolerance.

Identification and isolation of CD1a positive putative tumour infiltrating dendritic cells in human breast cancer.

Identification of dendritic cells (DC) in human tissues has been technically problematic due to the lack of truly specific immunohistochemical markers for DC's. Human dendritic cells express CD1a glycoprotein at certain points in their life cycle. CD1a positive cells are present in many human tumours and have been associated with improved survival. However, little information exists concerning the separation of DC from human tumours. The current study reports that human breast carcinomas have low densities of CD1a positive cells with dendritic morphology, and details are shown of a technique for successful separation of these cells from tumour tissues.

Importance of localized skin infection in tick-borne encephalitis virus transmission.

Arboviruses are transmitted to vertebrates by the "bite" of infected arthropods. Events at the site of virus deposition are largely unknown despite increasing evidence that blood-sucking arthropods immunomodulate their skin site of feeding. This question is particularly relevant for ixodid ticks that feed for several days. To examine events under conditions mimicking tick-borne encephalitis (TBE) virus transmission in nature (i.e., infected and uninfected Ixodes ricinus ticks feeding on the same animal), infected adult and uninfected nymphal ticks were placed in one retaining chamber (skin site A) and uninfected nymphs were placed within a second chamber posteriorly (skin site B) on two natural host species, yellow-necked field mice (Apodemus flavicollis) and bank voles (Clethrionomys glareolus). Virus transmission from infected to uninfected cofeeding ticks was correlated with infection in the skin site of tick feeding. Furthermore, virus was recruited preferentially to the site in which ticks were feeding compared with uninfested skin sites. Viremia did not correspond with a generalized infection of the skin; virus was not detected in an uninfested skin site (C) of 12/13 natural hosts that had viremia levels > or = 2.0 log10 ic mouse LD50/0.02 ml blood. To characterize infected cells, laboratory mouse strains were infested with infected ticks and then explants were removed from selected skin sites and floated on culture medium. Numerous leukocytes were found to migrate from the skin explants of tick feeding sites. Two-color immunocytochemistry revealed viral antigen in both migratory Langerhans cells and neutrophils; in addition, the migratory monocyte/macrophages were shown to produce infectious virus. The results indicate that the local skin site of tick feeding is an important focus of viral replication early after TBE virus transmission by ticks. Cellular infiltration of tick feeding sites, and the migration of cells from such sites, may provide a vehicle for transmission between infected and uninfected cofeeding ticks that is independent of a patent viremia. The data support the hypothesis that viremia is a product, rather than a prerequisite, of tick-borne virus transmission.

Dendritic cell loss from nonlymphoid tissues after systemic administration of lipopolysaccharide, tumor necrosis factor, and interleukin 1.

Dendritic cells (DC) in nonlymphoid organs can internalize and process foreign antigens before migrating to secondary lymphoid tissues to initiate primary immune responses. However, there is little information on which stimuli promote migration of DC from the tissues. Systemic administration of lipopolysaccharide (LPS), which induces in vivo production of cytokines, led to a reduction in the numbers of major histocompatibility complex class II-positive (Ia+) leukocytes in mouse hearts and kidneys: > 95% of DC were depleted 1-3 d after injection of 50 micrograms LPS. Several lines of evidence indicated that this response was due to migration of DC rather than loss of Ia expression or cytotoxic effects. In skin of treated mice, the number of Ia+ epidermal Langerhans' cells (LC) was reduced, and "cords" of Ia+ leukocytes became evident in the dermis. The latter cells expressed little NLDC145 and may have originated from recruited or resident DC progenitors. Systemic administration of recombinant tumor necrosis factor (rhTNF)-alpha resulted in a decrease in numbers of Ia+ cells in heart and kidney and of epidermal LC, and it also induced dermal cords. Administration of a rh-interleukin (IL)-1 resulted in a decrease in Ia+ cells only in renal medulla, appeared to activate a subset of epidermal LC, and induced dermal cords. Similar microgram doses of rhIL-2 had no obvious effect. Treatment with a neutralizing anti-TNF antiserum before LPS administration inhibited the depletion of LC from skin but not from heart or kidney. Therefore, TNF-alpha and IL-1 alpha may promote DC migration from nonlymphoid tissues and may have differential effects on different DC populations, but it is unclear whether they act on DC directly or indirectly (e.g., via other cytokines).

Developmental changes predispose the fetal thymus to positive selection of CD4+CD8- T cells.

