|Home | About | Journals | Submit | Contact Us | Français|
SSO1273 of Sulfolobus solfataricus was identified as a cell surface-bound protein by a proteomics approach. Sequence inspection of the genome revealed that the open reading frame of sso1273 is associated in an operon-like structure with genes encoding all the remaining components of a canonical protein-dependent ATP-binding cassette (ABC) transporter. sso1273 gene expression and SSO1273 protein accumulation on the cell surface were demonstrated to be strongly induced by the addition of a peptide mixture (tryptone) to the culture medium. The native protein was obtained in multimeric form, mostly hexameric, under the purification conditions used, and it was characterized as an oligopeptide binding protein, named S. solfataricus OppA (OppASs). OppaASs possesses typical sequence patterns required for glycosylphosphatidylinositol lipid anchoring, resulting in an N-linked glycoprotein with carbohydrate moieties likely composed of high mannose and/or hybrid complex carbohydrates. OppASs specifically binds oligopeptides and shows a marked selectivity for the amino acid composition of substrates when assayed in complex peptide mixtures. Moreover, a truncated version of OppASs, produced in recombinant form and including the putative binding domain, showed a low but significant oligopeptide binding activity.
Sulfolobus solfataricus is an obligate aerobe that grows in hot and acidic environments either chemolithotrophically by oxidizing metal cations (Fe2+ or S−) or heterotrophically on simple sugars. It originates from a solfataric field with temperatures between 75°C and 90°C and pH values of 1.0 to 3.0 (9, 15). Within its environment, Sulfolobus can interact with a complex ecosystem consisting of a variety of primary producers and decomposers of organic matter. Moreover, biotopes such as the solfataric field of Sulfolobus contain decomposing materials of higher plants, including cellulose, starch, and proteinaceous compounds, that can act as potential carbon sources. Although S. solfataricus has been reported to grow on a wide variety of reduced organic compounds as the sole carbon and energy source (15), the nutrient utilization by this microorganism requires complex mechanisms of uptake and metabolism that are not yet well defined.
Numerous efforts have been directed toward the identification of the carbohydrate utilization strategy in this hyperthermophilic archaeon (18, 23). The metabolic pathways for the degradation of a variety of sugars have been studied in detail and provide evidence that S. solfataricus predominantly uses binding-protein-dependent ABC transporters for the uptake of carbohydrate compounds (1, 2, 13).
Archaeal ABC uptake systems are divided into two main classes: the carbohydrate (CUT) and the di-/oligopeptide uptake transporter classes (2). These transporter families use ATP hydrolysis to drive a unidirectional accumulation of solutes into the cytoplasm. The translocator components are composed of two integral membrane proteins, two peripheral membrane proteins that bind and hydrolyze ATP, and an extracellular substrate-binding protein (SBP). The SBP subunit captures and delivers the substrate to the translocon, and it is therefore considered to be one of the determinants of the transport specificity (2, 7, 10).
All sequenced genomes of archaea and thermophilic bacteria contain a large number of genes encoding putative ABC transport systems involved in the uptake of organic solutes. The preference of hyperthermophiles for ABC-type transporters could be important for the survival strategy in their natural habitat. In the nutrient-poor environments, such as hydrothermal vents or sulfuric hot springs, in which these organisms thrive, ABC transporters have the advantage that they can scavenge solutes at very low concentrations due to the high binding affinities of their SBP components. Furthermore, these transporters can catalyze translocation at a high rate, resulting in high internal concentrations of solutes. In contrast, secondary transport systems exhibit lower binding affinities, which make these systems less suitable for growth in extreme environments.
So far, attempts to predict the functional specificity of the ABC transporters using computational tools have been largely unsuccessful (2, 13, 20). For example, some characterized archaeal sugar transporters, based on the sequence identity and domain organization, were predicted to be di-/oligopeptide transporters (13, 20). These include the cellobiose/β-glucoside transporter system of Pyrococcus furiosus (20) and the maltose/maltodextrin and cellobiose/cello-oligomer transporters of S. solfataricus (13). However, genes encoding sugar-metabolizing enzymes are located in the vicinity of all these transport systems, suggesting that the location of the ABC operon can support the specific transport function.
Like oligopeptide binding proteins, MalE and CbtA bind a broad range of polymeric substrates (13, 20). In contrast, sugar-binding proteins usually exhibit a narrow substrate specificity that is often limited to monosaccharides. Therefore, it may well be that the substrate binding pocket of CbtA and MalE resembles that of the OppA family of binding proteins that can accommodate a range of short and long oligopeptides.
S. solfataricus contains 37 putative ABC transporters at the genome level (TransportDB, Genomic Comparisons of Membrane Transport systems [http://www.membranetransport.org/index.html]), but only a few of these systems have been functionally characterized. It is interesting that all of these are implicated in the uptake of mono-/oligosaccharides (1, 13, 20, 25).
