Career development for women in academic medicine: Multiple interventions in a department of medicine.

OBJECTIVE: To determine the gender-based career obstacles for women in an academic department of medicine and to report the interventions to correct such obstacles (resulting from the evaluation) and the results of these interventions. DESIGN: Intervention study, before-after trial, with assessment of faculty concerns and perceived change through structured, self-administered questionnaires. SETTING: The Department of Medicine, The Johns Hopkins University School of Medicine, Baltimore, Md. PARTICIPANTS: Full-time faculty. INTERVENTIONS: Multifaceted intervention from 1990 through 1995 to correct gender-based career obstacles reported by women faculty, including problem identification, leadership, and education of faculty, and interventions to improve faculty development, mentoring, and rewards and to reduce isolation and structural career impediments. MAIN OUTCOME MEASURES: Retention and promotion of deserving women faculty, salary equity, quality of mentoring, decreased isolation from information and colleagues, integration of women faculty into the scientific community, and decreased manifestations of gender bias. RESULTS: Junior women were retained and promoted, reversing previous experience, with a 550% increase in the number of women at the associate professor rank over 5 years (from 4 in 1990 to 26 in 1995). Interim 3-year follow-up showed a 183% increase in the proportion of women faculty who expected they would still be in academic medicine in 10 years (from 23% [7/30] in 1990 to 65% [30/46] in 1993). One half to two thirds of women faculty reported improvements in timeliness of promotions, manifestations of gender bias, access to information needed for faculty development, isolation, and salary equity. Men also reported improvements in these areas. CONCLUSIONS: The outcomes reported here indicate that it is possible to make substantive improvements in the development of women's careers, that an institutional strategy to this end can be successful in retaining women in academic medicine, and that such interventions are likely to benefit all faculty. Long-term interventions appear essential.

The role of protein kinase C in transmembrane signaling by the T cell antigen receptor complex. Effects of stimulation with soluble or immobilized CD3 antibodies.

Phorbol esters, such as phorbol myristate acetate (PMA), are known to be potent co-stimulants with calcium ionophores for activation of T lymphocytes. The most extensively studied intracellular effect of PMA is its ability to activate the cytoplasmic enzyme protein kinase C (pkC). Herein, we examined the role of pkC activation during T cell activation. During physiologic activation, this enzyme is activated by diacylglycerol which is generated through the hydrolysis of polyphosphoinositides. Therefore, we studied the activation of T lymphocytes induced by a synthetic diacylglycerol, dioctanoylglycerol. In contrast to PMA, this compound can be metabolized in T cells and presumably more closely mimics physiologic activation of pkC. Dioctanoylglycerol together with reagents that induce increases in intracellular free Ca2+ concentration, Ca2+ ionophores, or anti-cluster designation (CD)3 monoclonal antibodies (mAb) were able to induce interleukin 2 receptor expression and proliferation of T lymphocytes. Previous studies have demonstrated that the stimulation of T cells via the CD3/T cell antigen receptor complex by mAb against CD3 leads to an increase in cytoplasmic free Ca2+ and to an activation of pkC. Paradoxically, however, soluble CD3 antibodies do not cause proliferation of resting purified T cells. Inasmuch as immobilization of CD3 mAb has been shown to influence the agonist properties of such antibodies, we compared the ability of soluble and immobilized CD3 mAb to activate pkC. We demonstrated herein that soluble CD3 mAb cause only a very transient activation of pkC in the T cell leukemic line Jurkat. This pkC activation is markedly prolonged when Jurkat cells are stimulated with immobilized rather than soluble CD3 antibodies. These studies suggest that activation of pkC plays a major role in T cell activation and that the activation of pkC is influenced by the form in which CD3 mAb is presented to T cells.

Molecular events involved in regulating human interferon-gamma gene expression during T cell activation.

