Publications

Elevated basal serum tryptase levels are present in 4-6% of the general population, but the cause and relevance of such increases are unknown. Previously, we described subjects with dominantly inherited elevated basal serum tryptase levels associated with multisystem complaints including cutaneous flushing and pruritus, dysautonomia, functional gastrointestinal symptoms, chronic pain, and connective tissue abnormalities, including joint hypermobility. Here we report the identification of germline duplications and triplications in the TPSAB1 gene encoding α-tryptase that segregate with inherited increases in basal serum tryptase levels in 35 families presenting with associated multisystem complaints. Individuals harboring alleles encoding three copies of α-tryptase had higher basal serum levels of tryptase and were more symptomatic than those with alleles encoding two copies, suggesting a gene-dose effect. Further, we found in two additional cohorts (172 individuals) that elevated basal serum tryptase levels were exclusively associated with duplication of α-tryptase-encoding sequence in TPSAB1, and affected individuals reported symptom complexes seen in our initial familial cohort. Thus, our findings link duplications in TPSAB1 with irritable bowel syndrome, cutaneous complaints, connective tissue abnormalities, and dysautonomia.

Streptococcus sanguinis is a leading cause of bacterial infective endocarditis, a life-threatening infection of heart valves. S. sanguinis binds to human platelets with high avidity, and this adherence is likely to enhance virulence. Previous studies suggest that a serine-rich repeat adhesin termed SrpA mediates the binding of S. sanguinis to human platelets via its interaction with sialoglycans on the receptor GPIbα. However, in vitro binding assays with SrpA and defined sialoglycans failed to identify specific high-affinity ligands. To improve our understanding of the interaction between SrpA and human platelets, we determined cocrystal structures of the SrpA sialoglycan binding region (SrpA) with five low-affinity ligands: three sialylated trisaccharides (sialyl-T antigen, 3'-sialyllactose, and 3'-sialyl-N-acetyllactosamine), a sialylated tetrasaccharide (sialyl-Lewis), and a sialyl galactose disaccharide component common to these sialoglyans. We then combined structural analysis with mutagenesis to further determine whether our observed interactions between SrpA and glycans are important for binding to platelets and to better map the binding site for the physiological receptor. We found that the sialoglycan binding site of SrpA is significantly larger than the sialoglycans cocrystallized in this study, which suggests that binding of SrpA to platelets either is multivalent or occurs via a larger, disialylated glycan.

Host-derived proteases can augment or help to clear infections. This dichotomy is exemplified by cathepsin L (CTSL), which helps Hendra virus and SARS coronavirus to invade cells, but is essential for survival in mice with mycoplasma pneumonia. The present study tested the hypothesis that CTSL protects mice from serious consequences of infection by the orthomyxovirus influenza A, which is thought to be activated by host-supplied proteases other than CTSL. Ctsl-/- mice infected with influenza A/Puerto Rico/8/34(H1N1) had larger lung viral loads and higher mortality than infected Ctsl+/+ mice. Lung inflammation in surviving infected mice peaked 14 days after initial infection, accompanied marked focal distal airway bronchiolization and epithelial metaplasia followed by desquamation and fibrotic interstitial remodeling, and persisted for at least 6 weeks. Most deaths occurred during the second week of infection in both groups of mice. In contrast to mycoplasma pneumonia, infiltrating cells were predominantly mononuclear rather than polymorphonuclear. The histopathology of lung inflammation and remodeling in survivors was similar in Ctsl-/- and Ctsl+/+ mice, although Ctsl+/+ mice cleared immunoreactive virus sooner. Furthermore, Ctsl-/- mice had profound deficits in CD4+ lymphocytes before and after infection and weaker production of pathogen-specific IgG. Thus, CTSL appears to support innate as well as adaptive responses, which confer a survival advantage on mice infected with the orthomyxovirus influenza A.