Specialty differences in the management of asthma. A cross-sectional assessment of allergists' patients and generalists' patients in a large HMO

OBJECTIVE

To examine the differences in medical management and quality of life between patients with asthma who receive their primary asthma care from allergists and those who receive their care from generalists in a large health maintenance organization (HMO).

METHODS

We conducted a cross-sectional study of patients with asthma in a large HMO (Kaiser Permanente, Northwest Region, Portland, Ore). Participants were 392 individuals aged 15 through 55 years with physician-diagnosed asthma, taking antiasthma medications, reporting current asthma symptoms, and receiving asthma care from an allergist or from a generalist. Primary outcomes include characteristics of asthma, health care utilization, and quality of life.

RESULTS

Patients cared for by allergists tended to have more severe asthma than those cared for by generalists (P < .01). The allergists' patients tended to be older (38.6 +/- 9.6 years vs 35.7 +/- 12.6 years, P < .01), more atopic (91% vs 78%, P < .01), and more likely to report perennial (rather than seasonal) asthma (26% vs 36%, P < .04) than the generalists' patients. Patients receiving their primary asthma care from an allergist were considerably more likely than generalists' patients to report using inhaled anti-inflammatory agents (P < .01), oral steroids (P < .01), and regular (daily) breathing medications to control their asthma (P < .01). Allergists' patients were more likely to have asthma exacerbations treated in a clinic setting rather than an emergency department (P < .01). Furthermore, allergists' patients reported significantly improved quality of life as measured by several dimensions of the SF-36 scale (physical functioning, role emotional, bodily pain, and general health; P < .05).

CONCLUSIONS

These findings suggest that specialist care of asthma is of benefit for patients with asthma in a large HMO. Specifically, the allergists' patients conformed more closely to national asthma management guidelines and reported better quality of life than did the generalists' patients.

Synovial distribution of alpha d/CD18, a novel leukointegrin. Comparison with other integrins and their ligands

OBJECTIVE

To define the synovial distribution of the novel leukointegrin alpha d/CD18, and compare this with other members of the beta 2-integrin family of adhesion molecules, and their counter-receptors.

METHODS

Monoclonal antibodies to the CD3, CD14, CD29, CD68, beta 2-integrin, and immunoglobulin supergene families were used to immunohistologically define the distribution of these molecules in synovial tissue samples from normal subjects and osteoarthritis (OA) and rheumatoid arthritis (RA) patients.

RESULTS

The normal synovial lining cell layer (SLC) expresses CD68, vascular cell adhesion molecule 1, beta 1-integrin (CD29), the beta 2-integrins CD11b/CD18 (alpha m/beta 2, Mac-1), and alpha d/CD18, whereas CD11a/CD18 (alpha L/beta 2, lymphocyte function-associated antigen 1) and CD11c/CD18 (alpha x/beta 2, gp150/95) expression is generally absent. In RA synovitis, expression of beta 2-integrins in the SLC increases in proportion to the degree of hyperplasia. The ratio of cells in the SLC which express CD11c/CD18 increases substantially, approaching that of CD11b/CD18 and alpha d/CD18, while there is minimal increase in CD11a/CD18 expression. In the sublining areas of the tissues, aggregates and diffuse infiltrates of CD3/CD11a/ICAM-3+ lymphocytes are interspersed among CD68/CD14/CD11b/alpha d+ macrophages. A number of aggregates demonstrate intense alpha d staining of the lymphocytes. The synovial endothelium variably expresses intercellular adhesion molecule-1 (ICAM-1), ICAM-2, and vascular cell adhesion molecule 1 (VCAM-1), with minimal evidence of ICAM-3 expression.

CONCLUSION

The leukointegrin alpha d/CD18 is expressed constitutively by synovial macrophages and macrophage-like lining cells. In rheumatoid synovitis, the intense coexpression of this integrin and its known counter-receptor, ICAM-3, in the inflammatory infiltrates, suggests a potential role for this adhesion pathway in cellular interactions occurring the synovium.

