INFECTION AND IMMUNITY, Jan. 1993, p. 109-116 Vol. 61, No. 10019-9567/93/010109-08$02.00/0Copyright © 1993, American Society for Microbiology

Complete Nucleotide Sequence of the Actinomyces viscosusT14V Sialidase Gene: Presence of a Conserved Repeating

Sequence among Strains ofActinomyces spp.MARIA K. YEUNG

Department of Pediatric Dentistry, The University of Te-xas Health Science Center at San Antonio,7703 Floyd Curl Dnive, San Antonio, Te;xas 78284

Received 12 August 1992/Accepted 28 October 1992

The nucleotide sequence of the Actinomyces viscosus T14V sialidase gene (nanH) and flanking regions wasdetermined. An open reading frame of 2,703 nucleotides that encodes a predominately hydrophobic protein of901 amino acids (Mr, 92,871) was identified. The amino acid sequence at the amino terminus of the predictedprotein exhibited properties characteristic of a typical leader peptide. Five 12-amino-acid units that sharedbetween 33 and 67% sequence identity were noted within the central domain of the protein. Each unitcontained the sequence Ser-X-Asp-X-Gly-X-Thr-Trp, which is conserved among other bacterial and Trypano-soma sp. sialidases. Thus, the A. viscosus T14V nanH gene and the other prokaryotic and eukaryotic sialidasegenes evolved from a common ancestor. Southern hybridization analyses under conditions of high stringencyrevealed the existence of DNA sequences homologous to A. viscosus T14V nanH in the genomes of 18 strains offive Actinomyces species that expressed various levels of sialidase activity. The data demonstrate that thesialidase genes from divergent groups ofActinomyces spp. are highly conserved.

Members of the genus Actinomyces are predominantcolonizers of the human oral cavity (7, 14). Adherence ofActinomyces spp. in vitro to human buccal epithelial cells (5)and other mammalian cell types (9, 41) is facilitated by theenzymatic activity of sialidase, which exposes galactose- or

N-acetylgalactosamine-containing moieties that are the re-

ceptors for the surface fimbrial adhesins from these organ-isms (5, 7). A similar bacterium-host cell interaction wouldbe expected to occur in vivo, since sialidase activity ispresent in saliva and plaque fluids (32, 33). Several indige-nous plaque bacteria known to elaborate sialidase (2, 9, 16,27, 45) are the sources of the enzyme activity detected inplaque fluids. Results from previous studies demonstratedthat sialidase is produced by a large number ofActinomycesspp. (9, 27), although it is not known whether this group ofbacteria constitutes a major contributor to the levels ofsialidase activity in the plaque environment. However, it isnoteworthy that members of the indigenous Actinomycesspp. have evolved with multiple cell surface components,including fimbriae (7) and sialidase (9, 27), that function inconcert to enhance colonization of these organisms in theoral cavity. Moreover, sialidase activity in plaque fluidsinhibits adherence of other plaque bacteria, such as certainstrains of Streptococcus spp. (12, 26), that possess receptorsspecific for sialic acid-containing mucin-like glycoproteinson host tissues.

Sialidase from a number of microorganisms has beencharacterized, and more recently, the genes that encodeseveral of these proteins were cloned and their nucleotidesequences were determined. These include the sialidasegenes from strains of Clostridium spp. (35-37), Vibrio spp.(17), and others (3, 6, 20, 39). Comparison of the amino acidsequences predicted from the sialidase structural genes ofvarious bacteria revealed the presence of a short amino acidsequence referred to as the "Asp box" sequence (34).Significant global homology among the bacterial sialidaseswas not detected. However, amino acid sequence homolo-gies of between 28 and 78% were observed among the

109

sialidases from Clostridium perfringens A99, C. sordeliiG12, and C. septicum NC 0054714 (37). Whereas the pre-dicted primary protein sequences of these bacterial siali-dases have been determined, information concerning theenzyme-active sites of these proteins is not known.The Actinomyces viscosus T14V sialidase gene (nanH)

was identified as a first step toward elucidating the biochem-ical characteristics of the enzyme (53). Results from theprevious study showed that A. viscosus T14V nanH ismaintained stably in Escherichia coli. Moreover, high levelsof enzyme activity were obtained from the sialidase-positiverecombinant E. coli strain. The focus of the present studywas to determine the nucleotide sequence of A. viscosusT14V nanH and the upstream DNA region that containsregulatory sequences for the control of gene expression. Thepredicted amino acid sequence ofA. viscosus T14V sialidasewas compared with those of other bacterial (3, 6, 17, 20,35-37, 39), viral (1, 44), and Trypanosoma sp. (24, 31)sialidases. The sialidase gene from A. viscosus T14V alsowas used as a DNA probe to detect the presence of relatedsequences and the degree of sequence homology in 18 strainsof five Actinomyces species (23) that express various levelsof sialidase activity. Significant conservation ofA. viscosusT14V nanH was observed among members of the genusActinomyces.

MATERIALS AND METHODS

Bacterial strains. The bacterial strains and plasmids usedin this study are listed in Table 1. The media and growthconditions used forActinomyces sp. and E. coli strains weredescribed previously (8, 53).

Subcloning and nucleotide base sequencing. PlasmidspMY450-1 and pMY450-5, which contain the A. viscosusT14V sialidase gene on a 4.3-kbp EcoRI DNA fragmentinserted in opposite orientations in pSK+ (53), were purifiedby ethidium bromide-CsCl equilibrium gradient centrifuga-tion as described previously (40). A series of nested dele-

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

110 YEUNG

TABLE 1. Bacterial strains and plasmids

Sialidase activity'Strain or plasmid Cell associatedb or in Host origin; remark; reference(s)

sonicated extract' Extraceliular

A. viscosusT14V (1+ 2+)MG-1WVU627ATCC 19246A828T6-1600R-28M3902M4301

A. naeslundiiWVU45N16M4356M30

A. bovisW827ATCC 19012

A. israeliiW855ATCC 27037

A. odontolyticusATCC 17982X363

E. coliMV1184MY450-1MY450-1/11MY450-1/13MY450-1/19MY450-1/25MY450-1/23MY450-1/29

35371828942

2234

6382114

3342

23

3928

<0.0242485760<0.0251<0.02

Human; 8Human; 9Human; 50Human; 50Hamster; 8, 23Hamster; 8Rat; 8Monkey; 50Monkey; 50

Human; 8, 23Human; 23Monkey; 50Monkey; 50

Bovine; 50Bovine; 50

Human; 50Human; 50

Human; 8Human; 50

5353This studyThis studyThis studyThis studyThis studyThis study

Apr (Stratagene)Contains A. viscosus T14V nanH on a 4.3-kbEcoRI DNA fragment; 53

Contains the 4.3-kb EcoRI DNA fragment insertedin opposite orientation; 53

a Sialidases from the different preparations were incubated with 200 j±g of a,-acid glycoprotein (substrate) at 37'C for 1 h. Specific activity is expressed as

nanomoles of sialic acid released per hour (i) per 100 p,g of soluble protein or (ii) by 8 x 108 bacterial cells. The data shown are averages of three determinations.Actinomyces sp. bacterial cells were washed with phosphate-buffered saline (pH 7.2) and suspended in the same buffer to a density of 5 x 109/ml (A6w, 2.0).

Washed cells (8 x 108) were incubated with the substrate at 37'C for 1 h. Bacteria were removed at the end of the incubation period, and the amount of sialicacid in the supernatant fluid was determined.

c Soluble protein from sonicated cell lysates of E. coli carrying various plasmids prepared as described previously (53).d Extracellular sialidase in growth medium prepared by ammonium sulfate fractionation (60% saturation) as described previously (53).

tions was obtained from these plasmids by using exonucle-ase III and Si nuclease (GIBCO-BRL Life Technologies,Inc., Gaithersburg, Md.) by the method of Henikoff (19).Additional derivatives of pMY450-1 and pMY450-5 were

obtained by digesting the plasmids with restriction endonu-cleases and ligating the digested fragments to pSK+ linear-ized with the appropriate enzymes. E. coli MV1184 (53) wasused as the host for transformation (40).