The present work describes the isolation and characterization of the first functional ABC substrate binding protein from S. solfataricus belonging to the di-/oligopeptide transporter family, named S. solfataricus OppA (OppASs). We demonstrate that OppASs is an outer-cell-surface-anchored protein and that its expression is highly induced in the presence of a source of peptides in the culture broth. Furthermore, in vitro substrate specificity studies using complex oligopeptide mixtures indicate that OppASs is highly selective in peptide recognition.
Trypsin, dithiothreitol, iodoacetamide, NH4HCO3, bradykinin, xenopsin, and sodium deoxycholate were purchased from Sigma (St. Louis, MO). Acetonitrile, formic acid, and trifluoroacetic acid (TFA) were high-performance liquid chromatography (HPLC) grade and were from Romil. All other reagents and solvents of the highest purity were available from Baker. [3H]bradykinin (B3H) was from PerkinElmer.
S. solfataricus strain P2 (DSM 1617) was aerobically grown at 80°C in glycine-buffered Brock's medium (9) with 0.05% yeast extract and 0.2% sucrose at pH 3.2 (YS basal medium) or in YS medium supplemented with 0.2% tryptone (TYS medium). Escherichia coli BL21(DE3) RIL and the vector pET-28c(+) that produces a fusion protein with a His tag were purchased from Novagen (Madison). T4 DNA ligase and restriction endonucleases were purchased from New England BioLabs.
S. solfataricus cells (1 liter) were harvested from YS or TYS cultures at an optical density at 600 nm (OD600) of 1.2 and resuspended in 20 ml of 20 mM Tris-HCl, pH 6.5, containing 0.7 mM phenylmethylsulfonyl fluoride (PMSF). Cells were lysed by sonication (eight cycles with a 15-s pulse and 45 s off) using a Soniprep (B. Braun Labsonic U) equipped with a macroprobe. Unbroken cells were removed by low-spin centrifugation at 2,000 × g at 4°C for 20 min. Membranes were collected by ultracentrifugation at 100,000 × g at 4°C for 45 min. Pellets were washed four times with 20 mM Tris-HCl (pH 6.5), resuspended in 20 mM Tris-HCl (pH 6.5), and solubilized by incubation for 30 min at 37°C in the presence of 1% Triton X-100 and 0.7 mM PMSF. Insoluble material was removed by centrifugation at 250,000 × g at 4°C for 15 min, and the collected supernatants were harvested and extensively dialyzed against 20 mM Tris-HCl (pH 6.5) and 0.05% sodium deoxycholate. Protein concentration was determined by the Bradford assay (8).
Aeropyrum pernix OppA (OppAAp) was obtained as reported by Palmieri et al. (27).
Proteins of membrane surface fractions were separated by 8% SDS-PAGE (20 cm by 20 cm) according to the method of Laemmli (22). The gel was stained with the Coomassie brilliant blue G250 (Bio-Rad).
In-gel digestion of selected protein bands was performed according to Shevchenko et al. (31).
Peptide mixtures were analyzed by a quadrupole time-of-flight instrument (Q-Star Elite; Applied Biosystems) coupled online to a nano-HPLC system (Ultimate 3000; Dionex). The sample (5 μl out of 40 μl) was concentrated and desalinated using a C18 reverse-phase PepMap precolumn (length, 5 mm; pore size, 300 Å) (LC Packings, Sunnyvale, CA) and 2% acetonitrile in 0.1% formic acid-0.025% trifluoroacetic acid (TFA) at a flow rate of 30 μl/min for 5 min. Peptides were then separated by a C18 reverse-phase capillary column (length, 15 length; internal diameter, 75 μm; pore size, 300 Å) (LC Packings, Sunnyvale, CA) at a flow rate of 300 nl/min using a linear gradient of eluent B (98% acetonitrile in 0.1% formic acid and 0.025% TFA) in eluent A (2% acetonitrile in 0.1% formic acid and 0.025% TFA) from 5 to 50% in 30 min. A pulled silica capillary (outer diameter, 170μm; internal diameter, 100 μm; tip, 30-μm internal diameter) was used as a nanoflow tip. Collision-induced dissociation experiments were carried out in information-dependent acquisition (IDA) mode in the m/z acquisition range of 70 to 1,500. Nitrogen was used as the collision gas. Precursor ions were chosen as the two most intense peaks of each MS1 scan. Dynamic collision energy was applied depending on the mass and the charge of the precursor ion. Analyst QS, version 2.0, software (Applied Biosystems) was used to analyze raw data and to generate peak lists. Peak lists were used for tandem mass spectrometry (MS/MS) ion search in the Mascot Server, version 2.2, for protein identification. The Mascot Server was set up to search the NCBI nonredundant (nr) database with Archaeobacteria-specified taxonomy containing 177,984 sequences. Trypsin was specified as the digestion enzyme with a maximum of one missed cleavage site. Mascot was run with a fragment ion mass tolerance of 0.08 Da and a parent ion tolerance of 50 ppm. The carbamidomethyl derivative of cysteine was specified as a fixed modification while oxidation of methionine was specified as a variable modification in the Mascot search. Two independent nano-HPLC-electrospray ionization (ESI)-MS/MS experiments were performed for each sample.
