Band 381, Heft 7

Lysosomal degradation of the ganglioside GM2 by human βhexosaminidase A requires the presence of the GM2 activator protein as an essential cofactor. Here we demonstrate that GM2 activator mRNA is differentially expressed and mainly localized to the apical part of the epithelial cells of distal renal tubules and the collecting duct. In order to understand the mechanism underlying the regulation of the GM2 activator gene, we analyzed the genomic organization upstream exon 2 as well as the 5'flanking region. The GM2 activator gene spans about 16.8 kb with a first intron of 6.5 kb, and the transcription start is located at position -96 upstream from the ATG. DNA elements responsible for GM2 activator expression were identified in a PCRbased method of longdistance DNA walking. Sequence analysis revealed a 2.9 kb region upstream of the ATG that contained regulatory elements like CAAT boxes, Sp1 binding sites as well as AP1, and AP2 sites. Transfection experiments in COS-1 cells with a series of chimeras of 5'stepwise deletion mutants of the GM2 activator gene 5'flanking region and the secretory alkaline phosphatase (SEAP)reporter gene indicated that a genomic fragment encompassing 323 to +1bp had significant promoter activity. EMSA experiments showed that Sp1 and other transcription factors like AP1, AP2 and CCAATBox binding proteins are involved in GM2 activator gene regulation.

The transcription factor GLI1 is one of three vertebrate members of the GLI family, which is characterized by a highly conserved DNAbinding domain of five zinc fingers. We have analyzed whether human GLI1 is a target of PKA regulation. It was found that PKA inhibits GLI1 transcriptional activity. However, no evidence for proteolytic processing or for alteration in the subcellular distribution of GLI1 was obtained. The responsive PKA site (aa333-336) was localized to the second zinc finger of GLI1. Mutation of Ser336 revealed that PKA could also stimulate GLI1 transcriptional activity. Thus, our data demonstrate both negative and positive regulation of human GLI1 by PKA.

The efp gene encodes a protein that is essential for the growth and for the viability of Escherichia coli cells. Interruption of this gene results in cell death due to a defect in protein synthesis. We report here that the EFP protein, encoded by the efp gene, is required for in vitro reconstitution of polypeptide synthesis in a system programmed by a native template which contains each of the purified initiation factors, IF1, IF2, IF3; the elongation factors, EFTu, EFTs and EFG, and a protein called W that is required to eject tRNAs from ribosomes. The EFP protein is required for enhancing the rate and the extent of synthesis in the presence of all of the above factors. The EFP protein stimulates synthesis of poly(Phe) programmed with poly(rU) only if Nacetyl PhetRNA initiates the reactions under conditions that foster the dissociation of the 70S ribosome. Study of the ability of the ribosome to synthesize a number of fMetinitiated dipeptides from CCA amino acyl acceptors suggests that EFP acts to promote synthesis with acceptors that are poor donors for the the reconstituted peptidyl transferase.

An open reading frame (ORF) of snake muscle fructose 1,6-bisphosphatase (Fru-1,6-P[2]ase) was obtained by the RTPCR method with degenerate primers, followed by RACEPCR. The cDNA of Fru-1,6- P[2]ase, encoding 340 amino acids, is highly homologous to that of mammalian species, especially human muscle, with a few exceptions. Kinetic parameters of the purified recombinant enzyme, including inhibition behavior by AMP, were identical to that of the tissue form. Replacement of the Nterminal sequence of this enzyme by the corresponding region of rat liver Fru 1,6-P[2]ase shows that the activity was fully retained in the chimeric enzyme. The inhibition constant (K) of AMP at pH 7.5, however, increases sharply from 0.85M (wildtype) to 1.2mM (chimeric enzyme). AMP binding is mainly located in the Nterminal region, and the allosteric inhibition was shown not to be merely determined by the backbone of this region. The fact that the chimeric enzyme could be activated at alkaline pH by AMP indicated that the AMP activation requires the global structure beyond the area.

Cinnamomin, a new type II ribosomeinactivating protein (RIP), was found to be able to induce the release of calcein loaded in lecithin small unilamellar vesicles and the fusion or aggregation of the lecithin liposomes. Such induction could be promoted several fold by a pH 5.0 environment, a condition similar to that in endocytic vesicles. Lowering the pH from 7.5 to 5.0 evoked conformational changes of cinnamomin and unmasked its hydrophobic areas, including the exposure of 1-anilino-8-naphthalenesulfonate (1,8-ANS) binding sites of the molecule. Some tryptophan residues with affinity to acrylamide were demonstrated to participate in the lipidprotein interaction. The pH dependent fusogenicity of type II RIP might suggest its in vivo function as a fusogen to exert its cytotoxicity.