Single-stranded DNA was prepared from the culture su-

pematant fluid of each E. coli recombinant strain coinfectedwith helper bacteriophage M13K07 (46) as described previ-

ously (40). Single-stranded DNA templates also were ob-tained by incubation of plasmid DNA (5 ,ug) in a solutioncontaining 0.4 M NaOH and 0.4 mM EDTA (pH 8.0) at 37°Cfor 30 min. DNA sequencing was carried out by the dideoxy-chain termination procedure (42) by using modified T7 DNApolymerase (Sequenase version 2.0; United Stated Biochem-ical Corp., Cleveland, Ohio), [ao-35S]dATP (1,325 Ci/mmol;DuPont NEN Research Products, Boston, Mass.), and theuniversal M13 primer (46). DNA regions with high G+Ccontents were analyzed by substitution of dITP for dGTP.The sequencing reactions were analyzed on 6% polyacryl-

787435569S2

4367

12853421

5154

34

7058

<0.0267707970<0.0269<0.02

PlasmidspSK+pMY450-1

pMY450-5

INFECT. IMMUN.

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

NUCLEOTIDE SEQUENCE OF A. VISCOSUS T14V nanH 111

amide-8 M urea linear and buffer gradient gels (40). Thenucleotide sequence was assembled and analyzed by usingthe software packages (version 7.1) from the WisconsinGenetics Computer Group (13). Sequence homologysearches were performed by using the program FASTA (30)and the GenBank (release 72.0), EMBL (release 30.0), andSwissProt (release 22.0) data base libraries.

Southern analysis. Chromosomal DNA (2 ,ug) was digestedwith BamHI, and the DNA fragments were separated on 1%agarose gels. The procedures for agarose gel electrophore-sis, DNA transfer, and hybridization under conditions ofhigh stringency were as described previously (50). PlasmidpMY450-1 was digested with various restriction endonu-cleases (GIBCO-BRL), and the DNA fragments were elutedfrom agarose by using reagents from the Elu-Quik DNApurification kit (Schleicher & Schuell, Inc., Keene, N.H.).[a*-32P]dCTP (3,000 Ci/mmol; DuPont NEN)-labeled DNAprobes were prepared by nick translation as describedpreviously (53).Primer extension analysis. Total RNA from A. viscosus

T14V harvested at the mid-exponential and early stationaryphases of growth (optical densities at 660 nm, 0.5 and 0.8,respectively) was isolated by CsCl gradient centrifugation as

described previously (50). An 18-base oligonucleotide com-

plementary to the DNA sequence between bases 396 and 413(see Fig. 2) was prepared on a 391 DNA Synthesizer(Applied Biosystems, Inc., Foster City, Calif.). The primer(20 pmol) was labeled with [_y-32P]ATP (3,000 Ci/mmol;Dupont NEN) by incubation with T4 polynucleotide kinase(GIBCO-BRL) at 37°C for 30 min. Total RNA (50 ,ug) wasdenatured by incubation at 95°C for 1 min and precipitatedwith the labeled primer at -20°C for 30 min. The RNA-primer mixture was harvested by centrifugation at 10,000 x

g for 15 min at 4°C and dissolved in a hybridization solution(30 jLl) containing 80% deionized formamide, 40 mM PIPES[piperazine-N,N'-bis(2-ethanesulfonic acid, pH 6.4; SigmaChemical Co., St. Louis, Mo.], 400 mM NaCl, and 100 mMEDTA (pH 7.0). The RNA-primer mixture was heated at90°C for 2 min, transferred immediately to a 45°C water bath,and incubated at this temperature for 80 min. The annealedproduct was precipitated in ethanol and then dissolved in a

buffer containing 100 mM Tris-HCl (pH 8.3), 10 mM MgCl2,10 mM dithiothreitol, 50 mM KCl, 2 mM deoxynucleosidetriphosphates, and actinomycin D (50 jg/ml). RNasin RNaseinhibitor (50 U) (Promega Corp., Madison, Wis.) and avianmyeloblastosis virus reverse transcriptase (10 U) (GIBCO-BRL) were added to direct the extension reaction at 37°C for2 h. Unannealed RNA was digested with DNase-free RNase(1 ug/rml) (GIBCO-BRL) at 37°C for 30 min. A portion of theprimer extension product was analyzed on 6% polyacrylam-ide-8 M urea sequencing gels. Sequencing reactions of the4.3-kb EcoRI DNA fragment using the same 18-base syn-

thetic oligonucleotide primer also were analyzed.Determination of protein and enzyme levels and isoelectric

points. Expression of sialidase activity by Actinomycesstrains or E. coli subclones was monitored by spottingbacterial colonies onto Whatman 3MM filter papers satu-rated with a 150 ,uM concentration of the fluorogenic siali-dase substrate 2'-(4-methylumbelliferyl)-a-D-N-acetyl-neuraminic acid (MU-Neu5Ac, sodium salt; Sigma) (27) in0.17 M sodium acetate (pH 5.4). The filters were incubated at37°C for 15 min and examined under long-wavelength UVlight. Quantitation of enzyme activity in E. coli or A.viscosus protein preparations was carried out by measuringthe amount of sialic acid released from human a1-acidglycoprotein (Sigma). The assay conditions were described

previously (53). Determination of the cell-associated siali-dase of Actinomyces spp. in a washed bacterial cell suspen-sion also was previously described (53).

Preparation of extracellular sialidase by ammonium sul-fate (60% saturation) precipitation of the culture supematantfluids ofA. viscosus T14V or E. coli MY450-1 and sonicatedcell lysates from E. coli MY450-1 was done as describedpreviously (53). A. viscosus T14V cell-associated sialidasewas extracted from whole cells by digestion with lysozyme(500 ,ug/ml) at 37°C for 2 h. The extracted proteins wereseparated from undigested bacterial cells by centrifugation at15,000 x g and brought to 60% saturation with solid ammo-nium sulfate. The precipitated proteins were dissolved in abuffer containing 10 mM Tris-HCl (pH 7.0), 150 mM NaCl, 4mM CaCl2, and 0.2 mM phenylmethylsulfonyl fluoride anddialyzed against 500 volumes of the same buffer at 4°C.Protein concentrations were determined with bicinchoninicacid reagents (Pierce Chemical Co., Rockford, Ill.).

Isoelectric focusing of the various protein fractions wasperformed by the method of Rowley et al. (38). Solubleprotein (300 ,ug) was applied to 4% polyacrylamide nonde-naturing tube gels containing 2% (wt/vol) ampholine (pH 3 to10; Bio-Rad Laboratories, Richmond, Calif.). Proteins wereelectrofocused at a constant voltage of 500 V for 2 h at 4°C.Isoelectric focusing of proteins on similar gels containing0.25% CHAPS {3-[(3-cholamidopropyl)dimethyl-ammonio]-1-propanesulfonate; Sigma}, which removes nonspecificallyadhering hydrophobic proteins, also was performed. Afterelectrophoresis, the gels were incubated in 300 ,uM MU-NeuSAc in 0.17 M sodium acetate (pH 5.4) for 15 min at37°C. Protein bands exhibiting sialidase activity were de-tected by exposure to long-wavelength UV light. The rela-tive isoelectric points (pls) of the fluorescent bands weredetermined from a calibration curve prepared from a set ofpl markers (Bio-Rad) electrofocused under identical condi-tions. The pIs determined for human hemoglobin and cyto-chrome c electrofocused along with the protein sample underexamination were used as internal markers to account forelectrophoretic variations among gels.

Nucleotide sequence accession number. The GenBank ac-cession number for the sequence presented in this paper isL06898.