Beta 2-integrin/intercellular adhesion molecule (ICAM) expression in the normal human intestine

The beta 2-integrin family of adhesion molecules and their ligands, the ICAM molecules, probably have an important role to play in the intestinal immune response. To date, any data published regarding the intestinal expression of beta 2-integrins and ICAMs have mostly described only mucosal expression. The presence in the intestine of alpha d, a newly described beta 2-integrin molecule, has not yet been determined. To understand further the expression of these molecules in all layers of the intestine, a comprehensive study was undertaken in both normal colon and normal small intestine specimens from 15 patients (colon n = 15, and ileum n = 6) who underwent surgical resections. Immunohistochemistry was performed for CD18, CD11a, CD11b, CD11c, alpha d, ICAM-1, ICAM-2, and ICAM-3. The colon and small intestine cases were grouped separately and each of the mucosal, submucosal, muscle and adventitial layers were scored for expression. The beta 2-integrin molecules were expressed by mononuclear cells in all layers of the bowel, predominantly in the mucosa and adventitia. CD11a showed the greatest mucosal expression and CD11b showed the greatest adventitial expression. alpha d exhibited expression in all bowel layers, as well, with most intense expression in the mucosa and adventitia. CD11c exhibited the least expression of all alpha subunit molecules. The expression of these alpha-chains at times appeared to be greater than that seen for CD18 (beta 2), raising the possibility of non-beta 2-containing complexes (i.e. alpha-chains associating with an alternate beta-chain). ICAM-1 was expressed on endothelium, particularly in the mucosa and rarely on mucosal mononuclear cells. ICAM-2 was predominantly on submucosal endothelium and rarely expressed in colon mucosa and never expressed in ileal mucosa. ICAM-3 was expressed by mononuclear cells throughout the bowel wall, but also on adventitial endothelium in selected cases. In summary, we are reporting a comprehensive immunohistochemical study of the differential expression of beta 2-integrins, including the newly described alpha d molecule, and the ICAM molecules in all layers of normal human colon and ileum. We raise the possibility that a second beta 2-integrin beta subunit may exist, and we report that ICAM-3 is expressed on endothelium, particularly in the adventitial layer.

The differential expression of a novel intestinal epithelial glycoprotein in various forms of inflammatory bowel disease

A13D8 is a monoclonal IgM antibody that identifies an as yet unknown antigen that is expressed intensely and ubiquitously in enterocytes. Immunohistochemically, it was shown that A13D8 has a granular supranuclear staining pattern in columnar epithelial cells of normal small intestine and the colon. In ulcerative colitis, this staining pattern was retained. However, during active inflammation, staining also was evident in goblet cells. To test whether this feature of goblet cell staining was unique to ulcerative colitis, tissue sections from a variety of colitides were examined. Crohn's disease, infectious colitis, and ischemic colitis had similar staining patterns to that seen with ulcerative colitis. There was significantly more inflammation in the biopsies from patients with ulcerative colitis and Crohn's disease with positive goblet cell staining than in the biopsies from those patients with negative goblet cell staining. Almost all positive goblet cell staining in ulcerative colitis and Crohn's disease occurred in biopsies that were actively inflamed, whereas there was rare staining in biopsies that were noninflamed (regardless of whether or not there was active inflammation elsewhere in the colon). Ileal goblet cells stained positively with A13D8 only in cases of active ileitis. In cases of collagenous colitis, with comparable degrees of inflammation to that seen in ulcerative colitis and Crohn's disease, there was rarely goblet cell staining and in graft-versus-host disease goblet cell staining of A13D8 was not observed. The binding of A13D8 to tissue sections was completely inhibited by N-acetyl-D-galactosamine. These results, in conjunction with immunochemical studies, suggest that the antibody recognizes an N-acetyl-D galactosamine-containing epitope on a glycoprotein(s). In conclusion, these data suggest that A13D8 recognizes a glycoprotein expressed by intestinal columnar epithelial cells and during specific inflammatory states, particularly those associated with a neutrophilic infiltrate, becomes evident in goblet cells. Further work is required to establish the exact nature of this molecule and whether it is a pro- or anti-inflammatory factor.

The correlation between neuropsychological and neuroanatomic changes over time in asymptomatic and symptomatic HIV-1-infected individuals

To determine the relationship between neuroanatomic and neuropsychological changes in both asymptomatic and symptomatic HIV-1-infected individuals, we conducted a longitudinal study of 47 HIV-infected individuals, 15 of whom were asymptomatic and 32 of whom had either AIDS-related complex or AIDS. To measure neuroanatomic change over a 30-month period, we conducted quantitative MRI measures of bicaudate/brain ratio (BCR) and bifrontal/ brain ratio. A comparison of change over time between BCR and neuropsychological performance showed a correlation between increase in atrophy and worsening in certain cognitive functions. The correlation held for both asymptomatic and symptomatic groups, with more pronounced changes in the symptomatic group.

