SUMMARY Interleukin-16 (IL-16) acts as a chemoattractant for CD4+ cells, as a modulator of T-cell activation, and plays a key role in asthma. This report describes the cytokine-inducing effects of IL-16 on total peripheral blood mononuclear cells (PBMC) and PBMC subpopulations. While CD4 + T lymphocytes did not secrete cytokines in response to rhIL-16, CD14+ CD4 + monocytes and maturing macrophages secrete IL-1b, IL-6, IL-15 and tumour necrosis factor-a (TNF-a) upon rhIL-16 stimulation. The mRNA species for these four cytokines were detected as early as 4 hr poststimulation, with protein being secreted by 24 hr. Secretion of IL-1b and IL-6 by total PBMC was dose dependent, with maximal secretion being observed using 50 ng/ml rhIL-16. However, for IL-15 or TNF-a maximal secretion by total PBMC occurred with all concentrations between 5 ng/ml to 500 ng/ml rhIL-16. Puri®ed monocytes/macrophages secreted maximal concentrations of all four cytokines in the presence of 500 ng/ml rhIL-16, except for monocytes where maximal secretion of IL-15 was, interestingly, observed with only 50 ng/ml rhIL-16. The use of higher concentrations of rhIL-16 (1000 ng/ml) inhibited secretion of all four cytokines. While these IL-16-induced cytokines are likely to be involved in the immune system's response to antigen, the data suggest that IL-16 may play a key role in initiating and/or sustaining an in¯ammatory response.

Melanoma inhibitory activity (MIA) protein is a clinically valuable marker in patients with malignant melanoma, as enhanced values diagnose metastatic melanoma stages III and IV. Here we show that the recombinant human MIA adopts an SH3 domain-like fold in solution, with two perpendicular, antiparallel, three-and ®ve-stranded b-sheets. In contrast to known structures with the SH3 domain fold, MIA is a single-domain protein, and contains an additional antiparallel b-sheet and two disul®de bonds. MIA is also the ®rst extracellular protein found to have the SH3 domain-like fold. Furthermore, we show that MIA interacts with ®bronectin and that the peptide ligands identi®ed for MIA exhibit a matching sequence to type III human ®bronectin repeats, especially to FN14, which is close to an integrin a 4 b 1 binding site. The present study, therefore, may explain the role of MIA in metastasis in vivo, and supports a model in which the binding of human MIA to type III repeats of ®bronectin competes with integrin binding, thus detaching cells from the extracellular matrix.

The expression of specific membrane receptors for C3a was determined on guinea pig C3a-sensitive (gp R+) platelets and human polymorphonuclear leukocytes (hu PMNL). Binding studies with 125I-labeled C3a from gp or hu sources and Scatchard analysis applied to the binding data revealed the existence of two receptor classes on gp R+ platelets; a high-affinity class with about 200 binding sites/cell and Kd = 1.7 x 10(-9) M, and a relatively low-affinity class with Kd = 10(-8) M and about 500 sites/cell. Hu PMNL express a homogeneous receptor class with Kd = 3 x 10(-8) M and 40,000 sites/cell. Molecular characterization of the C3a receptor on gp R+ platelets was achieved by (a) cross-linking photoaffinity-labeled receptors to bound 125I-labeled C3a; (b) photoaffinity labeling receptors with a 13-amino acid residue C3a analogue 125I-Nap-Ahx-13; and (c) use of chemical cross-linkers like disuccinimidylsuberate to cross-link receptors with 125I-C3a. All three techniques gave rise to very similar labeling patterns. With the photoaffinity labeling methods, a diffuse band pattern was observed with an apparent molecular mass of 95-123 kDa with 125I-C3a as label, and 85-105 kDa with 125I-Nap-Ahx-13 as label. Chemical cross-linking of 125I-C3a revealed three distinct bands with molecular masses of approximately 123, 108 and 95 kDa. Subtracting the contribution of the cross-linked ligands, the C3a receptor on gp R+ platelets appears to be a protein complex, consisting of one to three components with estimated molecular masses between 83-114 kDa.

No abstract

The enzymatic dissociation of acinar tissue by collagenase is a substantial step in the isolation of pancreatic islets. Although essential collagenase components have been purified, the variability in the activity of different batches limits long-term reproducibility of isolation success. The utilization of purified recombinant proteases would solve this problem. In the present study, pancreases from multiorgan donors were dissociated by means of digestion-filtration using either Liberase HI (n ‫؍‬ 51) or a recombinant collagenase blend (n ‫؍‬ 25). No significant differences were found regarding islet yield before and after purification, the percent of exocrine-attached islets, and final purity. However, the ratio between islet equivalents and islet numbers indicated a lesser fragmentation in islets isolated with recombinant collagenase (P < 0.01). In contrast, viability was slightly higher in islets isolated with Liberase (92.3 ؎ 0.8 vs. 85.6 ؎ 2.9%; P < 0.05). Insulin release during static glucose incubation was not different between experimental groups. Islet transplantation into diabetic nude mice resulted in sustained normoglycemia in either group until the graft was removed. These results demonstrated that viable human islets can be isolated using recombinant collagenase. Final optimization of this enzyme blend would offer continuous reproducibility of isolation success.