RESULTS

Expression of sialidase activity by pMY450-1 and its deriv-atives. Several deletion derivatives of pMY450-1 were ob-tained, and their restriction endonuclease maps were deter-mined (Fig. 1). Expression of sialidase activity by eachsubclone was assessed by the rapid filter spot assay (53). E.coli strains carrying pMY450-1/11, pMY450-1/13, pMY450-1/19, or pMY450-1/23 exhibited blue fluorescence with inten-sities comparable to that observed in E. coli MY450-1 afterincubation with the fluorogenic substrate. In contrast, nofluorescence signal was detected with E. coli carryingpMY450-1/25, which contained the 2.3-kb KpnI-EcoRI DNAfragment, or pMY450-1/29, which contained a deletion ofapproximately 400 bp in close proximity to the rightmostEcoRI site of pMY450-1 (Fig. 1). The enzyme activity insonicated extracts of these subclones was determined byusing human a1-acid glycoprotein (Table 1), one of severalnatural substrates for A. viscosus T14V sialidase (53). En-zyme activity was below the level of detection for E. coliMY450-1/25 and MY450-1/29. Thus, the DNA region be-tween the leftmost SstI and rightmost EcoRI sites onpMY450-1 in the orientation depicted in Fig. 1 must contain

VOL. 61, 1993

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

112 YEUNG

E EPK BSm K Se K 8.8

L A 60V InI I I

INFECr. IMMUN.

s E

.LJ0 1 2 3 4Kb

FIG. 1. Physical maps of pMY450-1 an(expression of sialidase activity by each subclthe rapid filter paper spot assay. Plus and npresence and absence, respectively, of a bluSymbols: , pSK+ vector DNA; _, A.M, deleted DNA. Restriction enzymes: I

KpnI; B, BamHI; S, SalI; Sm, SmaI; Ss, Ss,

s8dm wud PW'' the sequence information needed to encode a functional A.viscosus T14V sialidase that specifically binds to and cata-

+ PW4501 lytically hydrolyzes the substrates tested.+ pMY4501/11 Nucleotide and predicted amino acid sequences ofA. visco-+ PMY450.1/13 sus T14V sialidase. The nucleotide sequence of the 4.3-kb

PMY50-1/ 3 EcoRI fragment of pMY450-1 was determined on bothPMY4O1/19 strands (Fig. 2; note that the inserted DNA fragment is in the

- pM450 /23 opposite orientation to that shown in Fig. 1). An open+ ~pM4501/29 reading frame (ORF) of 2,703 nucleotides (between bases- pMY4501/29 376 and 3,078) that was preceded by a putative ribosomal

binding site (Fig. 2, underlined sequence) (43) was identifiedsix bases upstream from the translational initiation codon

d its derivatives. The (ATG). This reading frame encoded a protein of 901 aminolone was monitored by acids with a calculated molecular weight of 92,871. Theninus signs denote the calculated pl of the predicted sialidase was 5.72. Resultsae fluorescence signal. from isoelectric focusing analyses showed the presence of aviscosus T14V DNA; single band (relative pI, 5.45) in the solubilized cell extracts

E, EcoRI; P, PstI; K, and extracellular proteins fromA. viscosus T14V and E. colitI. MY450-1 that stained for sialidase activity after incubation

with the fluorogenic substrate. A similar pI value wasobtained for the band exhibiting enzyme activity whenisoelectric focusing was carried out in polyacrylamide gels

TI 72

TTTGCGACTCGCCCGACTTTGCGAATCGA ATGTTCCGTTACCTTACCGTGAGACGTCTGACATCTGCCGTCTACGATCTCCTGT

TGAAGGCAAGGCGCCTCGGTCCCTAGAACGG&ACACCCATGACATCGCATAGTCCTTTCTCCCGGAGGCGCCTGCCGGCCCTCCTGMetThrSerHisSerProPheSerArgArgArgLeuPro1laLeuLeu

GGCTCCCTGCCACTGGCCGCCACCGGCCTGATGCCCGCCGCACCCCCGGCGCACGCCGTCCCCACGTCTGACGGCCTGGCCGACGTCGlyS-rL.uProLeuAlaAlaThrGlyLt uIu AlAaAaAlaProProAlaHisAlaValProThrSerAspGlyLeuAlaAspval

-________,....

ACCATCACGCAGGTGAACGCGCCCGCGGAGGCCTCTACTCCGTCGGCGATGTCATGACCTTCAACATCACCCTGACCAACACCAGCThrIleThrGlnvalAsnAlaProAlaAspGlyLeuTyrSerValGlyAspValMetThrPheAsnIleThrLeuThrAbnThrSer

GGCGAGGCCCACTCCTACGCCCCGGCCTCGACGAACCTGTCCGGGAACGTCTCCAAGTGCCGGTGGCGCAACGTCCCGGCCGGGACGGlyGruAlaHisSerTyrAlaProAlaS.rThrA.nLauSerGlyAsnValSerLysCysArgTrpArgAsnValProAlaGlyThr

ACCAAGACCGCACTGCACCGGCCTGGCCACGCACACGGTGACCGCCCGAGGACCTCAAGGCCGGTGCGCTTCACCCCCGCAGATCGCCTACThrLysThrAspCy.ThrGlyLeuAlaThrHisTrValThrAlaGluApLeuLysAlaGlyGlyPh.ThrProGlnIleAlaTyr

GAGGTCAAGGCCGTGGACTACCCCCGGAAGGCCCTGAGCACCCCCGAACGCATCAAGG CCCCGACGAGCCCAGTCAACACTCGluValLysAlaValGluTyrAlaGlyLy.AlaLAuSerThrProGluThrIleLy.GlyAlaThrSerProValLysAlaAInSer

CTGCGGGTCGAGTCATCACGCCGTCG G _ CGACACCGTCAGCTACACGGTGCGCGTGCGCTCGLAuArgValGluS.rsIeThrProS-rSerSerG1nGluA.nTyrLysL.uGlyAspThrValSerTyrThrValArgValArgSer

GTGTCGGACAACGATCAACGTCGCCGCCACCGAATCCTCCTTCGACGACCTGGGCCGCCAGTGCCACTGGGGCGGCCTCAAGCCGVal8SrAspLysThrIleA nValAlaAlaThrGluSrSerPheAspA8pLeuGlyArgGlnCysHisTrpGlyGlyL8uLysPro

GGCMAGGCGCCGTCTACAACTGCAAGCCGCTCACCCA CACGATCACGCAAGCCGACGTCGACGCCGGCCGCTGGACGCCATOGATCGlyLy.GlyAlaValTyrAsnCy.LysProL.uThrHisThrIleThrGlnAlaA.pValAspAlaGlyArgTrpThrProSerI1e

ACCCTGACGGCCACCGGAACCGACGGCGCCACCCTCCAOACGCTCACCGCCACCGGCAACCCGATCAACGTCGTCGGCGACCACCCGThrLuThrAlayThGlyrAspGlyAlaThrLuGlnTheuTLThrAlaThrGlyAnProIl AsnValValGlyAspHisPro

CAGGccAcr.c6cACVGGGCCGACG1rA8CGCGGAGCTGCCGGCCTCAATGAGCCAGGCCCAGCACCTGGCCGCCAACACGGCCGlnAlaThrProAlaProAlaProA pAlaS-rThrGluL uProAlaSerMt tSerGlnAlaGlnHisLeuAlaAlaAsnThrAla

ACCGACAACTACCGCATCCCGGCGATCACCACCGCC C CCCCAAGGACAACThrA.pAsnTyrArgIlProAlaIlloThrThrAlaProAsnGlyAspLeuL.uI.8erTyrAspGluArqProLysAspA-nGly

AACGGCGCAG5CGACGCCCCCAACCCGAACCACAT CTACGGTCCGCG_tCCTCA nGlyGlySerA pAlaProAsnProAsnHi*IlVaGrgASrTA plGy-TrpSAlohrSy

ATCCACCAGGCACGGACCGGCAAGAAGGTCGGCTACTCCGACCCGAGCTACGTCGTCGTCGG cIllHi GlnGlyThrGluThrGlyLysLysValGlyTyrSerAspProSerTyrValValAspHisGlnThrGlyThrIllPhisn

TTCCACGTCAAGTCCTACGACCAGGGCTGGGGCGGCTCGICGCGCGGCccGAcccGGAGAACCGGGGC ATCATCC.GGCCGAGGTGPheHisValLySerTyrAspGlnGlyTrpGlyGlySerArglyGTlyThrAspProGluAsnArgGlylI leGlnAlaGluVal

TCGACCTCCACGGACAACGGCTGG.ACCTGGACCGCACGCACGATCACCGAATC CAGAAGCGTGGACCGCGCGTTTC8rThrSerThrAspA nGlyTrpThrTrpThrHi,rgThrIlleTrAlaAspIl ThrLyslupLysProTrpThrAlaArgPhe...~ ... . ... ....