Fluorescent Nitric Oxide Cheletropic Traps (FNOCTs) were applied to specifically trap nitric oxide (NO) with high sensitivity. The fluorescent oquinoid ?electron system of the FNOCTs (? = 460 nm, ? = 600 nm) reacts rapidly with NO to a fluorescent phenanthrene system (? = 380 nm, ? = 460 nm). The cyclic nitroxides thus formed react further to nonradical products which exhibit identical fluorescence properties. Using the acid form of the trap (FNOCT-4), NO release by spermine NONOate and by lipopolysaccharide (LPS) activated alveolar macrophages were studied. A maximum extracellular release of NO of 37.5 nmol h[-1] (10[6] cells)[-1] from the macrophages was determined at 11 h after activation. Furthermore, intracellular NO release by LPSactivated macrophages and by microvascular omentum endothelial cells stimulated by the Ca[2+] ionophore A-23187, respectively, was monitored on the single cell level by means of fluorescence microscopy. After loading the cells with the membranepermeating acetoxymethylester derivative FNOCT-5,which is hydrolyzed to a nonpermeating dicarboxylate by intracellular hydrolases, NO formation by the endothelial cells started immediately upon stimulation, whereas start of NO production by the macrophages was delayed with a variation between 4 and 8 h for individual cells. These results demonstrate that the FNOCTs can be used to monitor NO release from single cells, as well as from NOdonating compounds, with high sensitivity and with temporal and spatial resolution.

Soluble proteins were isolated from leaves of the common ice plant Mesembryanthemum crystallinum L. in the CAM state of photosynthesis and tested for protease activity using amino acidβnaphthylamide (NA)derivatives in a search for proteolytic activity responsible for cleavage of the VATPase subunit B. This cleavage is suggested to occur at the peptide bond between Met192 and Glu193. At neutral pH MetNA was one of seven derivatives which were cleaved by proteases present in this fraction. Enzymes exhibiting proteolytic activity were separated from other soluble proteins by Superose 12-size exclusion FPLC. Incubation of partially purified protease with tonoplastenriched membrane vesicle fractions isolated from M. crystallinum in the C[3]state of photosynthesis led to a decrease in subunit B (55 kDa) protein amount and to the formation of the polypeptide D (32 kDa), which has been previously suggested to represent a fragment of subunit B. Cleavage of subunit B and the appearance of D also occurred during incubation of tonoplast vesicles in the presence of reactive oxygen species. In addition to D, the polypeptide E (28 kDa) appeared after incubation with protease and/or reactive oxygen species. Taken into account that D and E crossreacted with an affinity purified antiserum directed against subunit B, D as well as E might represent fragments of subunit B. These results open new perspectives with respect to the regulation of VATPase modification and turnover.

Multimerization of thyroglobulin (TG) takes place extracellularly in the thyroid follicle lumen and is regarded as a mechanism to store TG at high concentrations. Human thyroglobulin (hTG) has been shown to multimerize mainly by intermolecular disulfide crosslinks. We recently noted that TG of various mammalian species contains three highly conserved thioredoxin boxes (CXXC). This sequence is known to underlie the enzymatic activity of protein disulfide isomerase (PDI). As hTG formed intermolecular disulfide bonds in the absence of other proteins depending on the redox conditions and hTG concentration, the CXXCboxes of TG might provide the structural basis for selfassisted intermolecular crosslinking. To test this hypothesis we prepared a recombinant TG fragment containing the three thioredoxin boxes. This fragment exhibited a redox activity amounting to about 10 % of the activity of PDI at redox conditions supposed to be present in the extracellular space. This activity might be supplemented by the oxidizing system of the apical cell surfaces of thyrocytes facing the follicle lumen. Indeed, incubation of hTG with peroxidase and H[2]O[2] resulted in intermolecular disulfide bridge formation. Our results suggest a combined mechanism of selfassisted and peroxidasemediated disulfide bond formation leading to the intermolecular crosslinking of lumenal hTG.

A series of bivalent thrombin inhibitors was synthesized, consisting of a dphenylalanylprolylN?(methyl) arginyl active site blocking segment, a fibrinogen recognition exosite inhibitor part, and a peptidic linker connecting these fragments. The methylation of the P1 amino acid led to a moderate decrease in affinity compared with the unmethylated analog. In addition, it prevented the thrombin catalyzed proteolysis, independent of the P1' amino acid used. This is a significant advantage compared to the original hirulogs, which strictly require a proline as P1' amino acid to reduce the cleavage Cterminal to the arginyl residue. Several analogs were prepared by incorporation of different P1' amino acids found in natural thrombin substrates. The most potent inhibitor was I-11 [dCha ProN(Me)ArgThr(Gly)[5]DYEPIPEEAChadGlu] with a K of 37 pM. I-11 is highly selective and no inhibition of the related serine proteases trypsin, factor Xa and plasmin was observed. The stability of I-11 in human plasma in vitro was strongly improved compared to hirulog-1. In addition, a significantly reduced plasma clearance of I-11 was observed after intravenous injection in rats. Results from molecular modeling suggest that a strong reorganization of the hydrogen bonds in the active site of thrombin may result in the proteolytic stability found in this inhibitor series.

The mitochondrial FADdependent glycerol-3-phosphate dehydrogenase (mGPDH) is an essential component of the glycerol phosphate shuttle which transfers reduction equivalents from the cytoplasm into the mitochondria. We analyzed the distribution of different exon 1-containing transcripts by RTPCR in various tissues in vivo. Exon 1a was predominantly expressed in brain, brown adipose tissue and pancreas, exon 1b was ubiquitously expressed, and exon 1c was exclusively expressed in testis. In transient transfection assays the ubiquitous promoter B showed a detectable activity, whereas promoters A and C were completely silent. A deletion mutational analysis located the basal promoter B activity to a 316 bp core sequence upstream of the transcription start site.