Urine toxicology as an outcome measure in drug abuse clinical trials: must every sample be analyzed?

Clinical trials designed to establish the effectiveness of a pharmacotherapy for the treatment of drug abuse typically call for the collection and analysis of three urine samples per week to detect changes in drug use patterns. Examination of over 16,500 urine samples collected from 225 subjects during a one year buprenorphine/methadone clinical trial indicates that analysis of one weekly urine sample from those collected on a three-times-per-week fixed schedule provides essentially the same outcome information as analysis of all three weekly urines. Further, the percent of opiate-positive samples is constant across weekday, indicating that a single urine, randomly selected from those collected each week, is a valid indicator of treatment performance.

Penicillin-resistant Streptococcus pneumoniae keratitis

Although Streptococcus pneumoniae is usually extremely sensitive to penicillin, recent reports in the nonophthalmic literature indicate a rise in the prevalence of penicillin-resistant pneumococci. We report a case of penicillin-resistant S. pneumoniae keratitis in a 51-year-old woman with severe rheumatoid arthritis. This report is only the second such case in the ophthalmic literature and may indicate the emergence of penicillin-resistant strains of S. pneumoniae as ocular pathogens.

Correlation-based pattern recognition for implantable defibrillators

An estimated 300,000 Americans die each year from cardiac arrhythmias. Historically, drug therapy or surgery were the only treatment options available for patients suffering from arrhythmias. Recently, implantable arrhythmia management devices have been developed. These devices allow abnormal cardiac rhythms to be sensed and corrected in vivo. Proper arrhythmia classification is critical to selecting the appropriate therapeutic intervention. The classification problem is made more challenging by the power/computation constraints imposed by the short battery life of implantable devices. Current devices utilize heart rate-based classification algorithms. Although easy to implement, rate-based approaches have unacceptably high error rates in distinguishing supraventricular tachycardia (SVT) from ventricular tachycardia (VT). Conventional morphology assessment techniques used in ECG analysis often require too much computation to be practical for implantable devices. In this paper, a computationally-efficient, arrhythmia classification architecture using correlation-based morphology assessment is presented. The architecture classifies individuals heart beats by assessing similarity between an incoming cardiac signal vector and a series of prestored class templates. A series of these beat classifications are used to make an overall rhythm assessment. The system makes use of several new results in the field of pattern recognition. The resulting system achieved excellent accuracy in discriminating SVT and VT.

A multicenter trial of bupropion for cocaine dependence in methadone-maintained patients

We conducted a multi-site, placebo-controlled, randomized double-blind clinical trial comparing bupropion HCL (300 mg/day) to placebo for the treatment of cocaine dependence in methadone-maintained subjects. A total of 149 subjects at three sites participated in a 12-week study. Outcome measures included cocaine use, level of depression, and psychosocial functioning. Results showed no significant differences between placebo and bupropion. Exploratory analyses suggested a medication effect for the subset of subjects depressed at study entry. The need to target subgroups of cocaine abusers in future pharmacotherapy trials and the possible role of treatment readiness are discussed.

Inducible interaction of integrin alpha 2 beta 1 with calreticulin. Dependence on the activation state of the integrin

We have previously demonstrated an interaction between the highly conserved KXGFFKR sequence of the integrin alpha-subunit cytoplasmic domains and calreticulin. Since this highly conserved sequence motif has been implicated in the regulation of the integrin affinity state, we wanted to determine whether the calreticulin-integrin interaction also depended on the integrin affinity state, and whether calreticulin occupation of integrin via the KXGFFKR motif was involved in the regulation of the ligand affinity state. We now demonstrate that anti-integrin antibody- or phorbol 12-myristate 13-acetate (PMA)-induced activation of the alpha 2 beta 1 integrin on Jurkat cells, as determined by stimulation of adhesion to collagen type I, resulted in an increased amount of calreticulin bound to this integrin. alpha 2 beta 1 activation with either anti-beta 1 or anti-alpha 2 monoclonal antibodies resulted in a greater amount of calreticulin coimmunoprecipitating with this integrin. Inactivation by neutralizing anti-integrin antibodies abrogated the calreticulin-integrin interaction. A correlation was also found between PMA-induced alpha 2 beta 1 activation and the amount of calreticulin bound to this integrin. Furthermore, pretreatment of streptolysin O-permeablized Jurkat cells with an anti-calreticulin antibody resulted in a significant and specific inhibition of the adhesion to collagen type I that could be induced by antibodies to alpha 2 beta 1 or by PMA. These data suggest that the active, high affinity form of alpha 2 beta 1 binds calreticulin and that calreticulin binding to the alpha 2 cytoplasmic domain may be required for stabilizing the high affinity state of this integrin. The data presented here also demonstrate, for the first time, an inducible interaction of an integrin with an intracellular protein that occurs via the alpha subunit of the integrin.