The purification of SSO1273 was performed by SDS-PAGE electro-elution of the protein from an 8.0% polyacrylamide gel (1.5 mm thick; 20 by 20 cm) loaded with the surface membrane protein fraction of TYS culture cells. The sample was heated at 70°C for 20 min. Gel electrophoresis was carried out at 14 mA overnight at 4°C. A vertical strip of the gel was stained with Coomassie brilliant blue and used as the reference to cut the band of interest from the untreated portion of the gel. The excised gel band was subjected to electro-elution for 2 h at 50 mA and 25°C in a dialysis membrane tube containing 1 ml of SDS PAGE running buffer. After removal of SDS by extensive dialysis against 20 mM Tris-HCl (pH 6.5)-0.05% sodium deoxycholate, the purified protein was stored at 4°C for further analysis.
The molecular mass of the purified SSO1273 under native conditions was determined by gel filtration chromatography on a Superose 12 HR 10/30 column (Pharmacia Biotech) connected to the AKTA FPLC (fast protein liquid chromatography) Explorer system (Amersham Biosciences) and equilibrated in 25 mM Tris-HCl (pH 7.5)-0.05% sodium deoxycholate. The column was calibrated using a set of gel filtration protein standards (Bio-Rad), including thyroglobulin (670 kDa), bovine gamma globulin (158 kDa), chicken ovalbumin (44.0 kDa), equine myoglobin (17.0 kDa), and vitamin B12 (1.35 kDa).
Automated N-terminal degradation of the SSO1273 protein electroblotted onto polyvinylidene difluoride (PVDF) membrane (Bio-Rad) was performed using a Perkin-Elmer Applied Biosystems 477A pulsed-liquid protein sequencer.
The carbohydrate moiety analysis of SSO1273 was performed by lectin assay (Boehringer Manneheim). Purified SSO1273 and control standard glycoproteins (1 μg; supplied with the Boehringer glycan differentiation kit) were directly spotted onto PVDF membranes (Bio-Rad) and detected immunologically after they were bound to lectins conjugated with digoxigenin, following the manufacturer's instructions. All experiments were performed in duplicate on two different protein preparations.
Peptide binding activity of SSO1273 was tested using the octapeptide xenopsin or the nonapeptide bradykinin as ligands according to the assay procedure described by Palmieri et al. (27), with some modifications. Purified protein (0.2 nmol) was added to 50 mM sodium citrate buffer, pH 3.0, containing different amounts of the peptide substrate (from 0.2 to 1.0 nmol) in a final volume of 0.22 ml. The mixtures were incubated at 70°C for 30 min and successively put on ice to stop the reactions. Mixtures of the peptide substrates incubated with an equimolar concentration of bovine serum albumin (BSA) or trehalose binding protein from S. solfataricus were used as negative controls for peptide binding.
Peptide binding activity was evaluated by a direct binding assay, using the radio-ligand B3H as the peptide substrate. B3H (0.3 pmol) was incubated with different amounts of SSO1273 (from 1 to 5 pmol) in sodium citrate buffer, pH 3.0, in a final volume of 0.1 ml. The mixture was incubated at 70°C for 30 min and successively put on ice to stop the reaction. After incubation, unbound B3H was separated from the complex SSO1273-B3H by gel filtration chromatography using a prepacked Sephadex G-50 nick column (GE Healthcare). Control B3H sample incubated in the absence of SSO1273 was run in parallel. Fractions of 0.2 ml were collected and added with 1 ml of scintillation liquid (Ultima Gold; Camberra Packard). The radioactivity of each fraction was measured by a liquid scintillation counter (Camberra Packard). A mixture of B3H and BSA (12 pmol), treated in the same way, was used as a negative control.
The casein tryptic digest was prepared by dissolving 1.2 mg of bovine caseins (Sigma) in 1 ml of 50 mM NH4HCO3-2% acetonitrile, followed by in-solution enzymatic digestion using 0.06 mg of trypsin from bovine pancreas (Sigma) at 37°C for 2 h. The tryptic digest was ultrafiltered through a Microcon YM (Millipore) membrane with a cutoff of 10 kDa at 4°C and 12,000 × g. The resulting peptides in the eluate mixture (6 μg) were incubated with the SSO1273 protein (20 pmol) at 70°C for 30 min in 50 mM sodium citrate buffer, pH 3.0. After incubation, the SSO1273-peptide complexes were isolated by gel filtration chromatography on a Superdex 200 column connected to a Smart System (Pharmacia) and equilibrated with 20 mM ammonium bicarbonate buffer, pH 7.5. The collected fractions were analyzed by nano-HPLC-ESI-MS/MS as described above. A casein tryptic digest incubated in the absence of SSO1273 and subjected to the same treatment was used as a control sample for the identification of the detectable peptides.
