Hypoiodous acid as guest molecule in protonated water clusters: a combined FT-ICR/DFT study of I(H2O)n+.

Cationic water clusters containing iodine, of the composition I(H2O)n+, n = 0-25, are generated in a laser vaporization source and investigated by FT-ICR mass spectrometry. An investigation of blackbody radiation-induced fragmentation of size-selected clusters I(H2O)n+, n = 3-15, under collision-free conditions revealed an overall linear increase of the unimolecular rate constant with cluster size, similar to what has been observed previously for other hydrated ions. Above a certain critical size, I(H2O)n+, n greater than or approx. 13, reacts with HCl by formation of the interhalide ICl and a protonated water cluster, which is the reverse of a known solution-phase reaction. Accompanying density functional calculations illustrate the conceptual differences between cationic and anionic iodine-water clusters I(H2O)n+/-. While I-(H2O)n is genuinely a hydrated iodide ion, the cationic closed-shell species I(H2O)n+ may be best viewed as a protonated water cluster, in which one water molecule is replaced by hypoiodous acid. In the strongly acidic environment, HOI is protonated because of its high proton affinity. However, similar to the well-known H3O+/H5O2+ controversy in protonated water clusters, a smooth transition between H2IO+ and H4IO2+ as core ions is observed for different cluster sizes.

Time course of changes in cocaine self-administration behavior in rats during immunization with the cocaine vaccine IPC-1010.

RATIONALE: Following a 6-week immunization period consisting of three biweekly injections of the cocaine vaccine IPC-1010, the reacquisition of cocaine self-administration behavior in rats was previously shown to be reduced in a manner that was dependant on serum antibody level. The present studies were conducted to examine additional issues relevant to the clinical use of the vaccine. OBJECTIVES: One experiment was conducted to address the issue of whether exposure to cocaine during the immunization period would influence the ability of the vaccine to block cocaine self-administration. A second experiment was conducted to determine if the reductions in drug-seeking behavior and drug intake by the vaccine were behaviorally specific, or if behavior maintained by a non-drug reinforcer would be similarly affected. METHODS: Identical second-order schedules of cocaine (1 mg/kg) or food pellet (45 mg) delivery were used in rats. In both studies, the time course of changes in behavior during the 6-week immunization period was examined in vaccine and alum-treated control rats following baseline and extinction conditions. RESULTS: The cocaine vaccine IPC-1010 induced average serum antibody levels of 0.07 mg/ml and significantly reduced self-administration behavior during the 2-week period following the third vaccine boost in a subgroup of rats with serum antibody levels greater than the average value. Cocaine self-administration behavior at this time point significantly correlated with serum antibody level. IPC-1010 did not alter responding maintained by food throughout the immunization period although serum antibody levels reached a similar average of 0.06 mg/ml in this group of rats. CONCLUSIONS: These findings suggest that the reductions in drug-seeking behavior and drug intake after immunization with IPC-1010 did not result from a reduced ability of the rats to respond on the lever. Furthermore, daily exposure to cocaine during the immunization period did not influence the ability of the vaccine to reduce cocaine self-administration behavior that emerged gradually over time. These findings also confirm the need for a sufficiently high antibody level to blunt the reinforcing effects of cocaine.

Evaluation of anti-cocaine antibodies and a cocaine vaccine in a rat self-administration model.