GCGGCCTCGGGCCAGGGCAlaAla8erGlyGlnGly

GCGGTGCAGGCCGTCTCAlaValGlnAlaValSez

AAGGTCGT0GAGCTCTCCLy.ValValGluL.8uer

,TCCAGATTCAGCACGGGCCCCACGCCGGGCGCCTGGTGCAGCAGTACACG6TCA4l-GlnIlnGlnHisGlyProHti.AlaGlyArgLeuValGlnGlnTyrthrIllAm

SGCACGCCGATCSGGACCGGISlyThrProIllGlyThrGl

WCTCCGGCTCCGCTAAGGilySerGlyPheArgLysVa

;GACCGCCGGCGCC 1815rqThrAlaGlyGly

;CATGGATGAGAAC 1902LyM.tA.pGluAsn

GACGGTGGGCAGACCTGGAGCGA CGGTGTCCGACAAGAACCTGCCCGACTCGGTGGACAACGCCCAGATCATCCGAGCCTTCCCGAspGlyGlyGlnThrTrpSerGluroValSerAspLysAsnleuProAspservalAspAsnAlaGlnI1eI.eArgAlaPhePro

AACGCCGCGCCGGACGACCCGCGCGCCAAGGTGCTGCTGCTGAGCCACTCACCGAACCCGCGGCCGTGGTCGCGTGACCGCGGCACCAsnAlaAlaProAspA-pProArgAlaLysVal1LeuLeuLeuSerHi sSerProAsnProArgProTrpSerArgAspArgG lyThr

AT C ATGCTCCTGCGACGACGGCGCCTCCTGGACGAC GCAAGGTCTTCCACGAGCCCTTCGTCGGATACACGACGATCGCGGTG14erM tSerCysA pAspGlyAlaSerTrpThrTh |erLysVal PheHisGluProPheValGlyTyrThrThrIl1eAlaVaI1

411CAGTCCWACGGCAGCATCGGGCTGCTCAGCGAGGACGCCCACAACGGCGCCGACTACGGCGGCATCTGGTACCGCAACTTCACGATGGInSerAspGlySerI.eGlyLeuLeuSerGluAspAlaHisApnGlyAlaAspTyrGlyGlyIleTrpTyrArgAsnPheThrMet

AACTGGCTCGGCGAGCAGTGCGGCCAGAAGCCGGCGGAGCCGAGCCCGGCGCCGTCGCCGACGGCGGCACCCTCAGCGGCACCGACGAsnTrpLeuGlyGluGlnCysGlyGlnLysProAlaGluProSerProAlaProSerProThrAlaAlaProSerAlaAlaProThr

GAGAAGCCGGCCCCGTCGGCCGCGCCGAGCGCTGAGCCCACGCAGGCACCGGCACCATCCTCCGCGCCCGAGCCGAGCGCTGCGCCCGluLysProAlaProSerAlaAlaProSerAlaGluProThrGlnAlaProAlaProSerSerAlaProGluProSerAlaAlaPro

GAGCCGAGCAGCGCCCCGGCGCCGGAGCCCACGACCGCTCCGAGCACGGAGCCCACACCGGCTCCTGCGCCCAGCTCCGCACCTGAGGluProSerSerAlaProAlaProGluProThrThrAlaProSerThrGluProThrProAlaPoaPoAProserSerAlaProGlu

CAGACCGATGGGCCGACCGCTGCGCCCGCACCGGAGACGTCCTCTGCACCGGCCGCCGAACCGACGCAGGCCCCGACGGTGGCGCCTGlnThrA pGlyProThrAlaAlaProAlaProGluThrS*rSerAlaProAlaAlaGluProThrGlnAlaProThrValAlaPro

SJtTCTGTTGAGCCCACGCAGGCTCCGGGTGCGCAGCCGAGCTCAGCACCCAAGCCGGGGGCGACGGGTCGGGCCCCGTCGGTGGTGAACS-rValGluProThrGInAlaProGlyAlaGlnProSerSerAlaProLysProGlyAlaThrGlyArgAlaProS*rValVa lAr.n

CCGAAGGCGACCGGGGCGGCGACGGAGCCTGGGACGCCGTCATCGAGCGCGAGCCCGGCACCGAGCCGGAACGCGGCGCCGACGCCGProLyAlaThrGlyAlaAlaThrluProGlyThrProSerSerSerAlaSerProAlaProSr rArgAsnAlaAlaProThrPro

w TCGGCCGTCTGACGGCACCATGGCGCAGCCGACCGGTGGCGCCAGCGCGCCGAGTGCCLysProGlyNotCluProA pCluIleAspArgProSerAspGlyThrMetAlaGlnProThrGlyGlyAlaS*rAlaProSerAl a

GCGCCGACCAGGCCCCCGAACGCCGGCAGCACGCTGTCTCGCACGGGGACCAACGCGCTGCTGATCCTGGGCCTTGCGGGTGTCGCGAlaProThrGlnAlaAlaLysAlaGlySerArgL uS-rArgThrGlyThrAsnAlaL uLaul-eLeuGlyL uAlaGlyValAla

GTTGTCCGCGCCTA-CCTCGCTGCTGCGGGCTCGCCGTTCGAAGAACTGAACACGCCGACGAGCCGGTCATCCGGCSCTC A,GCACTGACTValValGlyGlyTyrLauLeuLeuArgAlaArgArqSerLysAsn***GAGGGGTGGGACGCCGGCGCGTCCCACCCCTCAGTGCTGTCCTAGGCGGCCGAGCTGACCGCCTCAGCGATCCCCGGCCTTCCTCAG

GCCCCAGACGCAAAAGATCTCCAAATCGGACACAAAGGCCCAGAACGGGGTCAGAAACGGCGCGACGAAACCGCATGATTCCAACG

TTCTGTTTCCCCCCTTGGCAACGCCGGGGCTCCTTTGTGTCGGATTTrGCACACCAACGCCCTCTAGCGCCCCTGCAATGCGGTCAGG

TTCGAGGGCCAGAGTCCGTGAGAGCCTGTTCGTATGCATCGAGCAGACGCCGGGCGTTCTTCAGTGACCGGAAGAGGTGGGCGGTCT

CCTGCCAACTCGCGTACTCGTCGGC CGGCATGAGGACCGCACGGCCGTTGTTCGAGACGACACGACCGCCTTCACTATCAGTAGTGG

TAGCGAACTTGGAGAATCACCAGGTCATCGCCAGCAACGAGGTACACGAGGCGATGCTCGTCAGTAATCCTTCGCGACCACGCGCCC

GAGGCGC'CGTACTTGAGCTGCTCGGGTTTCCCGATACCTCCGAAGGGATCGCGAAGCGCCGCATCAATGAGCGTGTTGACCCGTTTG

AGGATCCTGCGGTCCGCCTTCTGCCAGTGCTTATAGTCCTCCCAAGCATACGGATCCCATACGAGACGCATCAGTCCTCCCGGTCGA

GGTCATGCATCTCCGTCTGGCCCGCACGAGCCCGGTCATACGCATCAAGAAGGCGTCGCGCGTTCTCAGGTGATCGGAACAGGTAGG

CGGTCTCCTGCCAAGCAGCGTACTCGTCGGCGGGCATAAGAACAGCACTGCCCTTACGCGAGACAATCTCAACGGCGTCATGATCGG

CGTTGACTCGCTCGATGAGCGGGTACAACGTCTTACGCGCCTCGCTTGCACTGATAGACATCAGAGCACCTCTTCCCTCAGGCGTAC

K,pnIPsrCAGGATACGGTACCATCCGAGACGATAGACTGCAGGCTCTCTCACTCCCTCACACGGTCCGAGCAGCCACCTACTGCACCTCAACGG

CCTCAGGAGCAGCCACAGGGGCACTTCCTGCGAGGGGCTATCACTGAGGGGGTGCGGATGATGTCGAGCTCGTCGACTACCACTGAC

FIG. 2. Nucleotide sequence of the 4,284-bp EcoRI DNA fragment and the predicted amino acid sequence ofA. viscosus T14V sialidase.The 2,703-bp ORF begins at base 376 and terminates at base 3,078. The two transcriptional start sites (Ti and T2; bent arrows) for the sialidasemRNA and the putative ribosome-binding site (rbs; underlined) are indicated upstream of the translational initiation codon (ATG). Thepresumptive leader peptide at the amino-terminal end of the predicted protein is denoted by broken underlining. The five repeating12-amino-acid units (boxed) and the conserved amino acid residues (dots) within each repeating unit are highlighted.