The human T cell line Jurkat is a useful model of regulated T cell activation. After in vitro treatment of Jurkat with phytohemagglutinin (PHA) and phorbol ester (PMA), RNA transcripts of both interleukin 2 (IL 2) and interferon-gamma (IFN-gamma) appear, followed by secretion of both biologically active lymphokines. Employing the nuclear run-on technology, we first confirmed that the expression of both lymphokines after T cell activation is regulated at the transcriptional level. By using the recombinant approach of DNase I hypersensitivity mapping, we had previously localized a structurally unique, lymphocyte-specific genomic domain in the first intron of the human IFN-gamma gene that correlated with the transcriptional potential of that gene. By using several T cell lines that differ in their inducible expression of IFN-gamma, we have now localized several additional structural domains within the human IFN-gamma gene that appear to be coordinately involved in regulating expression. These include: a distal 5' flanking region site also seen only in T lymphocytes that can express the gene, a proximal, promoter-associated site that appears only after PHA/PMA-mediated IFN-gamma induction, and a second intronic site seen only in T cells whose IFN-gamma gene is selectively inactive. Collectively, our data suggest that T cell activation is accompanied by transcriptional level induction of lymphokine gene expression. In the case of IFN-gamma, T cell nuclei possess specific structural domains within the gene itself that seem to participate both positively and negatively in activation-mediated regulatory events.

Antinuclear antibodies in primary pulmonary hypertension.

The association of positive antinuclear antibodies with the clinical and hemodynamic features of 43 patients with primary pulmonary hypertension and 16 patients with secondary pulmonary hypertension was investigated. Each patient had determinations of antinuclear antibodies using a KB cell substrate immunofluorescent test. Of the patients with primary pulmonary hypertension, 40% had positive antinuclear antibodies at titers of 1:80 dilutions or greater. There were no differences between patients with primary pulmonary hypertension and positive antinuclear antibodies compared with those with negative antinuclear antibodies in relation to clinical or hemodynamic status. A 6% incidence rate of antinuclear antibodies was found in patients with secondary pulmonary hypertension, similar to that in the normal population. The clinical, hemodynamic, serologic and histologic similarity between patients with primary pulmonary hypertension and those with unexplained pulmonary hypertension associated with collagen vascular disorders suggests that primary pulmonary hypertension in some patients may represent a collagen vascular disease confined to the lungs. The frequency of positive antinuclear antibody tests would place primary pulmonary hypertension between rheumatoid arthritis and scleroderma in the spectrum of collagen vascular diseases. Further studies are necessary, however, before one might expect that immunosuppressive therapy would be beneficial to these patients.

Cholera toxin inhibits the T-cell antigen receptor-mediated increases in inositol trisphosphate and cytoplasmic free calcium.

The addition of monoclonal antibodies to the antigen receptor complex on the malignant human T-cell line Jurkat generates increases in inositol trisphosphate and in the concentration of cytoplasmic free calcium. Exposure of Jurkat cells to cholera toxin for 3 hr inhibited these receptor-mediated events and led to a selective, partial loss of the antigen receptor complex from the cellular surface. None of the effects of cholera toxin on the antigen receptor complex were mimicked by the B subunit of cholera toxin or by increasing intracellular cAMP levels with either forskolin or 8-bromo cAMP. These results suggest that a cholera toxin substrate can regulate signal transduction by the T-cell antigen receptor.

Differential effect of cyclosporin A on activation signaling in human T cell lines.

Different T cell lines, which can be induced to secrete interleukin 2 (IL-2) in vitro, were used to dissect the effect of cyclosporin A (CsA). The T leukemia cell Jurkat requires an increase in cytoplasmic calcium concentration ([Ca++]i) and phorbol myristate acetate (PMA) for the induction of IL-2 production, which is completely blocked by CsA. Another T cell line, HUT 78, also produces IL-2 in response to a rise in [Ca++]i and PMA; however, in HUT 78, PMA alone induces low levels of IL-2 production that is not blocked by CsA. After treatment with 5-azacytidine, HUT 78 cells produced maximal levels of IL-2 in response to PMA alone without requiring [Ca++]i increasing stimuli. In these cells no inhibitory effect of CsA on PMA-induced activation could be demonstrated. In addition, CsA does not inhibit PMA-induced translocation of protein kinase C. These data suggest that CsA does not globally inhibit IL-2 gene expression, but rather interferes with signaling events of T cell activation.