The same procedure (the incubation buffer was changed to 50 mM Tris-HCl, pH 7.5) was adopted to determine the substrate specificity of the archaeal ortholog of SSO1273, the APE1583 from A. pernix (27). The two assays were performed in parallel using the same preparation of casein tryptic digest described above. The experiments were performed in duplicate on three different protein preparations.
, the encoding region of the sso1273 gene between the positions +91 and +2489 bp was cloned lacking the sequences coding for the signal peptide at the 5′ end (nucleotides 1 to 90) and the threonine-rich domain at the 3′ end (deletion of 232 bp [sso1273Δ232]). The sso1273Δ232 gene was obtained by PCR from S. solfataricus P2 chromosomal DNA (4). The forward and reverse primers used in the amplification procedure were AppAup (5′-CAATCTGCCAGTTCCTCGCCTG-3′) and AppAdw (5′-GTTGTAGTGCTCGAGAGTTGTTATGGATGG-3′) containing an XhoI restriction site (underlined). AppAup was the primer starting with the 31st sso1273 translation codon. The XhoI restriction site in AppAdw was introduced to allow the insertion into the expression vector.
Amplification was performed for 40 cycles in a PCR Eppendorf apparatus under the conditions described by Saiki (29) using the enzyme Phusion DNA Polymerase (Finnzymes) and an annealing temperature of 65°C. The PCR product of ~2,400 bp was purified by a QIAquick PCR purification kit. After digestion with XhoI, the resulting DNA fragment was cloned into the NcoI/XhoI site of the expression vector pET-28c(+) (Novagen, Darmstadt, Germany), which had been linearized with NcoI, treated with Klenow DNA polymerase for filling in, and digested with XhoI.
The final construct was verified by sequencing. The resulting pET-sso1273Δ232 expression plasmid was used to transform competent E. coli BL21-CodonPlus(DE3)-RIL cells (Stratagene) according to the calcium chloride procedure described by Sambrook and Russell (30).
The cloning of a region coding for a putative binding domain from the sso1273 gene (between the positions +91 and +2079 bp) lacking a large portion of 642 bp (sso1273Δ642) at the 3′ end and the sequence coding for the signal peptide at the 5′ end, was also carried out. The amplification was performed for 40 cycles at an annealing temperature of 65°C using the chromosomal DNA template (50 ng) and the forward and reverse primers 5′-CAATCTGCCAGTTCCTCGCCTG-3′ and 5′-GGAGTCGACACTCGGATTCCACGCTA-3′, respectively, containing an SalI restriction site (underlined). The amplified digested fragment of ~2,000 bp was inserted into pET-28c(+), pretreated as for the cloning of the larger DNA fragment. The resulting pET28c-sso1273Δ642 vector was used to transform competent E. coli BL21-CodonPlus(DE3)-RIL cells.
Transformant E. coli cells were grown in LB medium containing 25 μg/ml kanamycin at 30°C. The overnight culture was diluted 100-fold in fresh LB medium, and cells were grown at 22°C. When the culture reached an OD600 of 0.6, zinc sulfate (0.25 mM) was added, and induction was performed with 0.4 mM isopropyl-β-d-thiogalactopyranoside (IPTG). After 4 h of growth, cells were lysed by sonication in lysis buffer (25 mM Tris-HCl [pH 7.5], 0.7 mM PMSF, 0.05% sodium deoxycholate), followed by centrifugation at 12,000 × g for 1 h to recover the supernatant used for protein purification.
During the purification, the recombinant proteins were monitored by SDS-PAGE analysis. The sample was mixed with 2 ml of His Select HF Nickel Affinity Gel (Sigma), preequilibrated in 25 mM Tris-HCl, pH 7.5 (buffer A), and gently shaken at 4°C overnight; the mixture was then packed into a column washed with buffer A, and the bound proteins were eluted by increasing stepwise histidine concentrations (10 mM, 30 mM, and 250 mM). Fractions containing the protein of interest were pooled, dialyzed against 25 mm Tris-HCl (pH 7.5), 40 mM l-histidine, and 0.05% sodium deoxycholate and applied onto a Superose 12 HR 10/30 column (Pharmacia Biotech) eluted with 25 mM Tris-HCl (pH 7.5)-0.05% sodium deoxycholate. Further purification was performed by hydrophobic chromatography on a HiTrap Phenyl Sepharose HP column equilibrated with 25 mM Tris-HCl (pH 7.5)-0.1 M ammonium sulfate buffer. A linear gradient of ammonium sulfate from 0.1 M to 0 in buffer A was applied for the elution of bound proteins. The purified recombinant protein was dialyzed against 25 mM Tris-HCl (pH 7.5), 40 mM l-histidine, and 0.05% sodium deoxycholate.
The N-terminal sequence analysis confirmed the identity of both the purified recombinant proteins SSO1273Δ232 and SSO1273Δ642.
Peptide binding activity of the recombinant SSO1273Δ642 was evaluated using the radio-ligand B3H, as described above.
