Crystal structure of NucB, a biofilm-degrading endonuclease.

Bacterial biofilms are a complex architecture of cells that grow on moist interfaces, and are held together by a molecular glue of extracellular proteins, sugars and nucleic acids. Biofilms are particularly problematic in human healthcare as they can coat medical implants and are thus a potential source of disease. The enzymatic dispersal of biofilms is increasingly being developed as a new strategy to treat this problem. Here, we have characterized NucB, a biofilm-dispersing nuclease from a marine strain of Bacillus licheniformis, and present its crystal structure together with the biochemistry and a mutational analysis required to confirm its active site. Taken together, these data support the categorization of NucB into a unique subfamily of the ßßα metal-dependent non-specific endonucleases. Understanding the structure and function of NucB will facilitate its future development into an anti-biofilm therapeutic agent.

Functional analysis of the GTPases EngA and YhbZ encoded by Salmonella typhimurium.

The S. typhimurium genome encodes proteins, designated EngA and YhbZ, which have a high sequence identity with the GTPases EngA/Der and ObgE/CgtAE of Escherichia coli. The wild-type activity of the E. coli proteins is essential for normal ribosome maturation and cell viability. In order to characterize the potential involvement of the Salmonella typhimurium EngA and YhbZ proteins in ribosome biology, we used high stringency affinity chromatography experiments to identify strongly binding ribosomal partner proteins. A combination of biochemical and microcalorimetric analysis was then used to characterize these protein:protein interactions and quantify nucleotide binding affinities. These experiments show that YhbZ specifically interacts with the pseudouridine synthase RluD (KD=2 microM and 1:1 stoichiometry), and we show for the first time that EngA can interact with the ribosomal structural protein S7. Thermodynamic analysis shows both EngA and YhbZ bind GDP with a higher affinity than GTP (20-fold difference for EngA and 3.8-fold for YhbZ), and that the two nucleotide binding sites in EngA show a 5.3-fold difference in affinity for GDP. We report a fluorescence assay for nucleotide binding to EngA and YhbZ, which is suitable for identifying inhibitors specific for this ligand-binding site, which would potentially inhibit their biological functions. The interactions of YhbZ with ribosome structural proteins that we identify may demonstrate a previously unreported additional function for this class of GTPase: that of ensuring delivery of rRNA modifying enzymes to the appropriate region of the ribosome.

Characterization of Salmonella typhimurium YegS, a putative lipid kinase homologous to eukaryotic sphingosine and diacylglycerol kinases.

Salmonella typhimurium YegS is a protein conserved in many prokaryotes. Although the function of YegS is not definitively known, it has been annotated as a potential diacylglycerol or sphingosine kinase based on sequence similarity with eukaryotic enzymes of known function. To further characterize YegS, we report its purification, biochemical analysis, crystallization, and structure determination. The crystal structure of YegS reveals a two-domain fold related to bacterial polyphosphate/ATP NAD kinases, comprising a central cleft between an N-terminal alpha/beta domain and a C-terminal two-layer beta-sandwich domain; conserved structural features are consistent with nucleotide binding within the cleft. The N-terminal and C-terminal domains of YegS are however counter-rotated, relative to the polyphosphate/ATP NAD kinase archetype, such that the potential nucleotide binding site is blocked. There are also two Ca2+ binding sites and two hydrophobic clefts, one in each domain of YegS. Analysis of mutagenesis data from eukaryotic homologues of YegS suggest that the N-terminal cleft may bind activating lipids while the C-terminal cleft may bind the lipid substrate. Microcalorimetry experiments showed interaction between recombinant YegS and Mg2+, Ca2+, and Mn2+ ions, with a weaker interaction also observed with polyphosphates and ATP. However, biochemical assays showed that recombinant YegS is endogenously neither an active diacylglycerol nor sphingosine kinase. Thus although the bioinformatics analysis and structure of YegS indicate that many of the ligand recognition determinants for lipid kinase activity are present, the absence of such activity may be due to specificity for a different lipid substrate or the requirement for activation by an, as yet, undetermined mechanism. In this regard the specific interaction of YegS with the periplasmic chaperone OmpH, which we demonstrate from pulldown experiments, may be of significance. Such an interaction suggests that YegS can be translocated to the periplasm and directed to the outer-membrane, an environment that may be required for enzyme activity.