A transferrin receptor antibody represents one signal for the induction of IL 2 production by a human T cell line.

We previously demonstrated a two-signal requirement for the activation of the human T cell lines Jurkat and HUT 78. Interleukin 2 (IL 2) production by these lines can be induced by phytohemagglutinin (PHA), T3 antibodies, or calcium ionophores, but only in combination with phorbol myristate acetate (PMA). To obtain further information about surface structures involved in T cell activation, we produced a monoclonal antibody that could substitute for PMA in the activation of HUT 78. This antibody, designated J64, induced IL 2 secretion by HUT 78 in combination with PHA, T3 antibodies, or calcium ionophores, however not by itself. J64 also had other PMA-like effects on HUT 78, such as an increase in IL 2 receptor expression and an inhibition of cell growth. J64 was shown to immunoprecipitate the transferrin receptor (TfR). However, it bound to an epitope different from those recognized by other TfR antibodies and different from the transferrin-binding site. In addition, other previously described TfR antibodies did not, like J64, function as activating stimuli for HUT 78. Possible mechanisms for activation signaling in T cells involving the TfR are discussed.

Regulation of expression of the human interferon gamma gene.

DNA fragments isolated from a genomic clone of human gamma interferon (IFN-gamma) as well as IFN-gamma cDNA were used to map potential regulatory regions of the IFN-gamma gene by DNase I-hypersensitivity analyses. In nuclei from the human T-cell line Jurkat, which can be induced to express the IFN-gamma gene, we observed a strongly hypersensitive site in the first intervening sequence that localized to the only intracistronic repeat element in the gene. DNase I mapping of Jurkat cells was compared to that of several other cell types, including B cells, macrophages, and epithelial cells. The presence of strong intronic hypersensitivity was found only in cells capable of expressing the IFN-gamma gene. No hypersensitivity was found in the 3' regions of the gene. Further, no hypersensitivity was observed when purified genomic DNA from Jurkat was analyzed, suggesting that DNA-protein interactions, and not simply DNA sequence alone, were responsible for DNase I hypersensitivity. The sequence AAGTGTAATTTTTTGAGTTTCTTTT, which is directly in the intronic hypersensitive area of IFN-gamma, is 83% homologous to a nearly identical sequence in the 5' flanking region of the interleukin 2 gene. In interleukin 2, the homologous sequence is about 300 base pairs upstream of that gene's promoter in an area of potential regulatory importance.

T cell activation: differences in the signals required for IL 2 production by nonactivated and activated T cells.

Soluble antibodies against the T3/antigen receptor complex alone are not sufficient to induce proliferation and interleukin 2 expression by T lymphocytes. An additional requirement is the presence of accessory cells (AC). In this model, AC provide at least two functions required for T cell activation: 1) the surface interaction of T3 antibodies with Fc receptors on the AC surface and 2) the production of soluble mediators such as interleukin 1 (IL 1). In the experiments reported here, these stimuli are represented by T3 antibodies immobilized onto Sepharose beads and by recombinant IL 1. In this study we investigated differences in activation requirements in resting and activated T cells. Resting T cells were represented by AC-depleted peripheral blood mononuclear cells (PBMC) or the T cell line Jurkat, which phenotypically resembles a resting T cell. Activated T cells were represented by T cell clones and the T cell line HUT 78, which express the activation molecules Ia and the IL 2 receptor (Tac). In resting cells, activation required the presence of three different signals: perturbation of the T3/antigen receptor complex by T3 antibodies, surface redistribution of T3/antigen receptor complexes, and presence of IL 1. In contrast, activated T cells require only perturbation and redistribution of T3/antigen receptor complexes and not IL 1 for the induction of proliferation or IL 2 production. Possible mechanisms of intracellular signaling for these stimuli are discussed.