Cloning and expression of hyaluronate lyase genes of Streptococcus intermedius and Streptococcus constellatus subsp. constellatus

Abstract

Hyaluronate lyase (HAase) genes of Streptococcus intermedius and Streptococcus constellatus subsp. constellatus were isolated. In S. constellatus subsp. constellatus, the deduced amino acid sequence of HAase was most similar to that of S. intermedius (68%), whereas the enzyme of S. intermedius was most similar to that of S. pneumoniae (72%). Upstream of the HAase gene on the opposite strands, an open reading frame of a putative glutathione peroxidase started in S. intermedius, and this arrangement was similar to that in S. pneumoniae but unlike that in S. constellatus subsp. constellatus. Cell lysates of Escherichia coli carrying each streptococcal gene showed HAase activity, demonstrating that each cloned gene actually coded for HAase.

Hyaluronate lyase gene, Zymography, Streptococcus intermedius, Streptococcus constellatus subsp. constellatus

1 Introduction

According to a recent recommendation, the anginosus group within the genus Streptococcus is divided into three species, one of which can be further divided into two subspecies: Streptococcus anginosus, Streptococcus intermedius, Streptococcus constellatus subsp. constellatus and Streptococcus constellatus subsp. pharyngis[1]. Although these bacteria are part of the normal flora in the oral cavity, urogenital region and intestinal tract, they frequently cause purulent infections in various body sites [2].

Hyaluronate lyase (HAase) is one of the possible pathogenic factors in S. intermedius and S. constellatus subsp. constellatus. Whiley et al. reported that approximately 90% of isolates in both species produced the enzyme [3], and no other oral streptococci showed such a high frequency of enzymatic activity. HAase may contribute to host tissue damage by degrading the major tissue component polysaccharides. Many pathogenic Gram-positive bacteria are known to be HAase producers, including Staphylococcus aureus[4], Propionibacterium acnes[5], Clostridium perfringens[6], Peptostreptococcus[7] and pathogenic streptococci (Streptococcus pyogenes[8], Streptococcus dysgalactiae[9], Streptococcus agalactiae[10], Streptococcus pneumoniae[11]).

In previous studies [12,13], we suggested that HAase of S. intermedius also degraded chondroitin sulfate A (CS-A) and C (CS-C), while the enzyme of S. constellatus subsp. constellatus could only degrade hyaluronic acid (HA). These observations, along with another experimental result — that an antiserum against crude HAase of S. intermedius weakly inhibited HAase of S. constellatus subsp. constellatus[13]– revealed that the enzymatic properties might be different between the two species.

The aims of this study were the cloning, sequence analysis and expression of HAase genes in S. intermedius and S. constellatus subsp. constellatus, and to compare the deduced amino acid sequences from enzyme to enzyme.

2 Materials and methods

2.1 Bacterial strains and culture conditions

The type strains (S. anginosus ATCC33397, S. intermedius ATCC27335 and S. constellatus subsp. constellatus ATCC27823) were used as the anginosus group strains. S. pneumoniae ATCC BAA-255 (= R6 strain) was also used for positive control in polymerase chain reaction (PCR). All strains were routinely cultured in GAM semi-solid medium without dextrose (Nissui, Tokyo, Japan).

2.2 Chemical and biological reagents

HA (sodium salt, from human umbilical cord), CS-A (sodium salt, from bovine trachea), 1-ethyl-2-[3-(1-ethylnaphtho[1,2-d]thiazolin-2-ylidene)-2-methylpropenyl]-naphtho-[1,2-d]thiazolium bromide (Stains-all) were purchased from Sigma Chemical, St. Louis, MO, USA. Oligonucleotide primers for PCR were synthesized by Sawady Technology or Bex, both of Tokyo, Japan. Taq DNA polymerase reaction mixture for PCR (Premix Taq, Ex Taq version) and restriction enzymes were purchased from TaKaRa Shuzo, Shiga, Japan. Other chemicals were special-grade products from Wako Pure Chemicals, Osaka, Japan, or from Nakalai Tesque, Kyoto, Japan.