S. solfataricus P2 cells were harvested when the cultures reached middle-late exponential phase and early stationary phase, namely, an absorbance at an OD600 of about 0.7 and 1.2, respectively. Total RNA was extracted by the guanidine thiocyanate method (30). For Northern blot analysis, the different RNAs extracted (14 μg) were electrophoretically separated together with molecular weight RNA standards (ready-to-use RiboRuler High Range RNA Ladder, 200 to 6,000 bases; Fermentas) in 1.2% agarose gel (12 by 15 cm) containing 10% formaldehyde and blotted onto a Hybond-XP nylon membrane (Amersham). Hybridization with α-32P-labeled sso1273 probe (a 934-bp Nco/ScaI restriction fragment of the coding sequence) was carried out at 65°C for 16 h in 5× Denhardt's reagent, 6× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 0.5% SDS, and 100 μg/ml sonicated and denatured salmon sperm DNA. After hybridization, blots were washed twice in 1× SSC-0.1% SDS for 10 min at room temperature and three times in 0.5× SSC-0.1% SDS for 15 min at 65°C. Signals were visualized by autoradiography and quantified by densitometry analysis using a Storm 860 Molecular Imager (GE Healthcare) and Quantity One software (Bio-Rad).
A replica filter was hybridized under the same conditions with the 23S rRNA gene sequence for the normalization of the specific signal quantification.
Substrate binding proteins (SBPs) of ABC systems involved in the transport of both sugars and oligopeptides are mainly expressed as membrane-associated proteins in both Bacteria and Archaea (2, 26). The expression of SBPs can be differentially induced in the presence of the specific ligands in the cell culture and during early stationary phase (10, 25). In this work S. solfataricus cells were grown using two different culture media, the peptide-poor YS and the peptide-rich TYS media, in order to differentially induce proteins belonging to the SBP family. Addition of tryptone, which is composed of a complex peptide mixture from pancreatic casein digest (TYS medium), to the yeast extract-sucrose (YS) culture medium led to a 2-fold decrease in the doubling time and an increase of about 20% in the cell density at the stationary phase.
The surface membrane protein fractions were prepared in parallel from both cultures in the late exponential phase of cell growth (OD of 1.2) and were analyzed by SDS-PAGE. As shown in Fig. Fig.1A,1A, surface proteins from YS and TYS cultures revealed similar patterns by SDS-PAGE but significant differences in the relative abundance of some protein bands. Particularly, two protein bands migrating with apparent molecular masses at about 120 kDa and 240 kDa (Fig. (Fig.1A)1A) were considerably more abundant in the TYS culture, suggesting that the expression levels of these proteins might be upregulated in response to the peptide sources added to the culture medium.
SBP components of ABC transporters in S. solfataricus have molecular masses which fall into the range of 60 to 200 kDa (Transporter DB [http://www.membranetransport.org/index.html]). Therefore, SDS-PAGE protein bands within this mass range (Fig. (Fig.1A)1A) were cut and analyzed by nano-HPLC-ESI-MS/MS. A protein annotated as SSO1273 (Table (Table1)1) was identified in both the 240-kDa and 120-kDa bands (Fig. (Fig.1A).1A). A similarity search in the main sequence databases revealed that SSO1273 shows significant identity with several hypothetical ABC-type oligopeptide binding proteins from different archaeal sources (from 25 to 98% of identity) and bacterial microorganisms (15 to 25% of identity), with the extracellular binding protein from Treponema denticola (TDE1072) as the most similar in bacteria. Furthermore, SSO1273 is one of the five archaeal components of the “predicted solute binding proteins” belonging to COG3889 of the Clusters of Orthologous Groups database, which is for the prediction of protein functions encoded in complete genomes. A 3D-PSSM search of the protein fold database shows that the SSO1273 topology most closely matches the periplasmic peptide-binding protein DppA from E. coli (Protein Data Bank [PDB] accession code 1DPE; PSSM E value of 0.0038; 95% certainty). All the other high-scoring matches are bacterial periplasmic binding proteins of ABC transporters, confirming the likelihood that the archaeal and bacterial OppA proteins share the same topologies.
Furthermore, the sso1273-encoded product revealed the characteristic domain organization of the archaeal ABC transporters belonging to the oligopeptide binding protein class (Fig. (Fig.2A).2A). Additional investigation using Signal P (http://www.cbs.dtu.dk/services/SignalP/) and the big-PI Predictor server (http://mendel.imp.ac.at/gpi/gpi_server.html) revealed a putative N-terminal signal peptide, M1-A30, and a potential C-terminal GPI modification site (omega site at S878) required for GPI lipid attachment of proteins to the cytoplasm membrane (Fig. (Fig.2A).2A). Usually, GPI-linked proteins contain a signal peptide at the N terminus and a C-terminal propeptide which is cleaved off and replaced by the GPI anchor. GPI anchoring is a ubiquitous posttranslational modification of surface-bound eukaryotic proteins, but the existence of GPI modification machinery (12) has also been suggested for archaea (16, 19). Further investigations are needed to experimentally determine whether OppASs is a true GPI-anchored protein.