M

75

162

249

2076

2 163

2250

23 3 7

2424

2511

2598

2685

2772

2859

2946

3033

3120

3207

3294

3381

3468

3555

3642

3729

3816

3903

3990

4077

4164

4251

4 284

336

423

sia

597

684

771

858

945

1032

1119

1206

1293

1380

1467

1554

1641

1728

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

NUCLEOTIDE SEQUENCE OF A. VISCOSUS T14V nanH 113

%Homology 1 2 3 4 5 6 7 8 9 10 11 12 1314 15161718 19

CGC CGC TCC ACC GAC GGC GGC AAG ACC TGG TCG GCG 1458Arg Arg SIR THR ASP GLY GLY LYS TSR TRP SER ALA 361

TCG ACC TCC ACG GAC AAC GGC TGG ACC TGG ACG CAC 167711 Ser Thr SER THR ASP Asn GLY Trp TSR TRP Thr His 434

GTC TAC TCC GAC GAC CAC GGG AAG ACG TGG CAG GCC 1869III Val Tyr BER Asp ASP His GLY LYS TSR TRP Gln ALA 498

GCC CAC TCC ACC GAC GGT GGG CAG ACC TGG AGC GAG 2013IV Ala His SER THR ASP GLY GLY Gln TER TRP SER Glu 546

TCG ATG TCC TGC GAC GAC GGC GCC TCC TGG ACG ACC 2202V Ser Met SER Cys ASP Asp GLY Ala Ser TRP Thr Thr 609

100

50

58

Mfg ~~~~~~Kblow-,1 W 23.0wee 7.' 9.4I

\\ 6.71~~~~go go 4 4.4_- 2.3- 2.0

67

33

CONSENSUS SIR ASP S GLY x TRSP

FIG. 3. Sequence comparison of the five repeating 12-amino-acidunits of A. viscosus T14V sialidase. The nucleotide and predictedamino acid sequences of the regions corresponding to the five (Ithrough V) 12-amino-acid units are aligned with the consensus Aspbox sequence (bottom line). Amino acid residues identical to thoseof the consensus sequence are in bold capital letters. Amino acids inunits II through V identical to those present in the first 12-amino-acid unit (I) are in capital letters. Percent homology between eachrepeating unit and the first 12-amino-acid unit is indicated on theright.

containing CHAPS, indicating the absence of contaminatinghydrophobic proteins that might have interfered with theelectrophoretic mobility of the enzyme (data not shown).The overall G+C content of the ORF was 71%, which was inaccord with the high G+C content (67%) of the A. viscosusT14V chromosome (10). Moreover, a bias for C or G at thethird position of each codon was observed (Fig. 2) and thecodon usage of the sialidase gene was similar to that of theActinomyces sp. fimbrial subunit genes (52).

Sequence analysis by the program of Kyte and Doolittle(25) revealed several prominent hydrophobic regions thatwere distributed in the amino-terminal end and centralregion of the predicted protein. The amino acid sequence atthe amino terminus resembled that of a typical signal peptide(48) (Fig. 2, amino acid residues marked by broken under-lining). Five 12-amino-acid repeating units were localizedwithin the central domain of the predicted protein (Fig. 2 and3) that shared between 33 and 67% sequence homology.Each repeating unit exhibited sequence similarity to theconserved Asp box sequence (34). Of the five repeating unitsshown in Fig. 3, four (blocks I through IV) contained the fiveconserved residues (-Ser-X-Asp-X-Gly-X-Thr-Trp-) at posi-tions identical to those in the consensus Asp box sequence,while one (block V) contained a conserved substitution ofSer at the Thr position. A homology search of the GenBank(4) data base revealed significant sequence identity (32%over 272 amino acids) between the predicted amino acidsequence of the central domain ofA. viscosus T14V sialidaseand the partial protein sequence of Bacteroides fragilisTAL2480 sialidase (accession number M31663). Both se-

quences contained five Asp box sequence units with compa-rable distances between the units in each protein (39). Theresults agree with the previous finding that the Asp boxsequences and the distances between them are highly con-

served among sialidases from phylogenetically unrelatedorganisms (34).

Occurrence of DNA sequences homologous to A. viscosusT14V nanH among divergent groups of Actinomyces spp.

Sialidase activity was detected in all of 18 strains represent-

FIG. 4. Southern hybridization analysis of BamHI-digested Ac-tinomyces sp. chromosomal DNA to the EcoRI-KAnI fragment(nucleotide bases between positions 1 and 3,044, Fig. 1) correspond-ing to >95% of the A. viscosus T14V sialidase-coding region.Chromosomal DNAs from (lanes 1 through 19) A. viscosus T14V,MG-1, WVU627, ATCC 19246, A828, T6-1600, R-28, M3902, andM4301; A. naeslundii WVU45, N16, M4356, and M30; A. bovisW827 and ATCC 19012; A. israelii W855 and ATCC 27037; and A.odontolyticus ATCC 17982 and X363, respectively, were digestedwith BamHI, separated on a 1% agarose gel, and transferred to aGeneScreen filter. The filter was hybridized to 32P-labeled DNAprobes. The sizes of lambda DNA fragments obtained by digestionwith HindlIl are indicated on the right.

ing fiveActinomyces species (Table 1). Each strain gave bluefluorescence after incubation of the washed cell suspensionswith MU-Neu5Ac as monitored by a filter paper spot assay(data not shown). As shown in Table 1, the levels ofcell-associated and extracellular sialidase activities fromeach Actinomyces strain varied considerably. Among thestrains examined, those that produced high levels of thecell-associated enzyme also produced high levels of extra-cellular enzyme activity. It is not known whether thesesialidase-producing Actinomyces strains, like A. viscosusT14V, contained only one copy of the sialidase gene (53) orwhether the structural genes that encode sialidase from thesestrains were homologous to that from A. viscosus T14V.DNA sequences homologous to A. viscosus T14V nanH

were detected in the genomes of each of the 18 sialidase-producing Actinomyces strains under hybridization condi-tions that allowed a 10% nucleotide sequence mismatch (Fig.4). The DNA probe was the EcoRI-KpnI DNA fragment(Fig. 2, between bases 1 and 3,044) that contained the nanHstructural gene and the adjacent regulatory region. Consid-erable heterogeneity was observed in the genomic DNArelated to nanH among the different bacteria. However, thehybridization patterns were notably homogeneous amongthe rodentA. viscosus, monkeyA. viscosus, andA. naeslun-dii, A. bovis, and A. odontolyticus strains. A hybridizationprofile identical to that seen in Fig. 4 was obtained when a1.03-kb BglI-KpnI DNA fragment (Fig. 2, between bases1,282 and 2,318) that encodes the five repeating 12-amino-acid units was the DNA probe (data not shown).

Significant (89%) sequence homology was noted when thepredicted amino acid sequence ofA. viscosus T14V sialidasewas compared with that ofA. viscosus DSM 43798 sialidase(20). The predicted proteins from these two strains sharedthree regions of nearly 100% sequence identity (A, B, and C,corresponding to amino acids 1 to 315, 342 to 670, and 738 to850, respectively) that were interrupted by short regions ofnonsignificant homology (Fig. 5). The common B domaincontained the amino acid sequence (Fig. 5, enclosed by largebrackets) encoded by the Bgll-KpnI fragment used in South-ern analysis. The results suggest that extensive sequence

VOL. 61, 1993

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

114 YEUNG

AA viscosus T14V: MTSHSPFSRRRLPALLGSLPLAATGLIAAAPPAHAVPTSDGLADVTITQVNAPADGLYSV 60,, ,,,I---------lullllllllllllll;llllllllII ill 111111111111111111A. viscosus DSM43798: I. ..........II..AvicosuDS4379: MSHSPSRRRPALGSLP;ATLIA;APPAHAVPTSDGLADVTITQVNAPADGLYSVGDVMTFNITLTNTSGEAHSYAPASTNLSGNVSKCRWRNVPAGTTiKTDCTGLATHTVTAEDLKAG;GFTPQIAYEVKAVEYAGKALGDVMTFNITLTNTSGEAHSYAPASTNLSGNVSKCRWRNVPAGTTKTDCTOLATHTVTAEDLKAGGFTPQIAYEVKAVEYAGKALSTPETIKGATSPVKANSLRVES ITPSSSQENYKLGDTVSYTVRVRSVSDKTINVAATESSFDDIGRQCHWGGLMKPGGAVYNCK

