Jutta Heim

A new structural class of serine protease inhibitors revealed by the structure of the hirustasin–kallikrein complex

BACKGROUND Hirustasin belongs to a class of serine protease inhibitors characterized by a well conserved pattern of cysteine residues. Unlike the closely related inhibitors, antistasin/ghilanten and guamerin, which are selective for coagulation factor Xa or neutrophil elastase, hirustasin binds specifically to tissue kallikrein. The conservation of the pattern of cysteine residues and the significant sequence homology suggest that these related inhibitors possess a similar three-dimensional structure to hirustasin. RESULTS The crystal structure of the complex between tissue kallikrein and hirustasin was analyzed at 2.4 resolution. Hirustasin folds into a brick-like structure that is dominated by five disulfide bridges and is sparse in secondary structural elements. The cysteine residues are connected in an abab cdecde pattern that causes the polypeptide chain to fold into two similar motifs. As a hydrophobic core is absent from hirustasin the disulfide bridges maintain the tertiary structure and present the primary binding loop to the active site of the protease. The general structural topography and disulfide connectivity of hirustasin has not previously been described. CONCLUSIONS The crystal structure of the kallikrein-hirustasin complex reveals that hirustasin differs from other serine protease inhibitors in its conformation and its disulfide bond connectivity, making it the prototype for a new class of inhibitor. The disulfide pattern shows that the structure consists of two domains, but only the C-terminal domain interacts with the protease. The disulfide pattern of the N-terminal domain is related to the pattern found in other proteins. Kallikrein recognizes hirustasin by the formation of an antiparallel beta sheet between the protease and the inhibitor. The P1 arginine binds in a deep negatively charged pocket of the enzyme. An additional pocket at the periphery of the active site accommodates the sidechain of the P4 valine.

Alpha-factor-leader-directed secretion of recombinant human-insulin-like growth factor I from Saccharomyces cerevisiae. Precursor formation and processing in the yeast secretory pathway.

A synthetic gene coding for human-insulin-like growth factor I (IGFI) was fused to the leader sequence of yeast prepro-alpha-factor and expressed in Saccharomyces cerevisiae under the control of a glyceraldehyde-3-phosphate dehydrogenase promoter fragment. Recombinant IGFI was found inside yeast cells and secreted into the medium. The secreted IGFI migrated on SDS gels with the same electrophoretic mobility as authentic IGFI, i.e. at about 7.5 kDa. HPLC analysis of secreted IGFI revealed the presence of the correctly folded, genuine molecule as well as an isomeric byproduct of equal molecular mass but with two of the three disulfide bonds interchanged. Inside exponentially growing cells the 7.5-kDa IGFI was also found, along with up to four additional IGFI-related polypeptides of higher molecular mass. By endoglycosidase F treatment the three polypeptides between 19-26 kDa were converted to a single peptide of 17 kDa. Since this peptide also reacted with an anti-alpha-factor antibody, it represents most likely the unglycosylated alpha-factor--IGFI fusion precursor. Pulse-chase experiments established the precursor nature of the intracellular higher-molecular-mass IGFI species. Conversion of the primary translation product to the differently glycosylated IGFI precursor proteins and into the mature form occurred very rapidly, within 2 min. Rapid maturation was, however, not followed by an equally rapid secretion of the mature form into the medium: only after 30-40 min did IGFI appear outside the cells. We therefore postulate the presence of an as yet undefined Golgi or post-Golgi bottleneck representing a major obstacle in secretion of recombinant IGFI from S. cerevisiae cells.

Comparative microarray analysis of gene expression during apoptosis-induction by growth factor deprivation or protein kinase C inhibition

The transcriptional response of mouse pro-B cells to two different apoptotic stimuli was investigated. First, interleukin-3 (IL-3) deprivation was used to trigger programmed cell death in IL-3 dependent FL5.12 cells. Alternatively, cells were treated with the protein kinase C (PKC) inhibitor staurosporine. The temporal pattern of gene expression was followed with cDNA microarrays, covering over 8700 different mouse cDNA sequences corresponding to approximately 7900 unique genes. Messenger RNA levels of 315 genes were found to be regulated by more than twofold upon IL-3 removal, while 125 genes reacted to staurosporine treatment. Cross-comparison revealed an intersection of 34 genes similarly regulated in both pathways and thus representing candidates for common apoptosis regulators. For many expressed sequence tags (ESTs) our data suggest for the first time functions in the control of apoptosis, stress response or the cell cycle. IL-3 removal led to the repression of genes required for proliferation and to the induction of genes, linked to apoptotic and signaling pathways. Staurosporine caused predominantly activation of genes, some of which had previously been described to be involved in inflammation. Our findings indicate that cellular responses to both apoptotic stimuli influence various physiological pathways which had not previously been known to be linked.