The amino acid sequence of a novel tissueand phasespecific nuclear protein (SNP) has been determined, after purification from the nuclei of the oviduct of the lizard Podarcis sicula Raf. during the reproductive period of the seasonal growth. SNP has a pI of 9.0 and contains 81 amino acid residues with a molecular weight of 9211.88 ± 0.09. It shows a bipartite organization as the first 40 amino acids contain all 8 cysteinyl residues, while the last 41 amino acids contain 16 prolyl residues. Two more components have also been identified and characterized, with the first 79 amino acids matching SNP and missing one or two residues at the Cterminus. They have thus been named [des(Ala[81]) SNP1] and [des(Lys[80]Ala[81]) SNP2], respectively. The molecular weights are 9140.21 ± 0.83 for [des(Ala[81]) SNP1] and 9011 ± 0.09 for [des(Lys[80]Ala[81]) SNP2].

In mammals, betaine of the mitochondrial matrix is used in the cytosol by betainehomocysteine Smethyltransferase for methionine synthesis. The resulting dimethylglycine is shuttled back into the mitochondrial matrix for further degradation. Nanospray tandem mass spectrometry and Nterminal amino acid sequencing of microtubuleassociated proteins from rat liver tubulin revealed that betainehomocysteine Smethyltransferase is microtubule associated. This was confirmed by confocal laser scanning microscopy of HepG2 cells labeled with betainehomocysteine Smethyltransferase and αtubulinspecific monoclonal antibodies. The association of betainehomocysteine Smethyltransferase with the cytoskeleton may functionally integrate the mitochondrial and cytoplasmic compartments of choline degradation.

Nitric oxide (NO), produced in different cell types through the conversion of Larginine into Lcitrulline by the enzyme NO synthase, has been proposed to exert its action in several physiological and pathological events. The great propensity for nitrosothiol formation and breakdown represents a mechanism which modulates the action of macromolecules containing NOreactive Cys residues at their active centre and/or allosteric sites. Based on the human haemoglobin (Hb) structure and accounting for the known acidbase catalysed Cys?93-nitrosylation and Cys?93NOdenitrosylation processes, the putative amino acid sequence (Lys/Arg/His/Asp/Glu)Cys(Asp/Glu) (sites 1, 0, and + 1, respectively) has been proposed as the minimum consensus motif for CysNO reactivity. Although not found in human Hb, the presence of a polar amino acid residue (Gly/Ser/Thr/Cys/Tyr/Asn/Gln) at the 2 position has been observed in some NOreactive protein sequences (e.g., NMDA receptors). However, the most important component of the tri or tetrapeptide consensus motif has been recognised as the Cys(Asp/Glu) pair [Stamler et al., Neuron (1997) 18, 691 696]. Here, we analyse the threedimensional structure of several proteins containing NOreactive Cys residues, and show that their nitrosylation and denitrosylation processes may depend on the CysS? atomic structural microenvironment rather than on the tri or tetrapeptide sequence consensus motif.

The kallikreinkinin system is a mediator of inflammation in humans. In order to elucidate the range of expression of human tissue kallikrein and its substrates, high and low molecular weight kininogen, in inflammatory cells in vitro, we examined their biosynthesis in the HL-60 cell line by RTPCR and Southern blot analyses. Prominent expression of tissue kallikrein mRNA occurred in untreated promyelocytic cultures as well as in HL-60 cells that were induced to differentiate toward neutrophilic, monocytic, and macrophagic cells. Under the same inducing conditions, kininogen biosynthesis was undetectable at each differentiation state of HL-60 cultures. These results indicate that the myelomonocytic lineage of human leukocytes is a source of tissue kallikrein, which may be secreted as part of the inflammatory process.

Import of DNA from the cytoplasm into the mitochondrial matrix is an obligatory step for an in organello sitedirected mutagenesis or gene therapy approach on mitochondrial DNA diseases. In this context, we have developed an artificial DNA translocation vector that is composed of the mitochondrial signal peptide of the ornithine transcarbamylase (OTC) and a DNA moiety. While this vector is capable of directing attached passenger molecules to the mitochondrial matrix, the recognition of this artificial molecule by the endogenous mitochondrial signal peptide processing machinery as well as the cleavage of the peptide plays a pivotal role in the release of the attached DNA. To study the proteolytic processing of the artificial vector, various signal peptideDNAconjugates were treated with purified mitochondrial intermediate peptidase. When the leader peptide is directly linked to the DNA moiety without an intervening spacer, MIP processing is prevented. Cleavage of the peptide can be restored, however, when the first ten amino acid residues of the mature part of OTC are appended at the carboxyterminal end of the signal peptide. Our results show that artificial peptideDNAconjugates are recognized by the mitochondrial proteolytic machinery, and therefore an interference of the peptide with the DNA function can be excluded.