Logo: to the web site of Uppsala University

Several hematopoietic cells of the immune system store large amounts of proteases in cytoplasmic granules. The absolute majority of these proteases belong to the large family of chymotrypsin-related serine proteases. The chymase locus is one of four loci encoding these granule-associated serine proteases in mammals. The chymase locus encodes only four genes in primates, (1) the gene for a mast-cell-specific chymotryptic enzyme, the chymase; (2) a T-cell-expressed asp-ase, granzyme B; (3) a neutrophil-expressed chymotryptic enzyme, cathepsin G; and (4) a T-cell-expressed chymotryptic enzyme named granzyme H. Interestingly, this locus has experienced a number of quite dramatic expansions during mammalian evolution. This is illustrated by the very large number of functional protease genes found in the chymase locus of mice (15 genes) and rats (18 genes). A separate expansion has also occurred in ruminants, where we find a new class of protease genes, the duodenases, which are expressed in the intestinal region. In contrast, the opossum has only two functional genes in this locus, the mast cell (MC) chymase and granzyme B. This low number of genes may be the result of an inversion, which may have hindered unequal crossing over, a mechanism which may have been a major factor in the expansion within the rodent lineage. The chymase locus can be traced back to early tetrapods as genes that cluster with the mammalian genes in phylogenetic trees can be found in frogs, alligators and turtles, but appear to have been lost in birds. We here present the collected data concerning the evolution of this rapidly evolving locus, and how these changes in gene numbers and specificities may have affected the immune functions in the various tetrapod species.

Metallo-carboxypeptidases are exopeptidases with diverse expression and function, found in all kingdoms of life from bacteria to mammals. One of them, the carboxypeptidase A3 (CPA3), has become an important component of the mammalian immune system by its expression in mast cells. Mast cells (MCs) are highly specialized sentinel cells, which store large amounts of bioactive mediators, including CPA3, in very abundant cytoplasmic granules. Clinical studies have found an increased CPA3 expression in asthma but the physiological role as well as the evolutionary origin of CPA3 remains largely unexplored. CPA3 belongs to the M14A subfamily of metallocarboxypeptidases, which among others also includes the digestive enzymes CPA1, CPA2, CPB1 and CPO. To study the appearance of CPA3 during vertebrate evolution, we here performed bioinformatic analyses of homologous genes and gene loci from a broad panel of metazoan animals from invertebrates to mammals. The phylogenetic analysis indicated that CPA3 appeared at the base of tetrapod evolution in a branch closer to CPB1 than to other CPAs. Indeed, CPA3 and CPB1 are also located in the same locus, on chromosome 3 in humans. The presence of CPA3 only in tetrapods and not in fishes, suggested that CPA3 could have appeared by a gene duplication from CPB1 during early tetrapod evolution. However, the apparent loss of CPA3 in several tetrapod lineages, e.g. in birds and monotremes, indicates a complex evolution of the CPA3 gene. Interestingly, in the lack of CPA3 in fishes, zebrafish MCs express instead CPA5 for which the most closely related human carboxypeptidase is CPA1, which has a similar cleavage specificity as CPA3. Collectively, these findings clarify and add to our understanding of the evolution of hematopoietic proteases expressed by mast cells.

Cell lines of monocyte/macrophage origin are often used as model systems to study monocyte/macrophage biology. A relevant question is how similar these cell lines are to their in vivo counterparts? To address this issue, we performed a detailed analysis of the transcriptome of two commonly used human monocyte/macrophage cell lines, Mono Mac 6 and THP-1. Both of these cell lines originate from leukemic cells with myelo-monocytic characteristics. We found that both Mono Mac 6 and THP-1 represent cells of very immature origin. Their transcriptomes show more similarities to immature neutrophils than cells of the monocyte/macrophage lineage. They express significant levels of N-elastase, proteinase 3, cathepsin G, and azurocidin but very low levels of CD14, ficolin, and complement factor P. All major MHC class II genes are also expressed at low levels. They show high levels of lysozyme and low levels of one of the immunoglobulin Fc receptors, FCGRIIA, which is characteristic of both neutrophils and monocytes. THP-1, but not Mono Mac 6, also expresses the high-affinity receptor for IgG, FCGRIA. Both cell lines lack the expression of the connective tissue components fibronectin, proteoglycan 4, and syndecan 3, which are characteristics of tissue macrophages but are absent in blood monocytes, indicating that they originate from bone marrow precursors and not yolk sac-derived hematopoietic cells. Both of these cell lines seem, therefore, to represent cells arrested during early myelo-monocytic development, at a branch point between neutrophil and monocyte differentiation. Their very immature phenotype indicates that great care should be taken when using these cell lines as models for normal monocyte/macrophage biology.