The S. solfataricus genome organization around the sso1273 gene shows the presence of four downstream open reading frames (ORFs) oriented in the same transcription direction coding for two putative transporter permease proteins and two putative oligopeptide transporter ATP-binding proteins. The distance in nucleotides between each ORF ranges from 2 to 17 nucleotides. Therefore, this genomic region could be organized in an operon-like structure (Fig. (Fig.2B).2B). In addition, the analysis of the nucleotide sequences of the 5′ flanking sso1273 gene region showed a canonical archaeal TATA box and a putative TFB-responsive element (BRE) located immediately upstream of the TATA box (Fig. (Fig.2C),2C), which is revealed to be an important determinant in directionality of transcription in archaea (5).
Interestingly, the archaeal ortholog of sso1273, the newly characterized OppAAp from A. pernix (27), is arranged in an identical chromosomal organization (Fig. (Fig.2B)2B) and shows a sequence similarity of about 40% with SSO1273 (see Fig. S1 in the supplemental material). In addition, all the other components of the putative OppAAp ABC operon have comparable sequence lengths (Fig. (Fig.2B)2B) and significant levels of sequence identity (30 to 60%) with the S. solfataricus counterparts. However, the gene ape1583 encoding OppAAp is located 289 bp upstream to ape1582, and it is believed that this phenomenon represents a possible strategy to render the expression of OppA-encoding genes more independent with respect to the other subunits of the transport system.
Two other SBP components of ABC-type transporters were also identified in the TYS culture surface membrane fraction (see Table S1 in the supplemental material). They were the protein encoded by ORF sso2619, annotated as a putative dipeptide binding protein in all the sequence databases inspected, and the sugar transporter trehalose binding protein (TreS) encoded by ORF sso0999 (Fig. (Fig.1A)1A) and already characterized by Elferink et al. (13). Notably, the expression of TreS does not seem to be affected by the nutrient peptide supplementation, in contrast to SSO2619, which appears strongly increased in TYS medium with respect to YS culture (Fig. (Fig.1A1A).
The regulation of the expression of the sso1273 gene was examined in S. solfataricus P2 cells grown in yeast extract with or without tryptone, evaluating the relative RNA abundance in cultures grown to exponential and stationary phase. The Northern blot analysis using most of the sso1273 coding sequence as the probe revealed a single hybridization band showing signal intensity dependence on the presence of the peptide nutrient (Fig. (Fig.3).3). At least 3-fold higher RNA levels could be detected in cells grown in the presence of tryptone both at exponential and stationary phases. Therefore, the significant increase of the SSO1273 protein was dependent on the tryptic peptide digest from caseins as an inducer of the gene expression through a transcriptional control mode.
When the molecular size of the single hybridization band in Northern blotting (Fig. (Fig.3)3) was compared to the molecular weight RNA standards, it did not match the length of the whole-operon transcript (7,027 bases) (Fig. (Fig.3),3), but it was calculated to be about 2,800 bases, that is, a value matching the length (2,718 bases) of the monocistronically transcribed gene. This result clearly demonstrated that the sso1273 gene is autonomously expressed and is regulated by specific and independent regulatory sequences, as mentioned before (Fig. (Fig.2C2C).
However, in the intergenic region between sso1273 and the downstream ddpB-like gene, no TATA/BRE box was recognized, but there was a consensus sequence (5′-AGTGA-3′) located at −10 and −6 nucleotides to the start codon (Fig. (Fig.2B),2B), which could have a similar function as the Shine-Dalgarno sequence found in prokaryotic organisms (34). These findings support the reported hypothesis that the translation initiation mechanism for genes upstream of an operon is different from that of genes inside the operon (34).
Since classical chromatographic procedures failed, protein electro-elution from the 240-kDa band (Fig. (Fig.1A)1A) of an SDS-PAGE preparative gel was used to purify SSO1273. This approach was chosen because the in-gel digestion and nano-HPLC-ESI-MS/MS analysis of the 240-kDa molecular mass band resulted in the detection of a single protein, SSO1273, without other appreciable protein contaminants. The final yield (10%) of the SSO1273 purification strategy was about 0.12 mg from 1.2 mg of the surface membrane protein fraction (from 300 ml of cell culture) while the yield calculated on total protein extract was less than 0.2%. Edman degradation analysis of the purified protein revealed a unique N-terminal sequence, QSASSSPAST. The first amino acid corresponded to the +31 residue in the polypeptide translated from the sso1273 gene, suggesting that SSO1273 precursor is processed for translocation at the A30-Q31 site, cleaving off a 30-amino-acid signal peptide.