RATIONALE: Previous pre-clinical studies with an anti-cocaine monoclonal antibody left open several issues critical to assessing the effectiveness of a vaccine for altering cocaine self-administration behavior. OBJECTIVES: The objectives of this study were to determine, first, whether changes in self-administration behavior would be systematically related to antibody level and, second, how the antibody affected the self-administration of different doses of cocaine. METHODS: Two experiments were conducted using a second-order schedule of drug delivery in rats. The first was a passive-administration study using the anti-cocaine monoclonal antibody MO240 to examine the relationship between antibody level and cocaine self-administration behavior, and the second was an active-immunization study to examine the efficacy of the cocaine vaccine IPC-1010 for blocking various doses of self-administered cocaine. RESULTS: The passive-administration experiment with control and 4-mg or 12-mg MO240 treatments showed that antagonism of the 1 mg/kg cocaine training dose was dependent on antibody level. In animals whose serum antibody levels were sustained above 0.05 mg/ml, there was a sufficient amount of antibody to reduce drug-seeking behavior and drug intake. In the active-immunization experiment, the cocaine vaccine IPC-1010 induced average serum antibody levels of 0.08 mg/ml and reduced the reacquisition of behavior by 1 mg/kg cocaine. Antagonism of cocaine self-administration after immunization was evident across a range of doses of cocaine and was only apparent in animals whose serum antibody levels exceeded 0.05 mg/ml. Furthermore, there was no evidence that the antagonism was surmountable within the dose range examined (up to 5.6 mg/kg). CONCLUSIONS: Antagonism of cocaine self-administration across a range of doses is feasible after immunization with a cocaine vaccine as long as antibody levels are of a sufficient concentration.

Tissue engineering of bone. Cell based strategies.

Skeletal tissue regeneration requires the interaction of three basic biologic elements: cells, growth and differentiation factors, and extracellular matrix scaffolds. Therapeutic approaches for tissue engineered repair of bone defects have attempted to mimic the natural process of bone repair by delivering a source of cells capable of differentiating into osteoblasts, inductive growth and differentiation factors, or bioresorbable scaffolding matrices to support cellular attachment, migration, and proliferation. Sophisticated designs even have tried to combine two or more of these elements. The development of cell based approaches has advanced dramatically in recent years as an understanding of musculoskeletal cell biology improves. Cell based approaches do not depend on the presence of local osteoprogenitors for the synthesis of new bone and, as a result, they particularly are attractive for patients who have a diminished pool of these progenitors, or in whom the host tissue bed has been compromised. This review highlights the development of cell based approaches for the tissue engineering of bone, and offers perspectives on the optimal elements for success. Although logistical and regulatory issues remain to be solved, cell based therapies for the repair of clinically significant bone defects rapidly are approaching clinical feasibility.

Development of a therapeutic vaccine for the treatment of cocaine addiction.

No pharmacotherapies have yet been approved for the treatment of cocaine addiction. One new approach is to block the effects of cocaine with anti-cocaine antibodies induced by a therapeutic cocaine vaccine. A cocaine vaccine has been developed which induces a cocaine-specific antibody response in rodents. The antibody binds to cocaine in the circulation and can be shown to inhibit the ability of cocaine to enter the brain. Furthermore, anti-cocaine antibody can inhibit cocaine self-administration in rats. These data suggest that a cocaine vaccine may be a powerful therapeutic tool. The intent is to immunized motivated patients with the vaccine as part of a comprehensive treatment program. If the patient uses cocaine after being vaccinated, the antibody will inhibit the reinforcing activity of cocaine and decrease the likelihood of relapse.

Cholera toxin B subunit binding to an antigen-presenting cell directly co-stimulates cytokine production from a T cell clone.

Cholera toxin (CT) is a powerful immunomodulator with strong adjuvant activity. Much of this activity is retained by the binding component alone, cholera toxin B subunit (CTB). Little is known about the mechanism of the immunomodulatory activity of CTB. In this study, both CT and CTB were found to dramatically enhance IL-4 production from a T cell clone stimulated with antigen and the B cell hybridoma LB as antigen-presenting cell (APC). Enhancement of cytokine production was seen following pretreatment of the APC with CT or CTB, while pretreatment of the T cells had no effect. Furthermore, stimulatory activity on the APC was stable to fixation with paraformaldehyde, demonstrating that the activity was mediated by a surface molecule on the APC. CT-pretreated APC also enhanced IL-4 production from anti-CD3 mAb-stimulated T cells, indicating that CT was providing a co-stimulatory signal. CT treatment of LB cells did not alter the expression of class II MHC molecules, CTLA-4 counter-receptors, LFA-1 or ICAM-1. When mAb were raised against the CT-pretreated APC, the only antibodies that were found to inhibit IL-4 production were those specific for CTB itself. The antibodies blocked even when the CT or CTB were already bound to the APC, arguing that co-stimulation was provided by a direct interaction with CTB. Blocking experiments suggested that APC-associated CTB molecules are interacting with non-GM1 receptors on the T cells. This novel finding of CTB-mediated T-B interaction provides one of the first potential mechanisms for the adjuvant activity of CTB.