2.3 Assay for HAase or CS depolymerase (CSase) activity

The zymographic method to detect HAase or CSase activity was previously described [12]. Briefly, a sample mixed with electrophoresis buffer without 2-mercaptoethanol was applied on a polyacrylamide gel (7.5%) containing 300 µg ml⁻¹ HA or CS-A at a final concentration, and sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS–PAGE) [14] was performed. The gel was washed twice with 2.5% Triton X-100 for 30 min at a time, and then incubated in 100 mM phosphate buffer (pH 6.2) at 37 °C. After the incubation period, the gel was fixed with 25% methanol and stained with Stains-all. The enzymatic activity can be seen as a pale pink band against a dark blue background. To determine the HAase activity quantitatively, the enzyme source was incubated with 1 mg ml⁻¹ HA aqueous solution in 50 mM phosphate buffer (pH 6.2) containing 0.01% bovine serum albumin (BSA) at 37°C. HAase products were then determined as N-acetyl-glucosamine equivalents by the Elson–Morgan method [15]. One unit of HAase activity is defined as the quantity of the enzyme that liberates 1 µmol of unsaturated disaccharide from HA per min at 37°C.

2.4 Screening of HAase gene

Each bacterial DNA was extracted from cells cultured overnight in GAM broth using a kit (Puregene DNA isolation kit for Gram-positive bacteria and yeast, Gentra Systems, Minneapolis, MN, USA). The DNA concentration of each sample was determined by the absorption value at 260 nm.

According to the HAase gene sequences of S. pneumoniae and S. agalactiae[16], two primer pairs, named HYL and HYL2, were designed for the PCR experiment: HYL-U: 5′-TGGTGGGATTATGAAATCGG-3′, HYL-D: 5′-TGATCGAGCGTCCACGACTC-3′, and HYL2-U: 5′-GGTGGAAACTTAGTTGATATGGG, HYL2-D: 5′-CATCACTCGTATACCAACCACGT-3′. These primer pairs amplify a 523-bp fragment (positions 960–1463) and a 736-bp fragment (positions 1128–1863) of the S. pneumoniae HAase gene (GenBank accession number: L20670), respectively. The reaction mixture (20 µl) containing 10 µl of Premix Taq, 4 pmol each of upper primer and lower primer and 50 ng of DNA template in a 0.5 ml tube (Thermo-tube, ABgene House, Surrey, UK) was set into a thermal cycler (Robocycler Gradient 40 temperature cycler with hot top, Stratagene, La Jolla, CA, USA). The PCR reaction was performed under the following conditions: 94°C for 4 min, 30 cycles of 94°C for 1 min, 56°C for 1 min, 72°C for 1 min, and then a final step at 72°C for 5 min. The product was checked by 1% agarose gel electrophoresis in TAE buffer. The gel was stained by ethidium bromide and then visualized on an UV transilluminator (UVP, Upland, CA, USA). A primer pair named HYL-N (HYL-N-U: 5′-ATCGACCACACCAATGTTGC-3′, HYL-N-D: 5′-GAGCGTCCACGACTCATATC-3′), which amplified a 200-bp (positions 1278–1477) fragment within the HYL or HYL2 pair, was also used for nested- or semi-nested PCR. To amplify the inner region, 1 µl of the 10- to 100-fold-diluted first PCR product was used as a template.

For direct sequencing of PCR products, a DNA fragment band was cut out of an agarose gel and recovered using a spin column kit (NucleoSpin Extract kit, Macherey-Nagel, Duren, Germany). DNA sequencing was performed by a dye-terminator method using an automatic sequencer (ABI Prism Model 377, Applied Biosystems, Foster City, CA, USA). For gene walking, the bacterial DNA was digested by each restriction enzyme, leaving a 3′ overhang (Kpn I, Pst I, Bsp 1286I). A walking kit (TOPO Walker kit, Invitrogen, Carlsbad, CA, USA) was used according to the instruction manual.