Matriptase Induction of Metalloproteinase-Dependent Aggrecanolysis In Vitro and In Vivo: Promotion of Osteoarthritic Cartilage Damage by Multiple Mechanisms.

OBJECTIVE: To assess the ability of matriptase, a type II transmembrane serine proteinase, to promote aggrecan loss from the cartilage of patients with osteoarthritis (OA) and to determine whether its inhibition can prevent aggrecan loss and cartilage damage in experimental OA. METHODS: Aggrecan release from human OA cartilage explants and human stem cell-derived cartilage discs was evaluated, and cartilage-conditioned media were used for Western blotting. Gene expression was analyzed by real-time polymerase chain reaction. Murine OA was induced by surgical destabilization of the medial meniscus, and matriptase inhibitors were administered via osmotic minipump or intraarticular injection. Cartilage damage was scored histologically and aggrecan cleavage was visualized immunohistochemically using specific neoepitope antibodies. RESULTS: The addition of soluble recombinant matriptase promoted a time-dependent release of aggrecan (and collagen) from OA cartilage, which was sensitive to metalloproteinase inhibition and protease-activated receptor 2 antagonism. Although engineered human (normal) cartilage discs failed to release aggrecan following matriptase addition, both matrix metalloproteinase- and aggrecanase-mediated cleavages of aggrecan were detected in human OA cartilage. Additionally, while matriptase did not directly degrade aggrecan, it promoted the accumulation of low-density lipoprotein receptor-related protein 1 (LRP-1) in conditioned media of the OA cartilage explants. Matriptase inhibition via neutralizing antibody or small molecule inhibitor significantly reduced cartilage damage scores in murine OA, which was associated with reduced generation of metalloproteinase-mediated aggrecan cleavage. CONCLUSION: Matriptase potently induces the release of metalloproteinase-generated aggrecan fragments as well as soluble LRP-1 from OA cartilage. Therapeutic targeting of matriptase proteolytic activity reduces metalloproteinase activity, further suggesting that this serine proteinase may have potential as a disease-modifying therapy in OA.

Biophysical and kinetic analysis of wild-type and site-directed mutants of the isolated and native dehydroquinate synthase domain of the AROM protein.

Dehydroquinate synthase (DHQS) is the N-terminal domain of the pentafunctional AROM protein that catalyses steps 2 to 7 in the shikimate pathway in microbial eukaryotes. DHQS converts 3-deoxy-D-arabino-heptulosonate-7-phosphate (DAHP) to dehydroquinate in a reaction that includes alcohol oxidation, phosphate beta-elimination, carbonyl reduction, ring opening, and intramolecular aldol condensation. Kinetic analysis of the isolated DHQS domains with the AROM protein showed that for the substrate DAHP the difference in Km is less than a factor of 3, that the turnover numbers differed by 24%, and that the Km for NAD+ differs by a factor of 3. Isothermal titration calorimetry revealed that a second (inhibitory) site for divalent metal binding has an approximately 4000-fold increase in KD compared to the catalytic binding site. Inhibitor studies have suggested the enzyme could act as a simple oxidoreductase with several of the reactions occurring spontaneously, whereas structural studies have implied that DHQS participates in all steps of the reaction. Analysis of site-directed mutants experimentally test and support this latter hypothesis. Differential scanning calorimetry, circular dichroism spectroscopy, and molecular exclusion chromatography demonstrate that the mutant DHQS retain their secondary and quaternary structures and their ligand binding capacity. R130K has a 135-fold reduction in specific activity with DAHP and a greater than 1100-fold decrease in the kcat/Km ratio, whereas R130A is inactive.

Modulation of the ligand binding properties of the transcription repressor NmrA by GATA-containing DNA and site-directed mutagenesis.