Selection of a competent T-cell repertoire is dependent on complex interactions between immature thymocytes and components of the thymic stroma. These events may be preserved in vitro by excising developing thymus rudiments and maintaining them under carefully controlled conditions in fetal thymus organ cultures (FTOC). Using this approach, we have shown that the ability of C57B1/6 thymi to sustain positive selection of mature CD4+CD8- cells is profoundly influenced by the day of gestation on which they are excised: while thymocytes from day 14 rudiments fail to progress beyond the CD4+CD8+ stage of the developmental pathway, day 15 and day 16 thymi support the differentiation of CD4+CD8- thymocytes. Importantly, day 16 thymocytes transferred to day 14 deoxyguanosine-treated rudiments are likewise arrested at the CD4+CD8+ stage, suggesting that the thymic microenvironment of day 14 rudiments, rather than the state of differentiation of the thymocytes they contain, is responsible for the block in positive selection. Our studies of the stromal elements of day 14 rudiments have, however, revealed no obvious deficiencies in the cell types represented, or their expression of class II major histocompatibility complex (MHC) determinants. Furthermore, we have been unable to circumvent the blockage in positive selection by the addition of certain cytokines expressed late during gestation. These results suggest that subtle changes occurring at day 15 of ontogeny render the thymic microenvironment capable of positive selection.

Systemic lipopolysaccharide recruits dendritic cell progenitors to nonlymphoid tissues.

Dendritic cells (DC) are thought to be the "passenger leukocytes" that sensitize the recipients of organ transplants against graft antigens and trigger allograft rejection. DC originate from MHC class II-negative (Ia-) progenitors in the bone marrow, which enter the tissues and develop into migratory cells with the specialized capacity to initiate primary immune responses. There is little information on which stimuli recruit DC progenitors to the tissues. Systemic administration of LPS to mice depletes Ia+ leukocytes from heart and kidney but recruits Ia- leukocytes (Roake JA, et al., see footnote 6). When these leukocytes were isolated and cultured overnight, Ia+ low density leukocytes developed that could stimulate primary T cell responses in vitro. Hearts from LPS-treated mice were transplanted to allogeneic recipients. One to 4 days after grafting, Ia+ donor cells were present in recipient spleens, localized to peripheral white pulp, and associated with CD4+, but not CD8+, T cells. Cells with the migratory characteristics of DC, therefore, originated from Ia- progenitors in the transplanted hearts. We conclude that LPS recruits Ia- DC precursors to the heart and kidneys. Hearts from LPS-treated donors were rejected by allogeneic recipients at the same tempo as normal hearts, implying that Ia- DC progenitors might ultimately contribute to heart graft rejection (direct sensitization). However, since hearts from cyclophosphamide-treated donors, which do not give rise to Ia+ cells in recipient spleens, were also rejected at a similar tempo, indirect sensitization could also play a role in heart graft rejection in this model.

Isolation and characterization of dendritic cells from mouse heart and kidney.

Dendritic cells (DC) are thought to be distributed throughout lymphoid and most nonlymphoid tissues. Single cell suspensions were prepared from mouse hearts and kidneys. Subsets of MHC class II-positive (Ia+) leukocytes from both sources expressed markers such as CDw32 Fc receptors, F4/80, and complement receptor type 3 (CD11b/CD18). The capacity of these cells to initiate primary in vitro immune responses was assessed using oxidative mitogenesis and allogeneic mixed leukocyte responses. After fractionation by density centrifugation, cell sorting, immunomagnetic bead separation, or cell panning, the stimulatory activity of kidney cell suspensions was found to reside in the low density, Ia+ leukocyte fractions after overnight culture (day 1). In contrast, freshly isolated (day 0) cells had considerably less or no activity in these assays. However, depletion of Ia+ or CD45+ cells on day 0 followed by overnight culture removed the stimulatory activity on day 1. Therefore, day 0 kidney cells contain Ia+ leukocytes that can acquire or up-regulate their stimulatory activity during overnight culture. Similar observations were made for cells isolated from hearts, except that a population of uncharacterized nonleukocytes with stimulatory activity was detected on day 0 but not day 1. The phagocytic capacity of the leukocytes was then examined. Subsets of Ia+ cells phagocytosed zymosan, as shown by two-color flow cytometry and other immunofluorescence studies, and the zymosan-positive cells from kidney were able to initiate primary responses. Overall, these data demonstrate the existence of DC in kidneys and hearts, and suggest that in situ these cells resemble immature rather than mature DC.