STPETIKGATSPVKANSLRtVESITPSSSQENYKLGDTVSYTVRVRSVSDKTINVAAT111SFDDIGR1CHWGGLKPGKGAVYNCKPLTHTITQADVDAGRWTPSITLTATGTDGATLQTLTATGNPINVVGDHPQATPAPAPDASTELPASMSQAQHLAANTDYRI

P1THTITQADVDAGRWTPSITLTATGTDGATLQTLTATGNPINVVGDHPQATPAPAPDAS,.,11..-QAQHLAANT#NRPAITTAPNGDLLISYDERPKDNGNGGSDAPNPNHIVQ RRTGGTSA PTIHQGTETGKKVGYSDPSYVVDHQTGTIFN--FH

PAIPPPPMGTCSSPTTSARRTTATAAATTPNPNHIV RSDGTSA PTIHQGTETGKKVGYSDPSYVVDHQTGTIFNFHV

KSYDQGWGGSRGGTDPENRGI IQAE STSTDNGTH RTITAITXD PWAFFAASGQGIQIQHGPHAGRLVQQYTIRTAGGIillllllllllllllli11111111111iT11111i11111111111111111111111111111111111111111111STPAETIKGASPVHKTWQSL0GTPIGTGMENXVVLDTSY08VDGSLMLNS SSFDDRCGGJPK AQY

llilllllllllillllllll ill1111111111111111111111111111111111111111111111111111111iIIlIi

AFPNAAPDDPRAKVLLLSHSPNPRPWCRDRGTI SCDDGASWI KVFHEPFVGYTTIAVQSDGST0QLAJTTAHNGADYGG8YW

YRNFTMNWLGEQCGQKPAEPS PAPSPTAAPSAAPTEKPAPSAAPSAEPTQAPAPSSAPEPSAAPEPSSAPAPEPTTAPSTEIPTPYIRlllilllllllGCllKAESPRRllllll:::: RRRPRPRRLSPRRHRHHPPRPSRA:1: 1::::1:::::::HR ::

YRFTMNWLGEQCGQKPAEPLSSPGRRRHPQRRRRPRRALPRRHHPPRP~~VFEPVYTI DSRALRPSRAGPGAGAHDRSEGAH

APAPSSAPEQTDGPTAAPAPETSSAPAAEPTQAPTVAPSVEPTQAPGAQPSSAPKPGATGRAPSVVNPKATGAATEPGTPSSSA: : liii 11111111111 111lllllllllllllllllllliliilllllliiGSCAQSAPEQTDGPTAAPAPETSSAPAAEPTQAPTVAPSVEPTQAPGAQPSSAPKPGATGRAPSVVNPKATGAATEPGTPSSSA

SPAPSRNAAPTPKPGMEPDEIDRPSDGTMAQPTGGASAPSAAPTQMKAGSRLSRTGTNALLILGLAGVAVVGGYLLLRARRSK1111111111111111111111111111111111P:::::::::: :: ::SPAPSRNAAPTPRPGMEPDEIDRPSDGTMAQPTGAPARRVPRRRRRRRPAAGCLARDQRAADPGPCGCRGCRRVPAAAGSPFEE

60

144

144

228

228

312

: 312

396

396

; 480

; 480

t 564

t 564

1 648

1 648

I 732

! 731

K 816

i 15

900

C 899

FIG. 5. Comparison of the predicted amino acid sequence ofsialidase from A. viscosus T14V and DSM 43798. The predictedproteins were aligned by using the FASTA program of Pearson andLipman (30). Three regions (A, B, and C) of nearly 100% sequenceidentity that are separated by short regions with little or no homol-ogy are indicated. The amino acid sequence encoded by the BglI-KpnI DNA fragment used as a DNA probe in Southern hybridiza-tion studies described in the text is bracketed. Identical (verticallines) and conserved substituted (colons) amino acid residues andthe 12-amino-acid repeating units (boxed) are indicated.

homology over the entire sialidase molecule may be ex-pected among strains of the same Actinomyces species.

Determination of the transcriptional initiation site(s) for A.viscosus T14V nanH. The 18-base synthetic oligonucleotide5'-GGCAGGCGCCTCCGGGAG-3', which is complemen-tary to the nucleotide sequence between positions 396 and413 (Fig. 1), was used in primer extension analyses tohybridize to A. viscosus T14V total RNA. Two transcrip-tional initiation sites, designated Ti and T2, were obtainedwith cellular RNA from both the mid-exponential and earlystationary phases of growth (Fig. 6A). The intensities of thetwo bands were comparable in reactions with RNA ex-

tracted from cells in the mid-exponential phase of growth. Incontrast, the signal observed with the second band was lessintense than that obtained with the first band in the reactionusing RNA from cells in the early stationary phase of growth(Fig. 6A, lane 2). Thus, the transcriptional start sites of theA. viscosus T14V nanH mRNA were mapped to thymidineat position -125 (Ti) and cytosine at position -104 (T2),respectively, relative to the translational start codon (Fig. 2).Alignment of the 5' nucleotide base sequences upstream ofeach transcriptional start site revealed significant homologybetween bases -14 and -18 and -25 and -53, respectively,with respect to the positions of Ti and T2 (Fig. 6B).

DISCUSSION

Sequence analysis of A. viscosus T14V nanH and itspredicted protein revealed the presence of unique featuresthat are characteristic of sialidases from other bacteria (34).The calculated molecular weight (92,871) of the predictedprotein and the molecular weight (100,000) determined pre-viously by maxicell analysis of the protein expressed by E.coli carrying A. viscosus T14V nanH (53) were similar. The

B-60 - 5 C - o -30 -20 -10 +1

5' -GTGTGUC0GC0CCC; CGC0GCGG5 GGACC0GAG0CCCCGGGCJTCA90GG CGJRGC0GTC0GCCT 3'iIIIIIII111111111 11 H11I, 11 III

5' ARJGRCCGARGCGGCCGCGGCGT PG0GPCCGRGCGGTCGGCCT-TG-GCGAC-CGCCCGfCTTTGC 3'-60 -50 e3 - 3C -20 -10 t1

FIG. 6. (A) Determination of the transcriptional start site of theA. viscosus T14V nanH mRNA. The product of primer extensionusing total RNA harvested from the mid-exponential (lane 1) andearly stationary (lane 2) phases of growth was analyzed on a 6%polyacrylamide-8 M urea sequencing gel. The sequencing reactions(ACGT) of pMY450-1 using the same synthetic oligonucleotide asthat employed in primer extension analysis also were analyzed. Thepositions of extension products Ti and T2 and the correspondingnucleotides (underlined) on the transcribed DNA strand are indi-cated. (B) Sequence comparison of the 5' regions upstream of thetranscriptional start sites of A. viscosus T14V nanH mRNA. Thenucleotide sequences of the 5' region upstream of Ti (uppersequence) and T2 (lower sequence) were aligned by introduction ofone space (dash). Identical nucleotides (vertical lines) and thetranscriptional initiation nucleotide (underlined) designated as the+1 base of each region are indicated.

calculated pI (5.72) of the predicted protein and the observedpI (5.45) determined for the cell-associated and extracellularsialidases from A. viscosus T14V and E. coli carrying nanHwere in close agreement. The predicted A. viscosus T14Vsialidase contained five 12-amino-acid units (Fig. 2 and 3)that were homologous to the conserved Asp box sequencepresent in sialidases from other bacteria (3, 6, 17, 20, 35-37,39), viruses (1, 41), and Trypanosoma spp. (24, 31). Resultsfrom Southern analysis showed that A. viscosus T14V nanHwas highly conserved among sialidase-producing strains ofActinomyces spp. (Fig. 4). Taken together, the data demon-strate that the Actinomyces sp. sialidase genes are membersof the same subfamily that share a common ancestor with theother prokaryotic and eukaryotic sialidase genes.Each of the 18 strains of Actinomyces spp. examined in

the present study expressed cell-associated and extracellularsialidase activities (Table 1). Quantitation of enzyme activityfrom various subcellular fractions from each strain indicatedthat >80% of the total enzyme was cell bound while less than10% was cell free (51). The mechanism that mediates exportof A. viscosus T14V sialidase across the cell cytoplasmic

A.;

,.\ .:

,^

;t

t EJ A C G T r,.__ ;::.