A microarray-based, integrated approach to identify novel regulators of cancer drug response and apoptosis.

DNA microarrays are powerful tools for the analysis of gene expression on a genomic scale. The importance of individual regulatory events for the process under study can however not be deduced unequivocally without additional experiments. We devised a strategy to identify central regulators of cancer drug responses by combining the results of microarray experiments with efficient methods for phenotypic testing of candidate genes. We exposed murine FL5.12 pro-B cells to cisplatin, camptothecin, methotrexate or paclitaxel, respectively and analysed the patterns of gene expression with cDNA microarrays. Drug-specific regulatory events as well as intersections between different apoptotic pathways, including previously studied responses to staurosporine and interleukin-3 (IL-3) deprivation, were identified. Genes shared by at least three pathways were chosen for further analysis. Ectopic expression of three such genes, TEAP, GP49B, and Lipin1 was found to have an anti-proliferative effect on pro-B cells. Interestingly, we identified hemoglobin alpha as a strong pro-apoptotic regulator. While hemoglobin-expressing cells were growing normally in the presence of IL-3, they displayed accelerated apoptosis with similar kinetics as Bax overexpressing cells upon IL-3 removal. The pro-apoptotic effect of hemoglobin was suppressed by Bcl-2 and was characterized by enhanced stimulation of caspase activity.

Novel BNIP1 variants and their interaction with BCL2 family members.

By PCR and EST database searches we have identified three novel BNIP1 splice variants, and found that one of them, BNIP1‐b, contains a highly conserved BH3 domain. The BNIP1 gene has been assigned to chromosome 5q33–34. Using in vitro protein‐protein interaction assays, all BNIP1 variants were shown to interact with BCL2 and also with BCL2L1 (previously Bcl‐xL). These interactions are BH3‐independent. Furthermore, the BNIP1 variants cannot interact with BAX. The results suggest that the BNIP1 variants are novel members of the BCL2 family but function through a mechanism different from other BH3‐only members.

In-vitro processing of yeast α-factor leader fusion proteins using a soluble yscF (Kex2) variant

SummaryThe Saccharomyces cerevisiae KEX2 gene encodes the membrane-bound endoprotease yscF, which is responsible for the site-specific endoproteolytic cleavages at pairs of basic amino acid residues in the α-factor precursor. In order to obtain soluble yscF activity, a mutant KEX2 gene lacking 600 bp coding for the C-terminal 200 amino acids was constructed. Expression of the truncated KEX2 gene in yeast led to the secretion of an active soluble yscF protein (yscFs). The soluble yscF protein is able to efficiently cleave heterologous protein precursors in-vitro, as demonstrated for α-factor leader-hIGF1 and α-factor leader-hirudin fusion proteins.

In-vitro cleavage of a fusion protein bound to cellulose using the soluble yscFs (Kex2) variant

SummaryIn order to show site-specific cleavage of fusion proteins with an engineered soluble yscF variant, we have constructed a fusion gene encoding eglinC from Hirudo medicinalis and the cellulose-binding domain from the Cellulomonas fimi exoglucanase (Cex). The two fusion partners were separated by a Lys-Arg-containing recognition sequence for the yeast endoprotease yscF (Kex2).The fusion protein (eglinC-Cex) was expressed intracellularly in Saccharomyces cerevisiae. After disruption of the cells eglin-C-Cex was shown to bind to cellulose when present in total crude cell lysates. This step efficiently removed the majority of contaminating host proteins. While immobilized on cellulose, eglin-C-Cex could be cleaved into the expected fragments. Upon cleavage the eglinC part was released from the cellulose, while the purification tag, i.e. the cellulose binding domain, stayed bound to the cellulose matrix.