Efficacy of a therapeutic cocaine vaccine in rodent models.

Cocaine abuse is a major medical and public health concern in the United States, with approximately 2.1 million people dependent on cocaine. Pharmacological approaches to the treatment of cocaine addiction have thus far been disappointing, and new therapies are urgently needed. This paper describes an immunological approach to cocaine addiction. Antibody therapy for neutralization of abused drugs has been described previously, including a recent paper demonstrating the induction of anti-cocaine antibodies. However, both the rapidity of entry of cocaine into the brain and the high doses of cocaine frequently encountered have created challenges for an antibody-based therapy. Here we demonstrate that antibodies are efficacious in an animal model of addiction. Intravenous cocaine self-administration in rats was inhibited by passive transfer of an anti-cocaine monoclonal antibody. To actively induce anti-cocaine antibodies, a cocaine vaccine was developed that generated a high-titer, long-lasting antibody response in mice. Immunized mice displayed a significant change in cocaine pharmacokinetics, with decreased levels of cocaine measured in the brain of immunized mice only 30 seconds after intravenous (i.v.) administration of cocaine. These data establish the feasibility of a therapeutic cocaine vaccine for the treatment of cocaine addiction.

Cytokine production by T lymphocytes from young and aged mice.

We previously have shown that T cell proliferation in response to a primary signal through the CD3 epsilon chain and a costimulatory signal via the CD28 molecule is impaired in healthy, aged mice. To determine whether age-related alterations in cytokine production might explain the reduced proliferative responses of T cells from aged mice, we examined the secretion of the major T cell immunoregulatory cytokines, IFN-gamma, IL-4, and IL-2. Splenic T cells from young (2 to 4 mo) and aged (20 to 26 mo) mice were studied. T cells were stimulated with immobilized anti-CD3 epsilon chain mAb and soluble anti-CD28 mAb for 24 h. T cells from aged mice, when compared with young controls, showed increased IFN-gamma production, no difference in IL-4 production, and decreased IL-2 production. Most IFN-gamma was produced by CD8+ T cells, whereas most IL-2 and IL-4 was produced by CD4+ T cells. Both CD4+ and CD8+ T cells from aged mice produced significantly more IFN-gamma than corresponding cells from young mice. This increased production could be accounted for by increased numbers of CD4+CD44high and CD8+CD44high T cells in aged animals. CD4+CD44high and CD8+CD44high T cells from young mice produced comparable amounts of IFN-gamma as corresponding cells from aged mice. In contrast to unseparated splenic T cells, no age-related difference in IL-2 or IL-4 production by purified CD4+ T cells was observed. Similarly, when CD4+ T cells were further separated into CD44low and CD44high subpopulations, no age-related difference in IL-2 production was found. Therefore, we found no consistent evidence that diminished production of the major T cell growth factors, IL-2 and IL-4, is responsible for the age-related decrease in the proliferation of T cell subpopulations that were stimulated in vitro through the CD3 epsilon chain and costimulated via the CD28 molecule. The physiologic relevance of increased IFN-gamma production by T cells from aged mice is unknown.

An age-related decrease in rescue from T cell death following costimulation mediated by CD28.