Studies of mast cell biology are dependent on relevant and validated in vitro models. Here, we present detailed information concerning the phenotype of both freshly isolated human skin mast cells (MCs) and of in vitro cultures of these cells that were obtained by analyzing their total transcriptome. Transcript levels of MC-related granule proteins and transcription factors were found to be remarkably stable over a 3-week culture period. Relatively modest changes were also seen for important cell surface receptors including the high-affinity receptor for IgE, FCER1A, the low-affinity receptor for IgG, FCGR2A, and the receptor for stem cell factor, KIT. FCGR2A was the only Fc receptor for IgG expressed by these cells. The IgE receptor increased by 2-5-fold and an approximately 10-fold reduction in the expression of FCGR2A was observed most likely due to the cytokines, SCF and IL-4, used for expanding the cells. Comparisons of the present transcriptome against previously reported transcriptomes of mouse peritoneal MCs and mouse bone marrow-derived MCs (BMMCs) revealed both similarities and major differences. Strikingly, cathepsin G was the most highly expressed granule protease in human skin MCs, in contrast to the almost total absence of this protease in both mouse MCs. Transcript levels for the majority of cell surface receptors were also very low compared to the granule proteases in both mouse and human MCs, with a difference of almost two orders of magnitude. An almost total absence of T-cell granzymes was observed in human skin MCs, indicating that granzymes have no or only a minor role in human MC biology. Ex vivo skin MCs expressed high levels of selective immediate early genes and transcripts of heat shock proteins. In validation experiments, we determined that this expression was an inherent property of the cells and not the result of the isolation process. Three to four weeks in culture results in an induction of cell growth-related genes accompanying their expansion by 6-10-fold, which increases the number of cells for in vitro experiments. Collectively, we show that cultured human skin MCs resemble their ex vivo equivalents in many respects and are a more relevant in vitro model compared to mouse BMMCs for studies of MC biology, in particular human MC biology.

Cows produce saliva in very large quantities to lubricate and facilitate food processing. Estimates indicate an amount of 50-150 L per day. Human saliva has previously been found to contain numerous antibacterial components, such as lysozyme, histatins, members of the S-100 family and lactoferrin, to limit pathogen colonization. Cows depend on a complex microbial community in their digestive system for food digestion. Our aim here was to analyze how this would influence the content of their saliva. We therefore sampled saliva from five humans and both nose secretions and saliva from six cows and separated the saliva on SDS-PAGE gradient gels and analyzed the major protein bands with LC-MS/MS. The cow saliva was found to be dominated by a few major proteins only, carbonic anhydrase 6, a pH-stabilizing enzyme and the short palate, lung and nasal epithelium carcinoma-associated protein 2A (SPLUNC2A), also named bovine salivary protein 30 kDa (BSP30) or BPIFA2B. This latter protein has been proposed to play a role in local antibacterial response by binding bacterial lipopolysaccharides (LPSs) and inhibiting bacterial growth but may instead, according to more recent data, primarily have surfactant activity. Numerous peptide fragments of mucin-5B were also detected in different regions of the gel in the MS analysis. Interestingly, no major band on gel was detected representing any of the antibacterial proteins, indicating that cows may produce them at very low levels that do not harm the microbial flora of their digestive system. The nose secretions of the cows primarily contained the odorant protein, a protein thought to be involved in enhancing the sense of smell of the olfactory receptors and the possibility of quickly sensing potential poisonous food components. High levels of secretory IgA were also found in one sample of cow mouth drippings, indicating a strong upregulation during an infection. The human saliva was more complex, containing secretory IgA, amylase, carbonic anhydrase 6, lysozyme, histatins and a number of other less abundant proteins, indicating a major difference to the saliva of cows that show very low levels of antibacterial components, most likely to not harm the microbial flora of the rumen.