Migration of human polymorphonuclear leukocytes through a synovial fibroblast barrier is mediated by both beta 2 (CD11/CD18) integrins and the beta 1 (CD29) integrins VLA-5 and VLA-6

Polymorphonuclear leukocytes (PMNL) accumulate in joint fluid in inflammatory arthritides. We investigated the molecular mechanisms required for PMNL migration through a barrier of human synovial fibroblasts (HSF) grown on microporous filters, as a model of PMNL migration through synovial connective tissue and compared this process with PMNL migration through human dermal fibroblast (HDF) barriers and through human umbilical vein endothelium (HUVE). A small amount of PMNL migration occurred spontaneously only through the synovial fibroblast/filter unit (6-10%). Migration markedly increased through all cell monolayers when the chemotactic factors C5a, IL-8, or zymosan-activated plasma (containing C5adesArg) were added to form a chemotactic gradient. The migration induced by C5a, IL-8, or C5adesArg across HSF was partially inhibited (25-76% depending on stimulus) by mAb to CD18 (beta 2 integrin). The CD18-independent migration induced by IL-8 or C5adesArg was almost completely inhibited by mAbs to beta 1 integrin, but with C5a, inhibition by mAb to beta 1 integrin was only partial (40-50%). Inhibition by mAb to beta 1 integrin required treatment of the PMNL, but not the HSF and was only observed when the function of CD11/CD18 on PMNL was also blocked by a mAb. Treatment of PMNL with mAb to alpha 5 (VLA-5) plus alpha 6 (VLA-6) in combination, was required to inhibit CD18-independent migration through HSF to the degree observed with mAb to beta 1 integrin. There was no qualitative difference in the mechanisms utilized by PMNL for migration through HSF or HDF in response to chemotactic factors. In contrast, PMNL migration across HUVE was almost completely CD18-dependent (85%) with no role for beta 1 integrins. The results suggest that (a) PMNL migration through HSF in response to chemotactic factors utilizes both CD11/CD18 and beta 1 (CD29) integrins; (b) the VLA-5 and VLA-6 members of beta 1 integrins are involved in mediating migration; and (c) PMNL utilize similar mechanisms for migration through HSF and HDF, which are distinct from migration through HUVE.

Expression in Escherichia coli and affinity purification of a CKS-troponin I fusion protein

The human cardiac troponin I gene was subcloned and expressed at high levels in Escherichia coli as a fusion protein to CMP-KDO synthetase (CKS). Expression levels of the CKS-troponin I fusion were 8% of total cellular protein 4 h after induction with IPTG. The fusion was expressed primarily as an insoluble protein as shown by SDS-PAGE analysis. Expressed CKS-troponin I fusion from a crude lysate was antigenic against anti-CKS and anti-troponin I monoclonal antibodies in Western blots. The fusion was affinity-purified over a TnC affinity column using a urea-solubilized extract of a crude cell lysate. Serial dilutions of crude soluble extracts of the troponin I fusion were assayed in several microparticle enzyme immunoassays and found to exhibit similar immunogenic responses relative to cardiac troponin I isolated from human heart tissue.

Involvement of beta-1 and beta-2 adrenergic receptors in the antidepressant-like effects of centrally administered isoproterenol