In the case of S. intermedius, a primer prepared according to the sequence of S. pneumoniae (HYL6-U: 5′-TATTCTATGAAACAGGAACTGG-3′, positions 301–322) was also used for PCR amplification of the upper stream region.

2.5 Expression of HAase in Escherichia coli JM109

According to the sequencing data of each HAase gene, the PCR primer pair that could amplify the full length of the open reading frame (ORF) of the gene was designed (EXP-INT-U: 5′-CATGCAATCAAAAACAAAAAAGATGC-3′ and EXP-INT-D: 5′-GGGGTACCTTATTTCAATCGAAGACTAAAATAACCG-3′ for S. intermedius, EXP-CON-U: 5′-TATGAATTCAGAAA-TTAGAAATC-3′ and EXP-CON-D: 5′-GGGGTACCTTATTTCAGCCGAAGAACAACGTAACC-3′ for S. constellatus subsp. constellatus, respectively). Each downstream primer contained a sequence to add the recognition site of Kpn I at the end of ORF. The PCR reaction was performed in a 100 µl reaction mixture containing 20 mM Tris–HCl (pH 8.8), 10 mM KCl, 10 mM (NH₄)₂SO₄, 2 mM MgSO₄, 0.1% Triton X-100, 0.1 mg ml⁻¹ BSA, 25 U ml⁻¹Pfu DNA polymerase (Promega, Madison, WI, USA), 200 µM dNTPs, 0.2 µM of each primer pair, and 5 µg ml⁻¹ DNA template as a final concentration. The thermal cycling setting was: 94°C for 4 min, 30 cycles of 94°C for 1 min, 50°C (for S. intermedius) or 42°C (for S. constellatus subsp. constellatus) for 1 min, 72°C for 5 min, and then a final step at 72°C for 5 min. The amplified product was checked by agarose gel electrophoresis and collected by spin-column extraction from the gel. After phenol/chloroform extraction and ethanol precipitation, each DNA was cloned onto PinPoint Xa-1 vector (Promega) that was digested by Nru I and Kpn I, and introduced into E. coli DNA. For protein expression, the bacterial cells carrying recombinant plasmid were incubated in Luria–Bertani (LB) broth containing 100 µg ml⁻¹ ampicillin, 2 µM biotin and 100 µM IPTG at final concentration. The enzymatic expression was determined by SDS–PAGE and zymography.

3 Results

3.1 Characterization of HAase genes

Table 1 summarizes the PCR results using HAase primers. The larger PCR product of each species (the HYL-U–HYL2-D fragment for S. intermedius and the HYL fragment for S. constellatus subsp. constellatus) was expected to be a part of the target gene by semi-nested PCR, and used for DNA sequencing. The sequences of fragments containing putative HAase genes were obtained by gene walking (Fig. 1A).

1

PCR result using HAase gene primers

Species	HAase production	PCR amplification			Semi-nested PCR
		HYL	HYL2	HYL-U+HYL2-D	1st PCR	2nd PCR
S. anginosus	−	weak	−	−	HYL; weak	HYL-N; −
S. intermedius	+	−	+	+	HYL-U+HYL2-D; +	HYL2; +
S. constellatus subsp. constellatus	+	weak	−	−	HYL; weak	HYL-N; +
S. pneumoniae	+	+ (523 bp)	+ (736 bp)	+ (904 bp)

Size of amplified region was different from that in S. pneumoniae.

1

A: ORFs of sequenced fragments in S. intermedius and S. constellatus subsp. constellatus. PCR primers used for direct sequencing and restriction sites used for gene walking are shown. Primers designed according to the sequence of S. intermedius or S. constellatus subsp. constellatus gene have names with ‘I’ or ‘CON’. B: ORF order proximate to streptococcal HAase gene. Shadowed columns indicate the ORFs on the complement chains.

