Polyproline binding is an essential function of human profilin in yeast.

Structural analysis of human profilin has revealed two tryptophan residues, W3 and W31, which interact with polyproline. The codons for these residues were mutated to encode phenylalanine and the mutant proteins overexpressed in Eschericia coli. The isolated proteins were diminished in their ability to bind polyproline, whereas phosphatidylinositol 4,5-bisphosphate (PIP2) binding remained unchanged. In many strains of Saccharomyces cerevisiae, disruption of the gene encoding profilin, PFY1, is lethal. It was found that expression of the gene for human profilin is capable of suppressing this lethality. The polyproline-binding mutant alleles of the human gene were cloned into various yeast expression vectors. Each of the mutant genes resulted in suppression of the lethality of pfy1Delta. It was observed that the mutant protein expression levels paralleled the growth rates of the strains. The severity of various morphological abnormalities of the strains was also attenuated with increased protein levels, suggesting that profilin polyproline-binding mutations are deleterious to cell growth unless overexpressed. Both tryptophan mutations were combined to give a third mutant allele that was found both unable to bind polyproline and to suppress the lethality of a pfy1 deletion. Immunoprecipitation experiments suggested that the mutants were unaltered in their affinity for actin and PIP2. These data strongly suggest that polyproline binding is an essential function of profilin.     

Investigation of immunosuppressive properties of inactivated human immunodeficiency virus and possible neutralization of this effect by some patient sera

Retroviral infections are accompanied by immunosuppression in a variety of species. For feline leukemia virus, the immunosuppression has been ascribed to the transmembrane envelope protein, p15E, which suppresses the proliferative responses of cat, mouse, and human lymphocytes. A similar suppressive effect has been shown for a lysate of human immunodeficiency virus (HIV), strain HTLV-IIIB. Here we determined that detergent-disrupted HTLV-IIIB lystate exerted a strong suppressive effect on PHA-stimulated lymphocytes. Preparations of whole virions, a lysate of a local HIV isolate grown on MP-6 cells, and a commercially obtained UV and psoralene-inactivated lysate were examined and demonstrated to have a similar suppressive effect. The HIV lysate was not directly cytotoxic to lymphocytes and did not contain tumor necrosis factor or lymphotoxin. The HIV lysate specifically suppressed the proliferation of a range of hemopoietic cell lines from man and mouse including three EBV transformed CD4- and IL-2 receptor-negative B-cell lines. The lysate also suppressed the formation of human bone marrow colonies, whereas the lysate had only a slight or no effect on fibroblasts. The suppression of lymphocyte proliferation was not abrogated by addition of IL-2 or IL-1 and the HIV lysate inhibited the expression of IL-2 receptors on suboptimal PHA-stimulated mononuclear cells. The suppressive factor(s) has not been characterized in molecular terms, but suppressive activity was recovered in fractions with a molecular weight of about 67,000 and in both the glycoprotein fraction and in the glycoprotein-depleted fraction of the HIV lysate. Sera from one-third of a small series (N = 13) of individuals with antibodies to HIV seem to be able to neutralize the suppressive properties of HIV lysate in cultures.     

Degradation of extracellular matrix proteins by hemorrhagic metalloproteinases

The proteolytic activity of four hemorrhagic metalloproteinases (Ht-a, c, d, and e) isolated from the venom of the Western diamondback rattlesnake (Crotalus atrox) was investigated using isolated extracellular matrix (ECM) proteins. We determined that all of the proteinases are capable of cleaving fibronectin, laminin, type IV collagen, nidogen (entactin), and gelatins. However, none of the proteinases were proteolytic against the interstitial collagen types I and III or type V collagen. With all of the substrates listed above Ht-c and Ht-d produced identical digestion patterns, as would be expected for these isoenzymes. With fibronectin, Ht-a produces a different ratio of products from Ht-c and Ht-d, while Ht-e produces a unique pattern of digestion. Ht-e and Ht-a produced nonidentical patterns with the laminin/nidogen preparation although some similarity was shared between them as well as with the Ht-c/d digestion pattern. Similar results were also observed for these proteinases with nidogen 150 as the substrate. The type IV collagen digestion patterns by Ht-e and Ht-a were similar to the pattern observed with Ht-c/d but differed by two bands. The digestion patterns of the three gelatins produced by the proteinases show differences between Ht-c and Ht-d when compared to Ht-e and Ht-a. This investigation clearly shows that several of the ECM proteins are efficiently digested by these toxins. The proteinases have some digestion sites in common but show differing specificities. In addition, the range of ECM proteins digested by these hemorrhagic proteinases is nearly identical to that demonstrated by the ECM proteinase stromelysin (MMP-3). From these data, and the knowledge of the roles these ECM proteins have in maintaining basement membrane structural/functional integrity, one can envision that the degradation of these ECM proteins could readily lead to loss of capillary integrity resulting in hemorrhage occurring at those sites.