Central (i.c.v.) but not peripheral (i.p.) administration of isoproterenol reduced response rates and increased reinforcement rates of rats under a differential-reinforcement-of-low-response-rate 72-sec schedule in a dose-dependent manner. This effect of centrally administered isoproterenol was similar to effects produced by administration of proven antidepressant drugs. Propranolol antagonized the effect of centrally administered isoproterenol, suggesting that the antidepressant-like effect of this agonist was mediated by beta adrenergic receptors. In addition, both the beta-1 selective antagonist betaxolol and the beta-2 selective antagonist ICI 118,551 antagonized the effect of centrally administered isoproterenol in a dose-dependent manner. These antagonists exhibited similar potency, suggesting that both beta-1 and beta-2 adrenergic receptors were involved in the mediation of the antidepressant-like effect of centrally administered isoproterenol. In rats with down-regulated beta-2 adrenergic receptors, produced by repeated treatment with clenbuterol, isoproterenol still reduced response rates and increased reinforcement rates of rats under the differential-reinforcement-of-low-response-rate schedule. By contrast, the antidepressant-like effect of the beta-2 adrenergic agonist clenbuterol was attenuated by down-regulation of beta-2 adrenergic receptors. The present results indicate that stimulation of central beta adrenergic receptors by intraventricular administration of isoproterenol produces behavioral changes similar to those observed after administration of proven antidepressant drugs. Both beta-1 and beta-2 adrenergic receptors appear to mediate the antidepressant-like effect of isoproterenol.

Rapid automated enzyme immunoassay of serum amyloid A

Serum amyloid A (SAA), a sensitive acute-phase protein, is the precursor of AA fibrils in reactive amyloidosis. However, SAA is poorly immunogenic, and development and standardization of immunoassays of this protein have been difficult. We established an automated polyclonal/monoclonal microparticle capture enzyme immunoassay, standardized with the World Health Organization prospective reference standard for SAA. A stabilized concentrate of SAA was used for controls and calibrators. The assay range was 1-750 mg/L with CVs < 7% throughout. Plasma and serum gave identical results and no interferences were observed. Linear regression against radial immunodiffusion assay gave a slope of 1.04 (95% confidence interval 0.99-1.10), intercept of -9 mg/L (95% confidence interval -14-3), and residual SD (SEE) of 20 for samples containing < or = 200 mg/L (n = 173). In 105 apparently healthy adults the mean (SD) SAA concentration was 3.7 (3.6) mg/L, the median was 3.0 mg/L, and the range, 0.7-26.4 mg/L. In clinical acute-phase sera, values up to 2200 mg/L were seen. This method will facilitate measurement and investigation of SAA in clinical practice generally and in AA amyloidosis.

Expression of ICAM-R (ICAM-3), a novel counter-receptor for LFA-1, in rheumatoid and nonrheumatoid synovium. Comparison with other adhesion molecules

OBJECTIVE

To study the distribution of intercellular adhesion molecule receptor (ICAM-R, or ICAM-3), a novel ligand for the leukointegrin lymphocyte function-associated antigen 1 (LFA-1), in normal and rheumatoid synovial membranes and to compare this with the distribution of ICAM-1, ICAM-2, vascular cell adhesion molecule 1 (VCAM-1), and endothelial leukocyte adhesion molecule 1 (ELAM-1).

METHODS

We performed immunohistochemical analyses of frozen sections of normal and rheumatoid synovial tissue using monoclonal antibodies to the molecules examined.

RESULTS

ICAM-1 staining was detectable on the vascular endothelium and the synovial lining cells of both normal and rheumatoid synovial membranes. A variable proportion of lymphocytes infiltrating rheumatoid tissues expressed ICAM-1, ICAM-2 staining was demonstrable in the vascular endothelium of both normal and inflamed tissues, the latter demonstrating a significantly higher proportion of positive vessels. ELAM-1 staining was not detectable in normal synovial membranes but was seen on the endothelium of a limited number of rheumatoid synovial vessels, usually close to the synovial lining cell layer. VCAM-1 staining was intense in both normal and rheumatoid synovial lining cells, but vascular staining was weak in both. In contrast, ICAM-R staining was not detected in association with any synovial blood vessels, but was widely expressed by lymphocytes and macrophages. Cells of the lining layer did not stain for ICAM-R.

CONCLUSION

Although ICAM-R is a ligand for LFA-1 and shares considerable sequence homology with ICAM-1 and ICAM-2, it does not appear to be expressed by the endothelium of normal or inflamed synovial vessels. Intense expression of ICAM-R by rheumatoid synovial lymphocytes and macrophages suggests that it may play a role in processes requiring cell-cell contact, such as antigen presentation and homotypic aggregation.