_ _

_t G_ - AS ..T .; :_ [ t ]__ _ .i_ _

_ ;.1

_ !

_, aZ) > _ c.A

___ .' AA

r

INFECT. IMMUN.

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

NUCLEOTIDE SEQUENCE OF A. V7SCOSUS T14V nanH 115

membrane and its subsequent association with or releasefrom the cell wall is not known. Export of most bacterialproteins destined to locate in the cell envelope or in theextracellular medium is usually facilitated by signal peptides(29). However, protein export by a signal sequence-indepen-dent mechanism also has been demonstrated (11, 15). Se-quence analysis indicated that A. viscosus T14V sialidasemay be synthesized as a precursor with a putative signalsequence (Fig. 2). The first 29 residues at the amino terminushad the following basic signal sequence motif: a hydrophilicN-terminal region (4 basic amino acids) followed by a centralhydrophobic core (18 nonpolar amino acids) and a leadercleavage site (with a run of three alanine residues) at theC-terminal region. The carboxyl alanine residue is consid-ered the most favored amino acid adjacent to the signalsequence cleavage site (47). Moreover, the long hydrophilicregion of 11 residues (between amino acids 1 and 11) at theN terminus is characteristic of signal peptides in gram-positive bacteria (28). Interestingly, sialidase activity wasobtained from the periplasm and growth medium of E. colicarrying A. viscosus T14V nanH (53), suggesting that pro-cessing of the putative signal sequence also occurred in E.coli. Further studies are needed to determine the length ofthe putative leader sequence of the A. viscosus T14V siali-dase preprotein.Whereas the amino acids at the amino terminus of the

predicted A. viscosus T14V sialidase sequence may becrucial to the localization of the mature enzyme, those at thecarboxyl terminus are not required for enzyme activity.Results from deletion analyses indicated that E. coli MY450-1/19, which contained a 2.7-kb EcoRI-SstI DNA fragment,expressed sialidase activity while E. coli MY450-1/25, whichcontained a 2.3-kb EcoRI-KpnI DNA fragment, did not (Fig.1). Since the translational stop codon of the predictedsialidase ORF was located at nucleotide base 3,078 (Fig. 2),pMY450-1/19 would be expected to encode an N-terminaltruncated protein of 786 amino acid residues. Comparablelevels of enzyme activity were obtained for both E. coliMY450-1 and MY450-1/19 (Table 1), which indicated thatdeletion of 117 amino acids from the carboxyl terminus didnot affect the functional properties of the enzyme.The considerable variation of sialidase activity among the

18 different Actinomyces strains is noteworthy (Table 1).Various levels of sialidase activity also were obtained inprevious studies with other strains of Actinomyces spp. (9,28). It is not known whether the concentration of nanHmRNA correlated with the levels of sialidase in these organ-isms. Data from primer extension analysis indicated thatexpression of nanH from A. viscosus T14V is regulated bytwo tandem promoters, both transcribed during the mid-exponential and early stationary phases of growth (Fig. 6A).The presence of dual promoters, and thus enhanced tran-scription, may, in part, account for the high level of enzymeactivity in this strain. Further studies will be required todetermine the strength of each promoter and its response toenvironmental signals.Sequence similarity between the nucleotide bases at the

putative -10 and -35 regions ofA. viscosus T14V nanH andthe canonical E. coli promoter consensus sequence (18) orother promoter sequences (21, 22, 49), including thoserecognized by Streptomyces spp. (22, 49) with genomes withhigh G+C contents, was not observed. The results suggestthat Actinomyces spp. possess a distinct class of promotersequences. Interestingly, examination of the nucleotide se-quence 5' of the ORF revealed the presence of a sequence(269-fTGCAG-16 bp-TACCT1-296) that contains the highly

conserved bases (underlined) of the E. coli promoter se-quence (18). The -35 region of this sequence was in closeproximity to nanH transcriptional start site Ti (Fig. 2). It islikely that this sequence will be preferentially utilized in E.coli for expression of sialidase activity encoded by theinserted DNA. This would account for the stability of A.viscosus T14V nanH in E. coli and for the high level ofenzyme activity obtained from the latter host (53). Primerextension analysis using total RNA from E. coli MY450-1will allow comparison of the transcriptional initiation sitesfor nanH in E. coli and A. viscosus T14V. The presentresults indicate that conformity to the canonical E. colipromoter sequence is insufficient for signal transcription inA. viscosus T14V. Further studies are needed to character-ize the structure of the promoter sequence involved in themodulation of nanH expression in this organism.

ACKNOWLEDGMENTS

I thank M. Gamez for technical assistance in the preparation ofthe nested deletion clones for DNA sequencing and D. J. LeBlanc,L. N. Lee, and S. J. Mattingly for review of the manuscript.

This work was supported by Public Health Service grant DE08932from the National Institute of Dental Research.

REFERENCES1. Air, G. M., R. G. Webster, P. M. Colman, and W. G. Laver.

1987. Distribution of sequence differences in influenza N9neuraminidase of tern and whale viruses and crystallization ofthe whale neuraminidase complexed with antibodies. Virology160:346-354.

2. Beighton, D., and R. A. Whiley. 1990. Sialidase activity of the"Streptococcus millen group" and other viridans group strep-tococci. J. Clin. Microbiol. 28:1431-1433.

3. Berry, A. M., J. C. Paton, E. M. Glare, D. Hansman, andD. E. A. Catcheside. 1988. Cloning and expression of thepneumococcal neuraminidase gene in Escherichia coli. Gene71:299-305.

4. Bilofsky, H. S., and C. Burks. 1988. The GenBank geneticsequence data bank. Nucleic Acids Res. 16:1861-1863.

5. Brennan, M. J., J. 0. Cisar, and A. L. Sandberg. 1986. A160-kilodalton epithelial cell surface glycoprotein recognized byplant lectins that inhibit the adherence ofActinomyces naeslun-dii. Infect. Immun. 52:840-845.

6. Camara, M., T. J. Mitchell, P. W. Andrew, and G. J. Boulnois.1991. Streptococcuspneumoniae produces at least two distinctenzymes with neuraminidase activity: cloning and expression ofa second neuraminidase gene in Escherichia coli. Infect. Im-mun. 49:2856-2858.

7. Cisar, J. 0. 1986. Fimbrial lectins of the oral actinomyces, p.183-196. In D. Mirelman (ed.), Microbial lectins and aggluti-nins: properties and biological activity. John Wiley & Sons,Inc., New York.

8. Cisar, J. O., P. E. Kolenbrander, and F. C. McIntire. 1979.Specificity of coaggregation reactions between human oralstreptococci and strains of Actinomyces viscosus or Actinomy-ces naeslundii. Infect. Immun. 24:742-752.

9. Costello, A. H., J. 0. Cisar, P. E. Kolenbrander, and 0. Gabriel.1979. Neuraminidase-dependent hemagglutination of humanerythrocytes by human strains of Actinomyces viscosus andActinomyces naeslundii. Infect. Immun. 26:563-572.

10. Coykendall, A. L., and A. J. Munzenmaier. 1979. Deoxyribonu-cleic acid hybridization among strains ofActinomyces viscosusandActinomyces naeslundii. Int. J. Syst. Bacteriol. 29:234-240.

11. Delepelaire, P., and C. Wandersman. 1989. Protease secretionby Erwinia chrysanthemi. J. Biol. Chem. 264:9083-9089.

12. Demuth, D. R., E. E. Golub, and D. Malamud. 1990. Strepto-coccal-host interactions. Structural and functional analysis of aStreptococcus sanguis receptor for a human salivary glycopro-tein. J. Biol. Chem. 265:7120-7126.

13. Devereux, J., P. Haeberli, and 0. Smithies. 1984. A comprehen-

VOL. 61, 1993

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

116 YEUNG

sive set of sequence analysis programs for the VAX. NucleicAcids Res. 12:387-395.