SDS-PAGE analysis of the purified SSO1273 revealed the presence of only two bands corresponding to the apparent molecular masses of 240 and 120 kDa (Fig. (Fig.1B)1B) of the same protein. Moreover, gel filtration chromatography of SSO1273 on a Superose 12 column showed that the protein was eluted as a unique peak corresponding to a molecular mass of 640 kDa. These results suggest that the protein exists in a multimeric, mostly hexameric, form under the purification conditions used and only partially dissociates in SDS-PAGE into the dimer/monomer forms, possibly due to the high resistance frequently shown by archaeal proteins to thermal and/or SDS denaturation (14, 35). In fact, a single 120-kDa electrophoretic band was detected by SDS-PAGE when SSO1273 was subjected to extensive denaturing procedures before the analysis (data not shown). This value is in agreement with the molecular mass of the monomer sso1273-encoded product, taking into account the possible contribution of the sugar moiety. Indeed, the presence of oligomeric forms of SBPs in solution has already been described for some bacterial sugar binding proteins (6, 28, 36), showing associative properties which may be of relevance for membrane receptor functions.
Carbohydrate moieties of SSO1273 were characterized by lectin binding assay. Results indicated a strong positive reaction with the lectin GNA (Galanthus nivalis agglutinin), suggesting that SSO1273 is glycosylated by N-glycan(s) and contains mannose at the nonreducing end of a carbohydrate chain. GNA recognizes terminal nonreducing mannose α(1-3)-, α(1-6)-, or α(1-2)-linked to mannose and would essentially reveal only oligomannosides.
The binding activity of SSO1273 was assayed using bradykinin or xenopsin as ligands by two different assay procedures. In the differential assay, saturation analyses (Fig. (Fig.4A)4A) were performed, and Scatchard plots (Fig. (Fig.4B)4B) were used for the calculation of the affinity constants. A Kd (dissociation constant) value of 0.3 μM and a Bmax (maximum specific binding) of 1.8 μM were obtained, suggesting a high bradykinin-SSO1273 interaction. On the other hand, when the same analysis was carried out with xenopsin, no binding was observed (Fig. (Fig.4A).4A). The affinity of SSO1273 displayed toward bradykinin in the submicromolar range was comparable to the levels reported for the cognate archaeal carbohydrate-binding proteins so far characterized (2, 13). A direct binding assay was also performed using the radio-ligand B3H as the peptide substrate, following the complex formation by gel filtration chromatography. Results show that when B3H was gel filtered alone or with BSA (12 pmol) as a control protein, the entire radioactivity was recovered in a peak between 0.8 and 2.2 ml of elution volume. In the presence of the SSO1273 protein (1 pmol), 38% of radioactivity was found in an excluded peak eluted within a volume of 0.2 to 0.8 ml. A 5-fold higher concentration of the binding protein (5 pmol) produced an increase of up to 70% of the bradykinin-SSO1273 complex (see Fig. S2 in the supplemental material).
The same assays were also performed using the trehalose binding protein TreS from S. solfataricus. TreS was not able to bind xenopsin or bradykinin, thus suggesting that there is no overlapping affinity between sugar and peptide binding proteins. On the basis of these results SSO1273 was named OppASs.
A peptide mixture of trypsin-digested caseins was prepared to study the peptide selectivity of OppASs in a ligand-fishing experiment. These substrates model the peptide nutrient source (i.e., tryptone and pancreatic casein digest) used for S. solfataricus cultures in this study. After incubation of OppASs with the tryptic peptide mixture, the OppASs-ligand complexes were isolated by gel filtration chromatography, and the bound peptides were identified by nano-HPLC-ESI-MS/MS analysis.
In the purified OppASs-ligand complexes only 4 out of the 56 peptides experimentally detected in the control sample were identified (see Table S2 in the supplemental material). Interestingly, two of these (LNFLKK and HPHPHLSFMAIPPK) were recognized as low-abundance peptides in the control casein digest sample; therefore, their identification in the isolated OppASs-ligand complexes could be explained by their specific enrichment. Sequence analysis of the OppASs-bound peptides showed that they lack polar and/or charged residues in the first four positions and are rich in leucine and/or proline residues (Fig. (Fig.5A).5A). The same experiment performed in parallel with an archaeal SSO1273 ortholog, the recently characterized OppAAp from A. pernix (27), resulted in a significant difference in terms of the number and the nature of bound peptides, which suggests distinct binding properties of the two OppA proteins. Specifically, a considerably higher number of OppAAp-bound peptides (20 peptides) was detected, with relative intensity distributions similar to those observed in the control sample (see Table S2). These peptides showed an elevated variability in amino acid sequence, implying a broad substrate specificity (Fig. (Fig.5B5B).
Two differently truncated sso1273 coding sequences (sso1273Δ232 and sso1273Δ642) were expressed in E. coli. In order to enhance protein solubility inside the cells, both of the DNA subfragments lacking the N-terminal signal peptide coding sequences were differently truncated in the C-terminal region in correspondence to amino acid residue T799 or P663 of the polypeptide. The gene region sso1273Δ642, coding for the putative binding domain, was selected through multiple sequence alignments of SSO1273 with several bacterial OppA proteins. The two recombinant proteins were detected by SDS-PAGE analysis with apparent molecular masses of 90 kDa (SSO1273Δ232) and 65 kDa (SSO1273Δ642), values which are in good agreement with the theoretical values calculated for the tagged sequences. Although the expression levels of the two recombinant polypeptides were comparable, the 90-kDa truncated form showed high instability, mainly due to progressive precipitation under all the tested conditions (buffer composition, pH, salts, and solutes, such as glycerol, sucrose, Triton X-100, and deoxycholic acid, used as additives) for handling and storage.