Comparison of the results of bacterial cultures from multiple sites in chronic osteomyelitis of long bones. A prospective study

We evaluated the results of aerobic, anaerobic, and fungal cultures of specimens that had been obtained from multiple sites in thirty patients who had traumatic osteomyelitis with a sinus track. In each patient, we obtained specimens of material from the sinus track; specimens of purulent fluid, of soft tissue, and of bone obtained from curettage; and specimens from the bed of the involved bone. More than one organism grew on culture of the specimens from twenty-one of the patients; more than three organisms, from those of eleven patients; and ten organisms, from those of two patients. The same organisms grew on culture of the specimens from every site in only fourteen (47 per cent) of the thirty patients. We recommend that specimens of material from the sinus track; specimens of purulent material, of soft tissue, and of bone obtained from curettage; and specimens from the bed of the involved bone be obtained for culture before the treatment of chronic osteomyelitis with a draining sinus, so that as many of the infecting organisms as possible will be identified.

A five-year multicenter study of the susceptibility of the Bacteroides fragilis group isolates to cephalosporins, cephamins, penicillins, clindamycin, and metronidazole in the United States

Over 2800 clinical strains of the Bacteroides fragilis group were collected during a 5-year period from ten geographically separate sites and tested for their susceptibility to various antimicrobial agents using a broth microdilution method. Among the cephalosporins, ceftizoxime was the most active (13% resistance) and importantly exhibited relatively equal activity against both B. fragilis species and non-B. fragilis species. Cefotaxime exhibited similar activity with an overall resistance rate of 18%. Both ceftriaxone and cefoperazone were appreciably less active cephalosporins especially against non-B. fragilis species. With regard to cephamycins, cefoxitin (MIC90, 32 micrograms/ml) was more active than cefotetan (MIC90, > or = 256 micrograms/ml) and cefmetazole (MIC90, 64 micrograms/ml). Non-B. fragilis species were highly resistant to cefotetan and cefmetazole. Imipenem was highly active against all strains with the exception of four strains of B. fragilis. Ampicillin-sulbactam, amoxicillin-clavulanate, piperacillin-tazobactam, and cefoperazone-sulbactam were all highly active with resistance rates < 2%. No resistance was detected to metronidazole, whereas 14% of isolates were resistant to clindamycin. When compared with other studies, these findings underscore the wide variability in susceptibility patterns reported nationwide and the need to continue monitoring these patterns to aid in choosing the most active compounds for therapy.

Adhesion of peripheral blood lymphocytes of children with arthritis to human umbilical vein endothelial cells

To determine whether adhesion of peripheral blood lymphocytes (PBL) of patients with juvenile rheumatoid arthritis (JRA) may be enhanced, adhesion of PBL of children with JRA, children with seronegative spondyloarthropathies (SSA), age-appropriate and adult controls, to human umbilical vein endothelial cells (HUVEC) was assessed in vitro. B and CD4 T lymphocytes in initial, adherent, and non-adherent cell fraction were identified by flow cytometry. B lymphocytes of all the younger subjects combined had a higher adherence to activated HUVEC compared with B lymphocytes of the adult donors. Except for greater adherence of HLA-DR+ CD4 T cells, lymphocytes of children with JRA showed no enhanced adhesion to either unactivated or activated HUVEC. The percentage of B cells adherent to activated HUVEC in each of the subject groups was 1.5-3.6-fold higher than adherent CD4 T lymphocytes. Surface analyses indicated higher percentages of CD49d (alpha 4)+ and CD29 (beta 1)+ CD4 T lymphocytes in adherent cells, but less of a differential in CD49 (alpha 4)+ and no difference in CD29 (beta 1)+ B lymphocytes. There were fewer Leu-8 (L-selectin)+ B and Leu-8+ CD4 T cells among adherent cells. The data suggest a greater adhesive capacity of B lymphocytes compared with CD4 T lymphocytes which is unrelated to disease, and the possibility that B lymphocytes may utilize adhesion molecules distinct from those of CD4 T lymphocytes. Only a small subset of T cells of patients with JRA may have an enhanced capacity for adhesion to endothelium.