We previously reported that T cell proliferation in response to a primary signal through the T cell receptor (TCR) and a costimulatory signal via the CD28 molecule is impaired in healthy, aged mice. Here we extend these studies to examine factors which may be involved in this defect in T cells from aged mice. To determine if age-related changes in cytokine production might be responsible, splenic T cells from young (2-4 months) and aged (20-26 months) mice were stimulated with immobilized anti-CD3 epsilon and soluble anti-CD28 mAbs in the presence of exogenous IL-2, IL-4, IFN-gamma, IL-1 alpha, or IL-6. No improvement in the proliferative response of T cells from aged mice was found following the addition of any cytokine. In addition, the decreased proliferative response of T cells from aged mice was not caused by the enhanced production of IFN-gamma or other inhibitory factors. Interestingly, despite the age-related reduction in proliferation, no significant difference was found in the percentage of live cells entering the S, G2, or AM phase of the cell cycle in stimulated T cells from young and aged mice. Instead, anti-CD28-mediated costimulation was found to rescue T cells from young mice from anti-CD3epsilon-induced cell death, but did not rescue T cells from aged mice. This failure of T cells from aged mice to respond to costimulatory signals appears to contribute to the decreased proliferation observed from cultures containing these cells, and may be involved in many other age-related alterations in immunological responsiveness.

Molecular analysis of the same HIV peptide functionally binding to both a class I and a class II MHC molecule.

Although several peptides have been found to bind to both class I and class II molecules, the basis for this binding of the same peptide to two classes of MHC molecules has not been compared previously. We have analyzed one such peptide, P18 from the V3 loop of HIV-1 gp160, which we have previously shown to be recognized by CD8+ CTL with the class I molecule H-2Dd, and by CD4+ Th cells with the class II molecule I-Ad. With the use of truncated and substituted peptides, we found that the minimal core peptides are very similar, that the residues required for class I binding precisely fit the recently identified consensus motif for peptides binding to Dd (XGPX[R/K/H]XXX(X) [L/I/F]), and that at least three of the same residues are involved in binding to class II I-Ad. In addition, several of the same residues are involved in TCR interaction when the peptide is presented by class I and class II molecules. Modeling shows results to be consistent with the crystal structure of a peptide-class II MHC complex. Thus, the recognition of this versatile peptide by CD4+ Th cells with class II MHC molecules and by CD8+ cytotoxic T cells with class I MHC molecules is remarkably similar in both the core peptide used and the role of different residues in the ternary complex.

Aged T cells are hyporesponsive to costimulation mediated by CD28.

The ability of T cells to proliferate in response to a number of stimuli is impaired in healthy, aged individuals. T cells from young (2 to 4 mo) and aged (20 to 26 mo) mice were stimulated with immobilized anti-CD3 epsilon-chain mAb and soluble anti-CD28 mAb. T cells from young and aged mice proliferated comparably in response to anti-CD3 epsilon mAb alone. However, aged T cells were significantly less responsive to costimulation by anti-CD28 mAb. This decreased response of aged T cells was seen at all dosages tested of anti-CD3 epsilon and anti-CD28 mAbs and in the presence and absence of APC. Similar results were observed when costimulation was provided by B7-transfected L cell fibroblasts. T cells from young and aged mice had comparable expression of CD28 by flow cytometry, both before and after stimulation. The defect in response to CD28 was seen both in CD4+ and CD8+ T cells and in CD44lo (naive) and CD44hi (memory) T cells. The impaired response to costimulation mediated by CD28 on T cells from aged mice may be an important factor in reduced T cell responses associated with aging.

CD4+ class I-restricted T cells specific for HIV gp160 315-329.

Mature, circulating, alpha beta T cells express either CD4 or CD8. In the majority of cases, CD4+ T cells recognize Ag in association with class II MHC molecules, while CD8+ T cells recognize Ag in association with class I MHC molecules. In this report we describe CD4+ class I-restricted T cell hybridomas and normal clones specific for the peptide 315-329 of HIV gp160 strain IIIb in association with H-2Dd. Two models were formulated to explain how CD4+ class I-restricted T cells could arise. First, they could represent aberrant selection of CD4+ cells on class I MHC molecules in the thymus. Alternatively, they could have been selected normally on class II; in this case the cells would display cross-reactive recognition of 315-329 in association with H-2Dd and an unknown Ag in association with class II. To distinguish these models, a second specificity was determined for the T cell clones. The normal clones recognized the class I molecule H-2Kk as alloantigen and thus were presumably positively selected in the thymus on class I MHC. CD4 was shown to be functional in these cells in that anti-CD4 mAb inhibited their proliferation: however, both Ag- and Con A-induced responses were inhibited, suggesting that a negative signal was delivered by the anti-CD4 mAb.