NmrA is a negative transcription-regulating protein that binds to the C-terminal region of the GATA transcription-activating protein AreA. The proposed molecular mechanism of action for NmrA is to inhibit AreA binding to its target promoters. In contrast to this proposal, we report that a C-terminal fragment of AreA can bind individually to GATA-containing DNA and NmrA and that in the presence of a mixture of GATA-containing DNA and NmrA, the AreA fragment binds preferentially to the GATA-containing DNA in vitro. These observations are consistent with NmrA acting by an indirect route, such as by controlling entry into the nucleus. Deletion of the final nine amino acids of a C-terminal fragment of AreA does not affect NmrA binding. Wild-type NmrA binds NAD(+)(P+) with much greater affinity than NAD(P)H, despite the lack of the consensus GXXGXXG dinucleotide-binding motif. However, introducing the GXXGXXG sequence into the NmrA double mutant N12G/A18G causes an approximately 13-fold increase in the KD for NAD+ and a 2.3-fold increase for NADP+. An H37W mutant in NmrA designed to increase the interaction with the adenine ring of NAD+ has a decrease in KD of approximately 4.5-fold for NAD+ and a marginal 24% increase for NADP+. The crystal structure of the N12G/A18G mutant protein shows changes in main chain position as well as repositioning of H37, which disrupts contacts with the adenine ring of NAD+, changes which are predicted to reduce the binding affinity for this dinucleotide. The substitutions E193Q/D195N or Q202E/F204Y in the C-terminal domain of NmrA reduced the affinity for a C-terminal fragment of AreA, implying that this region of the protein interacts with AreA.

Kinetic analysis of the interaction between the QutA and QutR transcription-regulating proteins.

The QutR protein is a multidomain repressor protein that interacts with the QutA activator protein. Both proteins are active in the signal transduction pathway that regulates transcription of the quinic acid utilization (qut) gene cluster of the microbial eukaryote Aspergillus nidulans. In the presence of quinate, production of mRNA from the eight genes of the qut pathway is stimulated by the QutA activator protein. The QutR protein plays a key role in signal recognition and transduction, and a deletion analysis has shown that the N-terminal 88 amino acids are sufficient to inactivate QutA function in vivo. Using surface plasmon resonance we show here that the N-terminal 88 amino acids of QutR are able to bind in vitro to a region of QutA that genetic analysis has previously implicated in transcription activation. We further show that increasing the concentration of a full-length (missense) mutant QutR protein in the original mutant strain can restore its repressing function. This is interpreted to mean that the qutR mutation in this strain increases the equilibrium dissociation constant for the interaction between QutA and QutR. We propose a model in which the QutA and QutR proteins are in dynamic equilibrium between bound (transcriptionally inactive) and unbound (transcriptionally active) states.

Interaction between relaxase MbeA and accessory protein MbeC of the conjugally mobilizable plasmid ColE1.

MbeA and MbeC are two key proteins in plasmid ColE1 conjugal mobilization. Isothermal titration calorimetry was used to detect and quantify an interaction between MbeA and MbeC. As a result of this interaction, the affinity of MbeA for single stranded DNA increased. The interaction was confirmed in vivo using a bacterial two-hybrid system, which revealed that MbeA-MbeC complexes are formed through the amino-terminal region of MbeA and the carboxy-terminal region of MbeC. To the best of our knowledge, this is the first report of direct interactions between conjugative proteins encoded by a mobilizable plasmid.

The transcription repressor NmrA is subject to proteolysis by three Aspergillus nidulans proteases.

The role of specific cleavage of transcription repressor proteins by proteases and how this may be related to the emerging theme of dinucleotides as cellular signaling molecules is poorly characterized. The transcription repressor NmrA of Aspergillus nidulans discriminates between oxidized and reduced dinucleotides, however, dinucleotide binding has no effect on its interaction with the zinc finger in the transcription activator AreA. Protease activity in A. nidulans was assayed using NmrA as the substrate, and was absent in mycelium grown under nitrogen sufficient conditions but abundant in mycelium starved of nitrogen. One of the proteases was purified and identified as the protein Q5BAR4 encoded by the gene AN2366.2. Fluorescence confocal microscopy showed that the nuclear levels of NmrA were reduced approximately 38% when mycelium was grown on nitrate compared to ammonium and absent when starved of nitrogen. Proteolysis of NmrA occurred in an ordered manner by preferential digestion within a C-terminal surface exposed loop and subsequent digestion at other sites. NmrA digested at the C-terminal site was unable to bind to the AreA zinc finger. These data reveal a potential new layer of control of nitrogen metabolite repression by the ordered proteolytic cleavage of NmrA. NmrA digested at the C-terminal site retained the ability to bind NAD(+) and showed a resistance to further digestion that was enhanced by the presence of NAD(+). This is the first time that an effect of dinucleotide binding to NmrA has been demonstrated.

Structural and functional characterization of Salmonella enterica serovar Typhimurium YcbL: an unusual Type II glyoxalase.