On the other hand, the presence of histidine and sodium deoxycholate was able to stabilize in soluble form the SSO1273Δ642 polypeptide and allowed its purification to homogeneity by further size exclusion and hydrophobic chromatography steps. The assay performed with radiolabeled bradykinin revealed that the SSO1273Δ642 was, indeed, the protein region endowed with peptide binding ability (see Fig. S3 in the supplemental material), showing 30% activity toward bradykinin compared to that of the native OppASs. Moreover, substrate specificity assays on trypsin-digested caseins revealed that the recombinant form SSO1273Δ642 was able to bind 3 out of the 4 peptides found associated with the wild-type OppASs (see Table S2).
In this work, we identified a cell membrane-associated S. solfataricus protein, putatively involved in peptide uptake and highly expressed in peptide-rich growth medium. The protein, SSO1273, was characterized in both native and recombinant forms, and its peptide binding activity was investigated.
Indeed, under the experimental growth conditions used, two oligopeptide binding proteins, SSO1273 and SSO2619, were found to be upregulated by tryptone. This observation has already been reported for the genes sso2615 and sso2616 in the cytosolic cell extracts of S. solfataricus using a proteomics approach; these proteins are the putative ATPase components of the SSO2619 ABC transporter system (32). As a general mechanism for solute transport, S. solfataricus seems to preferentially hydrolyze ATP as the energy supplier for the entrance of di- and oligopeptides across its membrane rather than to benefit from the proton driving force generated by the wide intra- or extracellular differential pH (24). The previously described sugar binding protein TreS (13) was also detected from the surface membrane protein fraction analyzed. This result was expected since it is well known that the treS gene is induced by trehalose and that trehalose is present in significant amounts in the commercial preparations of the yeast extract (37) used as a basal nutrient for both types of cultures.
Several binding protein genes in archaea have been annotated as OppA, but in many cases the encoded proteins possess exclusively sugar-binding activity, and the corresponding genes are usually surrounded by ORFs related to sugar-degrading enzymes. Similarly, gene products assigned to the di-/oligopeptide class of transporters in Sulfolobus acidocaldarius and P. furiosus were later found to belong to the sugar-binding protein family (13, 20, 21). With respect to OppASs, no genes involved in sugar metabolism are closely located to the sso1273 or the other cognate genes coding for the ABC transporter, suggesting that the genetic loci for sugar and di-/oligopeptide uptake systems could be independent in the S. solfataricus genome. Therefore, OppASs is the first protein functionally demonstrated in S. solfataricus to be a real oligopeptide-binding protein. Actually, we defined a binding domain module by overexpressing in recombinant form a truncated version of the protein, which was verified to retain significant peptide binding activity and specificity similar to that of the wild-type OppASs. The S. solfataricus binding protein shows high structural similarity to the protein OppAAp, biochemically characterized from the hyperthermophile archaeon A. pernix and identified in the extracellular medium (27). These archaeal OppA proteins could play similar physiological roles since they share not only the overall domain organization and the operon-like structure of the encoding genes but also about 40% sequence similarity. However, the peptide binding assays evidenced significant differences in the characteristics of the bound peptides to the two OppA proteins, possibly due to sequence variations in the corresponding binding domains. These behaviors were also confirmed by the dissociation constant values measured using bradykinin and xenopsin as peptide substrates. Specifically, the Sulfolobus protein showed a strict preference for peptides with no polar or charged residues in the first four positions. This seems to be in contrast to the requirements needed for the nutritional role played by the Opp transport systems (11). In fact, the majority of the OppA-peptide contacts are formed via the backbone C=O and N-H groups, which facilitate sequence-independent binding of peptides (11).
The expression of an additional OppA, SSO2619, showing a low sequence identity with OppASs (about 20%) and belonging to a complete ABC uptake system, calls into question the physiological meaning of the coexistence of at least two oligopeptide transport systems induced by the peptide-rich conditioned medium. Preliminary data indicated that the Kd values of SSO2619 toward both of the peptide substrates bradykinin and xenopsin, showing no similarity in amino acid composition, are in the submicromolar range (0.25 μΜ and 0.35 μΜ for bradykinin and xenopsin, respectively), in agreement with those measured for OppA from A. pernix (27). These results could suggest a broad specificity for SSO2619 as described for OppAAp. Hence, Sulfolobus seems to have evolved complex machinery, requiring the concerted action of more than one di-/oligopeptide binding protein for peptide “fishing” and uptake, which could be involved in nutritional or nonnutritional functions.
We thank Giuseppe Ruggiero for expert assistance with bacterial cultures and Vitale Carratore for expert assistance with sequence analysis.
Published ahead of print on 9 April 2010.
†Supplemental material for this article may be found at http://jb.asm.org/.