Within the fragment of S. intermedius (total 4153 bp), two main ORFs existed: the HAase precursor gene (positions 804–3980), and the putative glutathione peroxidase gene on a complement chain (positions 16–492). Downstream of the stop codon (TAA) of the HAase precursor gene, an incomplete inverted repeat with two mismatches existed (positions 4000–4040) that may serve as a transcriptional terminator. Using the SignalP program, the HAase gene product was predicted to have a signal peptide region at the N-terminus (30 AA) [17]. A possible Gram-positive cell anchor, including an LPXTG motif followed by a transmembrane domain (predicted by TMpred [18]) and positive-charged amino acid residues, was found in the C-terminal region.

The sequenced region of S. constellatus subsp. constellatus (total 4763 bp) also contained two main ORFs. The putative coding region of the HAase precursor gene was from positions 1121 to 4402, and a 32-bp complete inverted repeat (positions 4409–4445), a possible terminator, was observed just downstream of the stop codon. As in the case of S. intermedius, the deduced protein was also predicted to have the signal peptide region at the N-terminus (37 AA) and the cell-anchored domain at the C-terminus. Another partial ORF, incomplete for the 3′ end (from <1 to 633), was on the opposite strand and showed the best BlastP hit with an unknown hypothetical protein of Bacillus halodurans (protein identification number BAB07584).

By the BlastP search, the deduced amino acid sequences of these streptococcal HAases showed significantly high similarities to the HAases produced by the Gram-positive bacteria (Table 2). A Pfam HMM search (http://pfam.wustl.edu/hmmsearch.shtml) within both of the deduced amino acid sequences revealed the typical conserved regions of polysaccharide lyase family 8, super-sandwich domain and C-terminal beta-sandwich domain. Fig. 2 shows the schematic of conserved regions detected by CDART (conserved domain architecture retrieval tool, http://www.ncbi.nih.gov/Structure/cdd) and the alignment of partial deduced amino acid sequences containing enzymatically active residues among streptococcal HAases [19,20]. The sequences of HAases were well conserved around the essential amino acid residues known in S. pneumoniae and S. agalactiae. Table 3 shows the molecular masses, pI values of the proteins deduced from the HAase genes in S. intermedius and S. constellatus subsp. constellatus, with or without signal peptide regions, that were calculated using GeneWorks version 2.45 (IntelliGenetics, Mountain View, CA, USA).

2

Similarity of deduced amino acid sequences in HAase among Gram-positive bacteria

Species of enzyme source (GenBank accession No.)	Length (AA)	Streptococcus constellatus subsp. constellatus (1093 AA)			Streptococcus intermedius (1063 AA)
		Identity (%)	Positives (%)	Expect	Identity (%)	Positives (%)	Expect
Streptococcus intermedius	1063	747/1085 (68%)	871/1085 (78%)	0.0	–	–	–
Streptococcus pneumoniae
TIGR4(NC_003028, AE007344)	1066	692/1085 (63%)	849/1085 (77%)	0.0	779/1068 (72%)	908/1068 (84%)	0.0
R6 (NC_003098, AE008409)	1078	695/1086 (63%)	851/1086 (77%)	0.0	782/1068 (73%)	910/1068 (84%)	0.0
Streptococcus agalactiae (Y15903)	1072	518/1070 (48%)	691/1070 (64%)	0.0	491/991 (49%)	661/991 (66%)	0.0
Streptococcus pyogenes (AF218838)	868	394/815 (48%)	528/815 (64%)	0.0	399/829 (48%)	542/829 (65%)	0.0
Streptococcus suis (AJ308328)	1164	428/1142 (37%)	634/1142 (55%)	0.0	422/1069 (39%)	609/1064 (56%)	0.0
Staphylococcus aureus (AP003136)	804	290/773 (37%)	444/773 (56%)	1×10−129	278/740 (37%)	420/740 (56%)	1×10−125
Propionibacterium, acnes (U15927)	752	161/627 (25%)	271/627 (40%)	2×10−55	155/617 (25%)	278/617 (44%)	2×10−53
Pedobacter heparinus (U27583)	700	158/642 (24%)	265/642 (40%)	1×10−25	146/645 (22%)	257/645 (39%)	1×10−22

Chondroitin AC lyase. All values were calculated by BlastP.