Granzymes A and K are two highly homologous serine proteases expressed by mammalian cytotoxic T cells (CTLs) and natural killer (NK) cells. The locus encoding these two proteases is the first of the hematopoietic serine protease loci to appear during vertebrate evolution. This locus is found in all jawed vertebrates including the cartilaginous fishes. Granzyme A is the most abundant of the different granzymes expressed by CTLs and NK cells and its potential function has been studied extensively for many years. However, no clear conclusions concerning its primary role in the immune defense has been obtained. In all mammals, there are only one copy each of granzyme A and K, whereas additional copies are found in both cartilaginous and ray finned fishes. In cichlids two of these copies seem to encode new members of the granzyme A/K family. These two new members appear to have changed primary specificity and to be pure chymases based on the amino acids in their active site substrate binding pockets. Interestingly, one of these gene copies is located in the middle of the granzyme A/K locus, while the other copy is present in another locus, the met-ase locus. We here present a detailed characterization of the extended cleavage specificity of one of these non-classical granzymes, a Zebra mbuna granzyme positioned in the granzyme A/K locus. This enzyme, named granzyme A2, showed a high preference for tyrosine in the P1 position of substrates, thereby being a strict chymase. We have also characterized one of the classical granzyme A/Ks of the Zebra mbuna, granzyme A1, which is a tryptase with preference for arginine in the P1 position of substrates. Based on their extended specificities, the two granzymes showed major similarities, but also some differences in preferred amino acids in positions surrounding the cleavable amino acid. Fish lack one of the hematopoietic serine protease loci of mammals, the chymase locus, where one of the major mast cell enzymes is located. An interesting question is now if cichlids have by compensatory mechanisms generated a mast cell chymase from another locus, and if similar chymotryptic enzymes have appeared also in other fish species.

Granzymes A and K are two highly homologous serine proteases expressed by mammalian cytotoxic T cells (CTL) and natural killer cells (NK). Granzyme A is the most abundant of the different granzymes (gzms) expressed by these two cell types. Gzms A and K are found in all jawed vertebrates and are the most well conserved of all hematopoietic serine proteases. Their potential functions have been studied extensively for many years, however, without clear conclusions. Gzm A was for many years thought to serve as a key component in the defense against viral infection by the induction of apoptosis in virus-infected cells, similar to gzm B. However, later studies have questioned this role and instead indicated that gzm A may act as a potent inducer of inflammatory cytokines and chemokines. Gzms A and K form clearly separate branches in a phylogenetic tree indicating separate functions. Transcriptional analyses presented here demonstrate the presence of gzm A and K transcripts in both CD4⁺ and CD8⁺ T cells. To enable screening for their primary biological targets we have made a detailed analysis of their extended cleavage specificities. Phage display analysis of the cleavage specificity of the recombinant enzymes showed that both gzms A and K are strict tryptases with high selectivity for Arg over Lys in the P1 position. The major differences in the specificities of these two enzymes are located N-terminally of the cleavage site, where gzm A prefers small amino acids such as Gly in the P3 position and shows a relatively relaxed selectivity in the P2 position. In contrast, gzm K prefers large amino acids such as Phe, Tyr, and Trp in both the P2 and P3 positions and does not tolerate negatively charged residues in the P2 position. This major distinction in extended specificities is likely reflected also in preferred in vivo targets of these two enzymes. This information can now be utilized for high-precision screening of primary targets for gzms A and K in search of their highly conserved but still poorly defined functions in vertebrate immunity.