Cell attachment to extracellular matrix substrates is inhibited upon downregulation of expression of calreticulin, an intracellular integrin alpha-subunit-binding protein

We have demonstrated recently that calreticulin, an intracellular calcium-binding protein, can interact with the alpha-subunits of integrin receptors via the highly conserved KXGFFKR amino acid sequence present in the cytoplasmic domains of all integrin alpha-subunits (Rojiani et al. (1991) Biochemistry 30, 9859-9866). Here we demonstrate that calreticulin can be co-localized by immunofluorescence as well as co-purified with integrins, that recombinant calreticulin can also interact with integrins, and that the interaction occurs predominantly via the N-domain of calreticulin, to a much lesser extent with the C-domain, but not at all with the proline-rich P-domain. To demonstrate a physiological role for the interaction of calreticulin with integrins, calreticulin expression was downregulated by treating cells with antisense oligonucleotides designed to inhibit the initiation of translation of calreticulin. Antisense oligonucleotides, but not sense or non-sense oligonucleotides, inhibited attachment and spreading of cells cultured in the presence of fetal bovine serum, and also of cells plated on individual extracellular matrix substrates in the absence of serum. The antisense oligonucleotide inhibited cell proliferation of anchorage-dependent cells slightly, but there was no effect on cell viability. The effect on cell attachment was similar to that achieved by treating cells with an antisense oligonucleotide designed to inhibit translation of the integrin alpha 3 subunit, which resulted in the inhibition of cell attachment to alpha 3 beta 1-specific substrates. The effect of the antisense calreticulin oligonucleotide on cell attachment was demonstrated to be integrin-mediated since antisense calreticulin treatment of Jurkat cells abrogated the stimulation of collagen cell attachment achieved by attachment-stimulating signalling anti-alpha 2 (JBS2) and anti-beta 1 (21C8) antibodies. The oligonucleotides did not affect the rate of cell proliferation of these cells. These results demonstrate a fundamental role of calreticulin in cell-extracellular matrix interactions.

Cell attachment to extracellular matrix substrates is inhibited upon downregulation of expression of calreticulin, an intracellular integrin alpha-subunit-binding protein

We have demonstrated recently that calreticulin, an intracellular calcium-binding protein, can interact with the alpha-subunits of integrin receptors via the highly conserved KXGFFKR amino acid sequence present in the cytoplasmic domains of all integrin alpha-subunits (Rojiani et al. (1991) Biochemistry 30, 9859–9866). Here we demonstrate that calreticulin can be co-localized by immunofluorescence as well as co-purified with integrins, that recombinant calreticulin can also interact with integrins, and that the interaction occurs predominantly via the N-domain of calreticulin, to a much lesser extent with the C-domain, but not at all with the proline-rich P-domain. To demonstrate a physiological role for the interaction of calreticulin with integrins, calreticulin expression was downregulated by treating cells with antisense oligonucleotides designed to inhibit the initiation of translation of calreticulin. Antisense oligonucleotides, but not sense or non-sense oligonucleotides, inhibited attachment and spreading of cells cultured in the presence of fetal bovine serum, and also of cells plated on individual extracellular matrix substrates in the absence of serum. The antisense oligonucleotide inhibited cell proliferation of anchorage-dependent cells slightly, but there was no effect on cell viability. The effect on cell attachment was similar to that achieved by treating cells with an antisense oligonucleotide designed to inhibit translation of the integrin alpha 3 subunit, which resulted in the inhibition of cell attachment to alpha 3 beta 1-specific substrates. The effect of the antisense calreticulin oligonucleotide on cell attachment was demonstrated to be integrin-mediated since antisense calreticulin treatment of Jurkat cells abrogated the stimulation of collagen cell attachment achieved by attachment-stimulating signalling anti-alpha 2 (JBS2) and anti-beta 1 (21C8) antibodies. The oligonucleotides did not affect the rate of cell proliferation of these cells. These results demonstrate a fundamental role of calreticulin in cell-extracellular matrix interactions.

Lymphocyte emperipolesis in human glial cells

Astrocytes have been observed to contain intact, viable lymphocytes within their cytoplasm (emperipolesis) in multiple sclerosis plaques and some brain tumors. This study characterizes the adhesive, emperipoletic and phagocytic properties of glial cells in culture. Human fetal and adult astrocytes engaged in adhesion and emperipolesis of lymphocytes. Emperipolesis, and not adhesion, was temperature- and cation-dependent. The CD8 and MHC Class I antigens played a role in emperipolesis. Lymphocytes most often remained viable within the cytoplasm of astrocytes but occasionally underwent lysis or caused disruption of the astrocyte intermediate filaments. The phenomenon of emperipolesis is distinct from phagocytosis, since microglia showed prominent phagocytic properties but did not engage in emperipolesis. Conversely, astrocytes were efficient emperipolites and rarely demonstrated phagocytic properties.