2

A: Schematic of conserved regions in streptococcal HAases detected by CDART. Vertically hatched bar, carbohydrate binding domain (pfam02018); gray bar, polysaccharide lyase family 8, super-sandwich domain (pfam02278); black bar, polysaccharide lyase family 8, C-terminal beta-sandwich domain (pfam02884); cross-hatched, Gram-positive anchor (pfam00746); #, probable Gram-positive anchor domain containing LPXTG motif (not significantly detected by CDART). Broken line-boxed regions including enzymatically active sites are aligned and shown in panel B. B: Partial alignment of streptococcal HAases. S. int, S. intermedius; S. con-con, S. constellatus subsp. constellatus; S. pn, S. pneumoniae; S. agal, S. agalactiae; S. pyo, S. pyogenes. *, active sites of HAase in S. pneumoniae[19]; †, essential amino acid residues of HAase in S. agalactiae[20]. Shadowed amino acid residues were common among all streptococcal HAases.

3

Molecular mass and pI value of HAase deduced from ORF

HAase producer	Enzyme form	Amino acids (AA)	Molecular mass (kDa)	pI value
S. intermedius	precursor	1063	120	9.46
	mature form	1033	117	9.35
S. constellatus subsp. constellatus	precursor	1093	124	6.48
	mature form	1056	120	6.11

Values of mature form were calculated without the signal peptide region.

The DNA sequences obtained in this study were submitted to GenBank as AF385684 for S. intermedius and AF385683 for S. constellatus subsp. constellatus.

3.2 Cloning of HAase genes and expression in E. coli JM109

By a zymographic method, several bands of HAase activity were detected in the cell lysate of E. coli JM109 carrying each streptococcal cloned gene (Fig. 3). The sample from E. coli carrying only a vector had no enzymatic activity. When using gels containing CS-A for zymography, the culture supernatant of S. intermedius and the cell lysate of JM109 carrying the S. intermedius gene showed the degradation activity. On the other hand, although the cell lysate of JM109 carrying the S. constellatus subsp. constellatus gene had activity to degrade HA, it did not show CS-A depolymerization activity. This enzymatic property was the same as that in the culture supernatant of S. constellatus subsp. constellatus.

3

A: Expression of HAase in E. coli JM109. Cell lysates of E. coli JM109 carrying plasmids were applied on 7.5% acrylamide gel (N) or 7.5% acrylamide containing 300 µg ml⁻¹ hyaluronate (HA), and then SDS–PAGE was performed. Protein bands on normal gel were detected by silver staining. HAase activity on HA-containing gel was detected by Stains-all staining after the incubation period. (−), JM109 carrying PinPoint vector (negative control); int hyl, carrying vector with S. intermedius HAase gene insertion; con-con hyl, carrying vector with S. constellatus subsp. constellatus HAase gene insertion. B: CS-A depolymerization activity of HAases. HAase activities of each streptococcal culture supernatant (Str) and the cell lysate of recombinant E. coli with each HAase gene (Rec) were determined by the Elson–Morgan assay. Then each sample, 300 µunits per lane as HAase, was applied to gel containing HA or CS-A and the depolymerization of the substrate was detected using the zymographic technique. Incubation period at 37°C was 3 h. int, S. intermedius; con-con, S. constellatus subsp. constellatus. HAase of S. intermedius, both natural type and recombinant type, showed CS-A depolymerization activity.