Antigen presenting cell-derived co-stimulatory signals can selectively regulate IL-2 and IL-4 production from a Th0 cell hybridoma.

Th0 cells are a subpopulation of Th cells that produce both IL-2 and IL-4. In order to study the selective regulation of lymphokines in Th0 cells, a Th0 cell hybridoma, GA15, was generated by fusing a Th2 clone with the thymoma BW5147. The pattern of lymphokines secreted by GA15 was found to depend on the antigen presenting cells (APC) used for stimulation. When GA15 was stimulated with antigen and splenocytes, both IL-2 and IL-4 were produced. However, when the B cell hybridoma LB was used as APC, only IL-2 was detected. Although LB cells could not stimulate IL-4 production, they were more potent than were spleen cells at inducing IL-2 production, demonstrating that the difference between the two APC populations was qualitative rather than quantitative. Similar results were seen upon stimulation with concanavalin A or anti-CD3 epsilon mAb. Regulation of interferon-gamma production paralleled regulation of IL-4 production. The failure of LB cells to stimulate IL-4 production was not due to inhibition or consumption of IL-4 and the activity could not be restored by adding IL-1, spleen cell supernatants or spleen cells lacking the appropriate MHC molecule. Finally, the parent Th2 clone used in the fusion showed a similar inability to produce IL-4 in response to LB cells. These data indicate that APC-derived co-stimulatory signals can selectively affect IL-2 and IL-4 production in a Th0 cell hybridoma. Therefore, the choice of APC may lead to selective stimulation of particular lymphokines by acting on Th0 cells.

Effect of prostaglandin E2 (PGE2) on IL-3/granulocyte-macrophage colony-stimulating factor production by T helper cells. Mode of stimulation and presence of costimulation can determine response to PGE2.

PGE2 is an immunomodulator that selectively inhibits the production of lymphokines associated with Th1 cells (IL-2 and IFN-gamma) but not Th2 cells (IL-4 and IL-5). We examined the effect of PGE2 on the production of IL-3 and granulocyte-macrophage (GM)-CSF from murine Th1 and Th2 clones. When the T cells were stimulated with Ag and APC, PGE2 inhibited IL-3/GM-CSF production from 3 Th1 clones and 1 Th2 clone, but enhanced production from 3 Th2 clones. A more specific bioassay demonstrated that IL-3 production was differentially affected by PGE2 in the Th clones. These data suggested that the effect of PGE2 on IL-3 production is dependent, not on a property of the lymphokine, but on a property of the T cell. The responsiveness to PGE2 did not consistently differ between Th1 and Th2 cells, and the observed heterogeneity in the response of Th2 clones did not correlate with the ability to induce increases in intracellular [Ca2+]. However, we postulated that signaling differences between the clones might explain the varied responsiveness to PGE2. If so, then the mode of stimulation might be expected to activate different pathways and thus affect the PGE2-responsiveness. Stimulation with ionomycin induced variable levels of IL-3/GM-CSF from the T cell clones. APC-derived costimulation dramatically enhanced IL-3/GM-CSF; the cells which produced high levels in response to ionomycin alone were not detectably costimulating each other. Interestingly, PGE2 enhanced IL-3/GM-CSF (and IL-3 alone in at least some cases) from cells stimulated with ionomycin alone, demonstrating that the mode of stimulation affects the PGE2-responsiveness. Addition of APC not only enhanced lymphokine production, but also altered the PGE2-responsiveness of the Th1 cells. In these cells, PGE2 either inhibited IL-3 and GM-CSF production or had no effect, in no case was the lymphokine production enhanced by PGE2 as it had been with ionomycin alone. These data indicate that the presence of APC-derived costimulatory signals can alter the effect of PGE2 on Th cell lymphokine production.