YcbL has been annotated as either a metallo-ß-lactamase or glyoxalase II (GLX2), both members of the zinc metallohydrolase superfamily, that contains many enzymes with a diverse range of activities. Here, we report crystallographic and biochemical data for Salmonella enterica serovar Typhimurium YcbL that establishes it as GLX2, which differs in certain structural and functional properties compared with previously known examples. These features include the insertion of an α-helix after residue 87 in YcbL and truncation of the C-terminal domain, which leads to the loss of some recognition determinants for the glutathione substrate. Despite these changes, YcbL has robust GLX2 activity. A further difference is that the YcbL structure contains only a single bound metal ion rather than the dual site normally observed for GLX2s. Activity assays in the presence of various metal ions indicate an increase in activity above basal levels in the presence of manganous and ferrous ions. Thus, YcbL represents a novel member of the GLX2 family.

Dinucleotide-sensing proteins: linking signaling networks and regulating transcription.

Differential binding of dinucleotides to key regulatory proteins can modulate their interactions with other proteins and, in some cases, can signal fluctuations in the cellular redox state, to produce changes in transcription and physiological state. The dinucleotide-binding proteins human HSCARG and yeast transcription repressor Gal80p are examples that offer exciting glimpses into the potential for dinucleotide-sensing proteins to couple fluctuations in dinucleotide ratios to changes in transcription and to act as networking agents linking different classes of signaling molecules.

Structural analysis of the recognition of the negative regulator NmrA and DNA by the zinc finger from the GATA-type transcription factor AreA.

Amongst the most common protein motifs in eukaryotes are zinc fingers (ZFs), which, although largely known as DNA binding modules, also can have additional important regulatory roles in forming protein:protein interactions. AreA is a transcriptional activator central to nitrogen metabolism in Aspergillus nidulans. AreA contains a GATA-type ZF that has a competing dual recognition function, binding either DNA or the negative regulator NmrA. We report the crystal structures of three AreA ZF-NmrA complexes including two with bound NAD(+) or NADP(+). The molecular recognition of AreA ZF-NmrA involves binding of the ZF to NmrA via hydrophobic and hydrogen bonding interactions through helices alpha1, alpha6 and alpha11. Comparison with an earlier NMR solution structure of AreA ZF-DNA complex by overlap of the AreA ZFs shows that parts of helices alpha6 and alpha11 of NmrA are positioned close to the GATA motif of the DNA, mimicking the major groove of DNA. The extensive overlap of DNA with NmrA explains their mutually exclusive binding to the AreA ZF. The presence of bound NAD(+)/NADP(+) in the NmrA-AreaA ZF complex, however, causes minimal structural changes. Thus, any regulatory effects on AreA function mediated by the binding of oxidised nicotinamide dinucleotides to NmrA in the NmrA-AreA ZF complex appear not to be modulated via protein conformational rearrangements.

The affinity of a major Ca2+ binding site on GRP78 is differentially enhanced by ADP and ATP.

GRP78 is a major protein regulated by the mammalian endoplasmic reticulum stress response, and up-regulation has been shown to be important in protecting cells from challenge with cytotoxic agents. GRP78 has ATPase activity, acts as a chaperone, and interacts specifically with other proteins, such as caspases, as part of a mechanism regulating apoptosis. GRP78 is also reported to have a possible role as a Ca2+ storage protein. In order to understand the potential biological effects of Ca2+ and ATP/ADP binding on the biology of GRP78, we have determined its ligand binding properties. We show here for the first time that GRP78 can bind Ca2+, ATP, and ADP, each with a 1:1 stoichiometry, and that the binding of cation and nucleotide is cooperative. These observations do not support the hypothesis that GRP78 is a dynamic Ca2+ storage protein. Furthermore, we demonstrate that whereas Mg2+ enhances GRP78 binding to ADP and ATP to the same extent, Ca2+ shows a differential enhancement. In the presence of Ca2+, the KD for ATP is lowered approximately 11-fold, and the KD for ADP is lowered around 930-fold. The KD for Ca2+ is lowered approximately 40-fold in the presence of ATP and around 880-fold with ADP. These findings may explain the biological requirement for a nucleotide exchange factor to remove ADP from GRP78. Taken together, our data suggest that the Ca2+-binding property of GRP78 may be part of a signal transduction pathway that modulates complex interactions between GRP78, ATP/ADP, secretory proteins, and caspases, and this ultimately has important consequences for cell viability.