Isolation and characterization of recombinant annexin V expressed in Saccharomyces cerevisiae

The cloning, purification and characterization of full-length annexin V, expressed intracellularly in Saccharomyces cerevisiae is detailed. Following homogenization in a glass bead mill, clarification by ultracentrifugation and fractional ammonium sulfate precipitation, the 319 amino acid protein was purified by column chromatography on phenyl-Sepharose and heparin-Sepharose. Annexin V elutes on reverse phase C4 silica as a single peak with greater than 97% homogeneity and is further characterized by a molecular mass of 34 kDa from electrophoresis under reducing conditions on SDS gels. Dynamic light scattering experiments reveal annexin V exists as a monomer in solution. Amino terminal Edman degradation afforded no sequence, therefore the carbamidomethylated protein was chemically cleaved with cyanogen bromide. Separation of the resulting peptide fragments on reverse phase HPLC followed by N-terminal sequencing and electrospray mass spectrometry supported the correct sequence as well as the existence of an acetyl blocking group on the N-terminus. The protein exhibits an isoelectric point of 4.73 by column chromatofocusing. Secondary structure predictions from CD spectroscopy indicate that the molecule is correctly folded. In anticoagulant assays, the purified protein exhibits dose-response effects in activated partial thromboplastin time (APTT) prolongation and doubles the clotting time of control human plasma at 70 micrograms ml-1. More specifically, in a factor Xa inhibition assay in which the activation of factor X via the tissue factor-factor VIIa complex is monitored by the cleavage of a factor Xa chromogenic substrate, recombinant annexin V exhibits a 50% inhibitory concentration (IC50) in the low nanomolar range.

Upregulation of alpha globin promotes apoptotic cell death in the hematopoietic cell line FL5.12

The function of alpha globin in the context of oxygen transport in erythroid cells is well described. Recently the expression of alpha globin was shown to be upregulated upon specific apoptotic stimuli like cytokine deprivation or cisplatin treatment in the hematopoietic pro-B cell line, FL5.12. In contrast to alpha globin, beta globin or globin-like genes were expressed at a very low level or were not expressed at all. Further, we found that alpha globin was not associated with heme. Apoptotic cells neither produced hemoglobin nor displayed a phenotype of cells differentiating down the erythroid lineage. Also other cell lines of variable differentiation status (NIH3T3, HeLa, K562) upregulated alpha globin during treatment with apoptosis-inducing agents. Under IL-3-deprived conditions GFP-alpha globin accelerated the progression of apoptosis comparable to GFP-Bax. GFP-alpha globin was expressed at a low level and enrichment of FL5.12 cells expressing GFP-alpha globin was difficult even in the presence of IL-3. Caspase-8, -9 and -3 as well as the proapoptotic factor Bax and cytochrome c were activated. Antisense alpha globin downregulated the expression of endogenous alpha globin und reduced caspase activity. Taken together these data indicate that alpha globin is a new and crucial factor in apoptosis especially supporting the mitochondrial pathway.

Hematopoietic transcription factor GATA-2 promotes upregulation of alpha globin and cell death in FL5.12 cells

Recently we showed that alpha globin is a novel pro-apoptotic factor in programmed cell death in the pro-B cell line, FL5.12. Alpha globin was also upregulated in various other cell lines after different apoptotic stimuli. Under withdrawal of IL-3, overexpression of alpha globin accelerated apoptosis in FL5.12. Here, we have studied how transcription of alpha globin is placed in the broader context of apoptosis. We used Affymetrix chip technology and RT QPCR to compare expression patterns of FL5.12 cells growing with or without IL-3 to search for transcription factors which were concomitantly upregulated with alpha globin. The erythroid-specific transcription factor GATA-2 was the earliest and most prominently upregulated candidate. GATA-1 was expressed at low levels and was weakly induced while GATA-3 was completely absent. To evaluate the influence of GATA-2 on alpha globin expression and cell viability we overexpressed GATA-2 in FL5.12 cells. Interestingly, high expression of GATA-2 resulted in cell death and elevated alpha globin levels in FL5.12 cells. Transduction of antisense GATA-2 prevented both increase of GATA-2 and alpha globin under apoptotic conditions and delayed cell death. We suggest a role of GATA-2 in apoptosis besides its function in maintenance and proliferation of immature hematopoietic progenitors.