Immunization of mice with human immunodeficiency virus glycoprotein gp160 peptide 315-329 induces both class I- and class II-restricted T cells: not all T cells can respond to whole molecule stimulation.

The V3 loop of human immunodeficiency virus (HIV) glycoprotein gp160 is of interest as a possible site for protective immune responses. This article examines the murine T cell response to peptide 315-329 derived from HIV gp160. Surprisingly, immunization with peptide in complete Freund's adjuvant induced class I-restricted T cells as well as class II-restricted T cells. These data suggest that this peptide may have the unusual ability to enter the class I antigen processing pathway. Strategies that employ V3 loop peptides to induce protective immunity must generate T cells that can recognize epitopes derived from whole molecules in vivo. Therefore, peptide-induced T cells were tested for their ability to respond to naturally processed forms of gp120 and gp160 whole-molecule preparations. Peptide induced class I-restricted cells were capable of recognizing transfectants expressing gp160. However, only one of two class II-restricted T cell lines was capable of recognizing soluble whole molecules. This indicates that peptide immunization induces T cells that recognize a class II-restricted determinant that is not generated during normal processing of whole molecules. We have also examined the response of peptide primed T cells to lipidated peptide antigens. Lipidated peptides are generally considered to have increased antigenicity and immunogenicity as compared to normal peptides. However, lipidation of peptide 315-329 damaged both the class I- and II-restricted determinants, indicating that lipidation is not always desirable. The data presented here highlight a potential serious problem in the use of peptide vaccines, in that peptide immunization may not always induce T cells that can protect against a viral challenge.

Pattern of lectin binding to murine T lymphocytes.

Lectins can be used to detect changes in cell-surface glycosylation. In this paper we examine the lectin-binding characteristics of murine T cells as measured by flow cytometry. A large panel of labelled lectins was used to stain naive, activated and resting murine T cells. Some lectins did not detectably bind any T cells, some bound to all T cells and some bound to only a subset of splenic T cells. Three lectins which preferentially bind to previously activated T cells were identified: Bandeiraea simplicifolia BS-I, Bauhinia purpurea and Lycopersicon esculentum. In addition, two lectins were found to bind preferentially to activated T cells: Glycine max and Triticum vulgaris. Finally, no differences were found in the ability of a large panel of lectins to bind to Th1 and Th2 clones. Lectin binding may be a powerful tool for distinguishing naive, activated and memory T cells.

Antibody responses to a cytochrome c peptide do not correlate with lymphokine production patterns from helper T-cell subsets.

This paper examines helper T-cell responses and antibody titres and isotypes following immunization with a peptide antigen in association with three different adjuvants. B10.A mice were primed with pigeon cytochrome c fragment 81-104 in association with the adjuvants complete Freund's adjuvant (CFA), incomplete Freund's adjuvant (IFA) and alum. Strong antibody responses, dominated by IgG1, were observed upon priming with CFA and IFA. In contrast, priming with alum induced a weak antibody response with little or no detectable antigen-specific IgG1. These differences did not correlate with differences in T-cell priming, as immunization with peptide in association with all three adjuvants induced comparable T-cell proliferative responses and frequencies of antigen-specific cells. In addition, no significant differences in interleukin-2 (IL-2), interferon-gamma (IFN-gamma) and IL-4 production could be found, suggesting that the adjuvants did not differentially affect Th1 and Th2 cells.

Prostaglandin E2 inhibits production of Th1 lymphokines but not of Th2 lymphokines.