4 Discussion

According to the deduced amino acid sequences of HAase in S. intermedius and S. constellatus subsp. constellatus, both streptococcal HAases could be anchored on the cell surfaces. Some bacterial proteins that mediate the interaction between the host components and bacteria, such as the fibronectin-binding proteins, collagen-binding proteins, and some enzymes that cleave large nutrients into transportable sizes (for example, dextranase and casein peptidase), have been shown to be cell-surface proteins [21]. As seen in Fig. 2A, the streptococcal HAases, except for S. agalactiae, commonly had the LPXTG motif at the C-terminus. In the case of the HAase in S. constellatus subsp. constellatus, the deduced amino acid sequence of HAase also contained the SGXG motif, possibly similar to a glycosaminoglycan attachment site of core protein as revealed by ProfileScan (http://hits.isb-sib.ch/cgi-bin/PFSCAN). Therefore, the cell-surface-anchored HAase might also act as a ligand for bacterial binding to proteoglycans by means of affinity and/or charge, if the enzyme did not digest the substrate rapidly. On the other hand, in both S. intermedius and S. constellatus subsp. constellatus, the HAase activities of culture supernatants were higher than those of bacterial cells (data not shown), thus suggesting that most of the produced enzymes were released from bacterial cells. If the free HAase was the dominant form in vivo as well as in vitro, the enzyme would have a chance to damage the tissue polysaccharide in a broader area compared to that with the cell-fixed form.

Based on the 16S rRNA gene sequence, the species among the anginosus group are close to each other and relatively far from S. pneumoniae, which is a member of the mitis group. However, the nucleotide sequences of HAase-coding regions were more similar between S. intermedius and S. pneumoniae (72%) than between S. intermedius and S. constellatus subsp. constellatus (69%). These similarity values were almost the same as those of the deduced amino acid sequences (Table 2). In addition, S. intermedius and S. pneumoniae showed the same order of the ORFs (glutathione peroxidase and HAase), which was not the case for S. constellatus subsp. constellatus (Fig. 1B; sequences of HAase and proximal genes [16,22–24]). Therefore, the similarity of individual genes among streptococci might not always be the same as the taxonomic relationship, perhaps because of the high frequency of interspecies recombination within the genus Streptococcus.

As shown in Table 3, the expected pI value of HAase of S. intermedius was extremely high. Shain et al. purified HAase from a clinical isolate of S. intermedius UNS 35, and reported the pI value as over 9.3 [25]. Although their purified enzyme had a smaller molecular size (approximately 80 kDa) than as observed in the zymographic experiments (approximately 115 kDa in both the type strain and some clinical isolates [12]), and was also smaller than the predicted molecular mass of the HAase gene of the type strain (117 kDa; without the signal peptide region), the pI value and enzymatic substrate range were similar. Therefore, at least, the HAase gene and the gene product of S. intermedius might be partially conserved within the species, and this high pI value might contribute to the binding of this positively charged enzyme to the negatively charged substrates.

When these putative HAase genes were each expressed in transformed E. coli JM109, both of the cell lysates of E. coli with cloned genes degraded HA, and only the sample from the cells with S. intermedius gene depolymerized CS-A. This indicated that each cloned gene actually coded the HAase and that the substrate range was also the same as that of the natural streptococcal enzyme. However, the molecular sizes of enzymatically active bands detected by zymography were smaller than those of streptococcal HAases and those predicted from the genes. This might be caused by proteinases of E. coli and/or truncated transcriptions of the inserted genes. To confirm this hypothesis, N-terminal amino acid sequencing and determination of transcriptional start points of streptococcal and recombinant HAases are needed.

In a previous study, both the pus from a patient monoinfected by S. intermedius and a sample from an experimental abscess of S. intermedius-inoculated mouse showed HAase activity with lower molecular size than that of the original S. intermedius enzyme, and all activities were equally inhibited by an antiserum against the crude S. intermedius HAase [13]. Therefore, by enzymatic digestion, the HAases with various molecular sizes might be generated from the whole enzyme, and some of them might play a role in tissue polysaccharide degradation even in vivo. Furthermore, the role of HAase in the pathogenicity of S. intermedius and S. constellatus subsp. constellatus should be investigated both in vivo and in vitro.

1

All nucleotide sequence information in this paper was submitted to GenBank, and the accession numbers were obtained (AF385683, AF385684).

Acknowledgements

The author would like to acknowledge Professor N. Maeda for her advice and critical comments.