Immunogenicity of a Yersinia pestis vaccine antigen monomerized by circular permutation.

Caf1, a chaperone-usher protein from Yersinia pestis, is a major protective antigen in the development of subunit vaccines against plague. However, recombinant Caf1 forms polymers of indeterminate size. We report the conversion of Caf1 from a polymer to a monomer by circular permutation of the gene. Biophysical evaluation confirmed that the engineered Caf1 was a folded monomer. We compared the immunogenicity of the engineered monomer with polymeric Caf1 in antigen presentation assays to CD4 T-cell hybridomas in vitro, as well as in the induction of antibody responses and protection against subcutaneous challenge with Y. pestis in vivo. In C57BL/6 mice, for which the major H-2(b)-restricted immunodominant CD4 T-cell epitopes were intact in the engineered monomer, immunogenicity and protective efficacy were preserved, although antibody titers were decreased 10-fold. Disruption of an H-2(d)-restricted immunodominant CD4 T-cell epitope during circular permutation resulted in a compromised T-cell response, a low postvaccination antibody titer, and a lack of protection of BALB/c mice. The use of circular permutation in vaccine design has not been reported previously.

GTP cyclohydrolase II structure and mechanism.

GTP cyclohydrolase II converts GTP to 2,5-diamino-6-beta-ribosyl-4(3H)-pyrimidinone 5'-phosphate, formate and pyrophosphate, the first step in riboflavin biosynthesis. The essential role of riboflavin in metabolism and the absence of GTP cyclohydrolase II in higher eukaryotes makes it a potential novel selective antimicrobial drug target. GTP cyclohydrolase II catalyzes a distinctive overall reaction from GTP cyclohydrolase I; the latter converts GTP to dihydroneopterin triphosphate, utilized in folate and tetrahydrobiopterin biosynthesis. The structure of GTP cyclohydrolase II determined at 1.54-A resolution reveals both a different protein fold to GTP cyclohydrolase I and distinctive molecular recognition determinants for GTP; although in both enzymes there is a bound catalytic zinc. The GTP cyclohydrolase II.GMPCPP complex structure shows Arg(128) interacting with the alpha-phosphonate, and thus in the case of GTP, Arg(128) is positioned to act as the nucleophile for pyrophosphate release and formation of the proposed covalent guanylyl-GTP cyclohydrolase II intermediate. Tyr(105) is identified as playing a key role in GTP ring opening; it is hydrogen-bonded to the zinc-activated water molecule, the latter being positioned for nucleophilic attack on the guanine C-8 atom. Although GTP cyclohydrolase I and GTP cyclohydrolase II both use a zinc ion for the GTP ring opening and formate release, different residues are utilized in each case to catalyze this reaction step.

The QUTA activator and QUTR repressor proteins of Aspergillus nidulans interact to regulate transcription of the quinate utilization pathway genese.

Genetic evidence suggests that the activity of the native QUTA transcription activator protein is negated by the action of the QUTR transcription repressor protein. When Aspergillus nidulans was transformed with plasmids containing the wild-type qutA gene, transformants that constitutively expressed the quinate pathway enzymes were isolated. The constitutive phenotype of these transformants was associated with an increased copy number of the transforming qutA gene and elevated qutA mRNA levels. Conversely, when A. nidulans was transformed with plasmids containing the qutR gene under the control of the constitutive pgk promoter, transformants with a super-repressed phenotype (unable to utilize quinate as a carbon source) were isolated. The super-repressed phenotype of these transformants was associated with an increased copy number of the transforming qutR gene and elevated qutR mRNA levels. These copy-number-dependent phenotypes argue that the levels of the QUTA and QUTR proteins were elevated in the high-copy-number transformants. When diploid strains were formed by combining haploid strains that contained high copy numbers of either the qutA gene (constitutive phenotype) or the qutR gene (super-repressing; non-inducible phenotype), the resulting diploid phenotype was one of quinate-inducible production of the quinate pathway enzymes, in a manner similar to wild-type. The simplest interpretation of these observations is that the QUTR repressor protein mediates its repressing activity through a direct interaction with the QUTA activator protein. Other possible interpretations are discussed in the text. Experiments in which truncated versions of the QUTA protein were produced in the presence of a wild-type QUTA protein indicate that the QUTR repressor protein recognizes and binds to the C-terminal half of the QUTA activator protein.

Domain structure and function within the QUTA protein of Aspergillus nidulans: implications for the control of transcription.