PGE2 is known to inhibit IL-2 and IFN-gamma production from Th cells and is widely viewed as a general immunosuppressant. However, PGE2 was found not to inhibit IL-4 production from Th2 clones, and IL-5 production from these clones was slightly enhanced. The same results were obtained with short term T cell lines, which indicates that the lack of inhibition of IL-4 and IL-5 production by PGE2 is a general phenomenon. PGE2 functions by increasing cAMP levels through activation of adenylate cyclase. Despite its failure to inhibit lymphokine release, PGE2 was capable of increasing cAMP levels in Th2 cells, and forskolin, a direct activator of adenylate cyclase, also did not inhibit IL-4 or IL-5 production. These data indicate that the failure of PGE2 to inhibit IL-4 and IL-5 production was not due to an inability of PGE2 to induce an increase in intracellular cAMP, and suggested instead that the expression of IL-4 and IL-5 in Th2 cells is insensitive to elevated cAMP levels. When Th0 clones were examined, PGE2 was again found to differentially affect IL-2 and IL-4 production in three of five clones tested. In two additional Th0 clones, both IL-2 and IL-4 production were inhibited. These data suggest that lymphokine production may be regulated on two different levels. First, Th1- and Th2-associated lymphokines may be differentially sensitive to intracellular signals such as cAMP. Second, T cell subsets may exist, including subsets of Th0 cells, with different signaling pathways. In addition, our data suggest that PGE2 may play an important role in regulating the development of a response dominated by Th1- or Th2-associated lymphokines.

Regulation and development of cytochrome c-specific IL-4-producing T cells.

The development of Ag-specific IL-4-producing T cells in short term cultures was examined. Freshly harvested lymph node cells from B10.A mice immunized with the cytochrome c fragment 81-104 did not produce detectable amounts of IL-4 upon Ag stimulation. However, after one cycle of bulk in vitro restimulation, the same cells could be stimulated to secrete IL-4. The development of Ag-specific IL-4-producing cells during the in vitro culture could be influenced by different culture conditions. When IL-4 and IL-1, in the presence of anti-IL-2R antibodies, were added to the bulk culture, secondary stimulation of the cells revealed increased levels of IL-4 production and constant or decreased levels of IL-2 production. The addition of IFN-gamma during the bulk culture led to a decrease in IL-4 production, yet had no effect on IL-2 production. When anti-IL-4 antibodies were present during the restimulation culture, no IL-4 was produced in the final stimulation assay. In addition, T cells cultured at high cell density produced more IL-4 in a secondary stimulation than did T cells cultured at low cell density. These results demonstrate that culture conditions during a short term culture of freshly harvested primed T cells may profoundly influence the development of IL-4-producing cells. This may be caused either by selective expansion or inhibition of preexisting IL-4-producing T cells, or by the differentiation of precursors of IL-4-producing cells.

In vivo priming of helper T cells in the absence of B cell activation.

This paper describes an adjuvant-free immunization regimen that results in the priming of T cells but not B cells. B10.A mice were primed s.c. with syngeneic spleen cells that had been pulsed with the peptide 81-104 derived from pigeon cytochrome c. The T cell response was measured by using a sensitive limiting dilution assay that measures lymphokine production. The precursor frequency of Ag-specific cells found in these mice was indistinguishable from the frequency found in mice primed in the footpads with 81-104 in CFA. A striking difference in antibody induction was found, however, when these two immunization regimens were compared. Mice primed with 81-104 in CFA developed significant serum antibody responses against the peptide, whereas mice primed with Ag-pulsed spleen cells produced no detectable anti-peptide antibodies. This lack of antibody did not result from detectable differences in the T cells that were primed: no differences were seen in IL-2 and IL-4 production or in the ability to provide help to B cells in vitro. In vitro stimulation with LPS suggested that the B cells were not primed by the Ag-pulsed spleen cells. The B cells were not tolerized, however, because boosting the mice with Ag in CFA resulted in the induction of an antibody response. The failure to induce an antibody response by priming with Ag-pulsed spleen cells was not caused by the site of immunization or the total amount of Ag used for priming. The critical variable may be the introduction of the Ag on the surface of an APC; in this form, B cell Ag recognition was apparently inefficient, whereas T cell Ag recognition was optimal.