Salmonella is a common food-borne, intracellular pathogen that causes infections in animals and humans. During infection of the host, flagella serve as important virulence factors that enable the bacterium to reach the site of infection and adhere to the host-cell. Flagella are complex organelles, and their expression is organized by an intricate, regulatory network, governed by the transcription factor FlhDC. While the transcriptional regulation of flagellar genes in Salmonellais well described, there is little known about post-transcriptional mechanisms. The RNA binding protein ProQ was found to promote motility, but the molecular mode of action remained unclear. Recently, the small RNA FlgO was found to be a contributing genetic factor for ProQ-dependent flagellar gene expression. FlgO was hypothesized to base-pair with the 5’untranslated region of the flhDC mRNA and thereby promote translation of the mRNA. As ProQ was positively affecting FlgO steady-state levels, I hypothesized that ProQ stabilizes or aids in processing of FlgO. Further, I investigated the effect of FlgO on translation of the flhDC mRNA. The results of this work showed that the stability and processing of FlgO is independent of ProQ, and in vitro binding assays showed no binding of ProQ to FlgO. However, the RNA binding protein Hfq was found to bind and stabilize FlgO. Further, I found that overexpression of FlgO increases flhDC mRNA translation and that this effect is independent of the small RNAs ArcZ, OmrA/B and OxyS, all of which are known to repress flhDC translation. The precise molecular mechanism of how FlgO stimulates translation of flhDCremains subject for future work.

Increasing numbers of small proteins with diverse physiological roles are being identified and characterized in both prokaryotic and eukaryotic systems, but the origins and evolution of these proteins remain unclear. Recent genomic sequence analyses in several organisms suggest that new functions encoded by small open reading frames (sORFs) may emerge de novo from noncoding sequences. However, experimental data demonstrating if and how randomly generated sORFs can confer beneficial effects to cells are limited. Here, we show that by upregulating hisB expression, de novo small proteins (<= 50 amino acids in length) selected from random sequence libraries can rescue Escherichia coli cells that lack the conditionally essential SerB enzyme. The recovered small proteins are hydrophobic and confer their rescue effect by binding to the 5 ' end regulatory region of the his operon mRNA, suggesting that protein binding promotes structural rearrangements of the RNA that allow increased hisB expression. This study adds RNA regulatory elements as another interacting partner for de novo proteins isolated from random sequence libraries and provides further experimental evidence that small proteins with selective benefits can originate from the expression of nonfunctional sequences.

BACKGROUND: Obstructive sleep apnea (OSA) is a common sleep-breathing disorder linked to increased risk of cardiovascular disease. Intermittent hypoxia and intermittent airway obstruction, hallmarks of OSA, have been shown in animal models to induce substantial changes to the gut microbiota composition and subsequent transplantation of fecal matter to other animals induced changes in blood pressure and glucose metabolism.

RESEARCH QUESTION: Does obstructive sleep apnea in adults associate with the composition and metabolic potential of the human gut microbiota?

STUDY DESIGN AND METHODS: We used respiratory polygraphy data from up to 3,570 individuals aged 50-64 from the population-based Swedish CardioPulmonary bioImage Study combined with deep shotgun metagenomics of fecal samples to identify cross-sectional associations between three OSA parameters covering apneas and hypopneas, cumulative sleep time in hypoxia and number of oxygen desaturation events with gut microbiota composition. Data collection about potential confounders was based on questionnaires, on-site anthropometric measurements, plasma metabolomics, and linkage with the Swedish Prescribed Drug Register.

RESULTS: We found that all three OSA parameters were associated with lower diversity of species in the gut. Further, the OSA-related hypoxia parameters were in multivariable-adjusted analysis associated with the relative abundance of 128 gut bacterial species, including higher abundance of Blautia obeum and Collinsela aerofaciens. The latter species was also independently associated with increased systolic blood pressure. Further, the cumulative time in hypoxia during sleep was associated with the abundance of genes involved in nine gut microbiota metabolic pathways, including propionate production from lactate. Lastly, we observed two heterogeneous sets of plasma metabolites with opposite association with species positively and negatively associated with hypoxia parameters, respectively.

INTERPRETATION: OSA-related hypoxia, but not the number of apneas/hypopneas, is associated with specific gut microbiota species and functions. Our findings lay the foundation for future research on the gut microbiota-mediated health effects of OSA.