QUTA is a positively acting regulatory protein that regulates the expression of the eight genes comprising the quinic acid utilization gene (qut) gene cluster in Aspergillus nidulans. It has been proposed that the QUTA protein is composed of two domains that are related to the N-terminal two domains-dehydroquinate (DHQ) synthase and 5-enolpyruvyl shikimate-3-phosphate (EPSP) synthase-of the pentadomain AROM protein. The AROM protein is an enzyme catalysing five consecutive steps in the shikimate pathway, two of which are common to the qut pathway. A genetic and molecular analysis of non-inducible qutA mutants showed that all 23 mutations analysed map within the N-terminal half of the encoded QUTA protein. One dominant mutation (qutA382) introduces a stop codon at the boundary between the two domains that were identified on the basis of amino acid sequence alignments between the QUTA protein and the N-terminal two domains of the pentafunctional AROM protein. The truncated protein encoded by mutant qutA382 has DNA-binding ability but no transcription activation function. A second dominant mutation (in strain qutA214) is missense, changing 457E-->K in a region of localized high negative charge and potentially identifies a transcription activation domain in the N-terminus of the EPSP-synthase-like domain of the QUTA protein. A series of qualitative and quantitative Northern blot experiments with mRNA derived from wild-type and mutant qutA strains supported the view that the QUTA protein regulates the expression of the qut gene cluster, including the qutA gene which encodes it. A series of Western blot and zinc-binding experiments demonstrated that a putative zinc binuclear cluster motif located within the N-terminus of the QUTA protein is able to bind zinc in vitro.

The negative transcriptional regulator NmrA discriminates between oxidized and reduced dinucleotides.

NmrA, a transcription repressor involved in the regulation of nitrogen metabolism in Aspergillus nidulans,is a member of the short-chain dehydrogenase reductase superfamily. Isothermal titration calorimetry and differential scanning calorimetry have been used to show NmrA binds NAD+ and NADP+ with similar affinity (average KD 65 microM) but has a greatly reduced affinity for NADH and NADPH (average KD 6.0 mM). The structure of NmrA in a complex with NADP+ reveals how repositioning a His-37 side chain allows the different conformations of NAD+ and NADP+ to be accommodated. Modeling NAD(P)H into NmrA indicated that steric clashes, attenuation of electrostatic interactions, and loss of aromatic ring stacking can explain the differing affinities of NAD(P)+/NAD(P)H. The ability of NmrA to discriminate between the oxidized and reduced forms of the dinucleotides may be linked to a possible role in redox sensing. Isothermal titration calorimetry demonstrated that NmrA and a C-terminal fragment of the GATA transcription factor AreA interacted with a 1:1 stoichiometry and an apparent KD of 0.26 microM. NmrA was unable to bind the nitrogen metabolite repression signaling molecules ammonium or glutamine.

PDZ domains facilitate binding of high temperature requirement protease A (HtrA) and tail-specific protease (Tsp) to heterologous substrates through recognition of the small stable RNA A (ssrA)-encoded peptide.

The Escherichia coli protease HtrA has two PDZ domains, and sequence alignments predict that the E. coli protease Tsp has a single PDZ domain. PDZ domains are composed of short sequences (80-100 amino acids) that have been implicated in a range of protein:protein interactions. The PDZ-like domain of Tsp may be involved in binding to the extreme COOH-terminal sequence of its substrate, whereas the HtrA PDZ domains are involved in subunit assembly and are predicted to be responsible for substrate binding and subsequent translocation into the active site. E. coli has a system of protein quality control surveillance mediated by the ssrA-encoded peptide tagging system. This system tags misfolded proteins or protein fragments with an 11-amino acid peptide that is recognized by a battery of cytoplasmic and periplasmic proteases as a degradation signal. Here we show that both HtrA and Tsp are able to recognize the ssrA-encoded peptide tag with apparent K(D) values of approximately 5 and 390 nm, respectively, and that their PDZ-like domains mediate this recognition. Fusion of the ssrA-encoded peptide tag to the COOH terminus of a heterologous protein (glutathione S-transferase) renders it sensitive to digestion by Tsp but not HtrA. These observations support the prediction that the HtrA PDZ domains facilitate substrate binding and the differential proteolytic responses of HtrA and Tsp to SsrA-tagged glutathione S-transferase are interpreted in terms of the structure of HtrA.