14. Ellen, R. P. 1982. Oral colonization by gram-positive bacteriasignificant to periodontal disease, p. 98-111. In R. J. Genco andS. E. Mergenhagen (ed.), Host-parasite interactions in peri-odontal diseases. American Society for Microbiology, Washing-ton, D.C.

15. Felmlee, T., S. Pellett, E.-Y. Lee, and R. A. Welch. 1985.Escherichia coli hemolysin is released extracellularly withoutcleavage of a signal peptide. J. Bacteriol. 163:88-93.

16. Fraser, A. G., and R. Brown. 1981. Neuraminidase productionby Bacteroidaceae. J. Med. Microbiol. 14:63-76.

17. Galen, J. E., J. M. Ketley, A. Fasano, S. H. Richardson, S. S.Wasserman, and J. B. Kaper. 1992. Role of Vbrio choleraeneuraminidase in the function of cholera toxin. Infect. Immun.60:406-415.

18. Harley, C. B., and R. P. Reynolds. 1987. Analysis of E. colipromoter sequences. Nucleic Acids Res. 15:2343-2361.

19. Henikoff, S. 1984. Unidirectional digestion with exonuclease IIIcreates targeted breakpoints for DNA sequencing. Gene 28:351-359.

20. Henningsen, M., P. Roggentin, and R. Schauer. 1991. Cloning,sequencing and expression of the sialidase gene fromActinomy-ces viscosus DSM 43798. Biol Chem. Hoppe-Seyler 372:1065-1072.

21. Hoopes, B. C., and W. R. McClure. 1987. Strategies in regula-tion of transcription initiation, p. 1231-1240. In J. L. Ingraham,K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger(ed.), Eschenchia coli and Salmonella typhimurium: cellularand molecular biology. American Society for Microbiology,Washington, D.C.

22. Hopwood, D. A., M. J. Bibb, K. F. Chater, G. R. Janssen, F.Malpartida, and C. P. Smith. 1986. Regulation of gene expres-sion in antibiotic-producing Streptomyces, p. 251-276. In I. R.Booth and C. F. Higgins (ed.), Regulation of gene expres-sion-25 years on. Symposium of the Society for GeneralMicrobiology. Cambridge University Press, Cambridge.

23. Johnson, J. L., L. V. H. Moore, B. Kaneko, and W. E. C. Moore.1990. Actinomyces georgiae sp. nov., Actinomyces gerencse-riae sp. nov., designation of two genospecies of Actinomycesnaeslundii, and inclusion of A. naeslundii serotypes II and IIIand Actinomyces viscosus serotype II in A. naeslundii genos-pecies 2. Int. J. Syst. Bacteriol. 40:273-286.

24. Kahn, S., T. G. Colbert, J. C. Wallace, N. A. Hoagland, and H.Eisen. 1991. The major 85-kDa surface antigen of the mamma-lian-stage forms of Trypanosoma cruzi is a family of sialidases.Proc. Natl. Acad. Sci. USA 88:4481-4485.

25. Kyte, J., and R. F. Doolittle. 1982. A simple method fordisplaying the hydropathic character of a protein. J. Mol. Biol.157:105-132.

26. LiUemark, W. F., C. G. Bloomquist, L. J. Fenner, P. J.Antonelli, and M. C. Coulter. 1989. Effect of neuraminidase onthe adherence to salivary pellicle of Streptococcus sanguis andStreptococcus mitis. Caries Res. 23:141-145.

27. Moncla, B. J., and P. Braham. 1989. Detection of sialidase(neuraminidase) activity in Actinomyces species by using 2'-(4-methylumbelliferyl)a-D-N-acetylneuraminic acid in a filter pa-per spot test. J. Clin. Microbiol. 27:182-184.

28. Neugebauer, K., R. Sprengel, and H. Schaller. 1981. Penicilli-nase from Bacillus licheniformis: nucleotide sequence of thegene and implications for the biosynthesis of a secretory proteinin a gram-positive bacterium. Nucleic Acids Res. 9:2577-2588.

29. Oliver, D. 1985. Protein secretion in Escherichia coli. Annu.Rev. Microbiol. 39:615-648.

30. Pearson, W. R., and D. J. Lipman. 1988. Improved tools forbiological sequence comparison. Proc. Natl. Acad. Sci. USA85:2444-2448.

31. Pereira, M. E. A., J. S. Mejia, E. Ortega-Barria, D. Matzilevich,and R. P. Prioli. 1991. The Trypanosoma cruzi neuraminidasecontains sequences similar to bacterial neuraminidases, YWTD

repeats of the low density lipoprotein receptor, and type IIImodules of fibronectin. J. Exp. Med. 174:179-191.

32. Perlitsh, M. J., and I. Glickman. 1966. Salivary neuraminidase.I. The presence of neuraminidase in human saliva. J. Periodon-tol. 37:368-373.

33. Rogers, R., E. Newbrun, and A. Tatevossian. 1979. Neuramini-dase activity in human dental plaque fluid. Arch. Oral Biol.24:703-705.

34. Roggentin, P., B. Rothe, J. B. Kaper, J. Galen, L. Lawrisuk,E. R. Vimr, and R. Schauer. 1989. Conserved sequences inbacterial and viral sialidases. Glycoconjugate J. 6:349-353.

35. Roggentin, P., B. Rothe, F. Lottspeich, and R. Schauer. 1988.Cloning and sequencing of a Clostridium perfringens sialidasegene. FEBS Lett. 238:31-34.

36. Rothe, B., P. Roggentin, R. Frank, H. Blocker, and R. Schauer.1989. Cloning, sequencing and expression of a sialidase genefrom Clostridium sordellii G12. J. Gen. Microbiol. 135:3087-3096.

37. Rothe, B., B. Rothe, P. Roggentin, and R. Schauer. 1991. Thesialidase gene from Clostridium septicum: cloning, sequencing,expression in Escherichia coli and identification of conservedsequences in sialidases and other proteins. Mol. Gen. Genet.226:190-197.

38. Rowley, P. T., M. Jacobs, C. Rosecrans, L. R. Weitkamp, andR. A. Doherty. 1972. High resolution analysis of hemoglobins:polyacrylamide isoelectric focusing. Biochem. Med. 6:553-560.

39. Russo, T. A., J. S. Thompson, V. G. Godoy, and M. H. Malamy.1990. Cloning and expression of the Bacteroides fragilisTAL2480 neuraminidase gene, nanH, in Escherichia coli. J.Bacteriol. 172:2594-2600.

40. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecularcloning: a laboratory manual, 2nd ed. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

41. Sandberg, A. L., L. L. Mudrick, J. 0. Cisar, M. J. Brennan,S. E. Mergenhagen, and A. E. Vatter. 1986. Type 2 fimbriallectin-mediated phagocytosis of oralActinomyces spp. by poly-morphonuclear leukocytes. Infect. Immun. 54:472-476.

42. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequenc-ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.USA 74:5463-5467.

43. Shine, J., and L. Dalgarno. 1975. Determinant of cistron speci-ficity in bacterial ribosomes. Nature (London) 254:34-38.

44. Steuler, H., W. Rohde, and C. Scholtissek. 1984. Sequence of theneuraminidase gene of an avian influenza A virus (A/parrot/Ulster/73, H7N1). Virology 135:118-124.

45. Tuyau, J. E., and W. Sims. 1975. Occurrence of haemophili indental plaque and their association with neuraminidase activity.J. Dent. Res. 54:737-739.

46. Vieira, J., and J. Messing. 1988. Production of single-strandedplasmid DNA. Methods Enzymol. 153:3-11.

47. von Heijne, G. 1986. A new method for predicting signalsequence cleavage sites. Nucleic Acids Res. 14:4683-4690.

48. von Heijne, G. 1988. Transcending the impenetrable: how pro-teins come to terms with membranes. Biochim. Biophys. Acta947:307-333.

49. Westpheling, J., M. Ranes, and R. Losick. 1985. RNA poly-merase heterogeneity in Streptomyces coelicolor. Nature (Lon-don) 313:22-27.

50. Yeung, M. K. 1992. Conservation of an Actinomyces viscosusT14V type 1 fimbrial subunit homolog among divergent groupsof Actinomyces spp. Infect. Immun. 60:1047-1054.

51. Yeung, M. K. Unpublished data.52. Yeung, M. K., and J. 0. Cisar. 1990. Sequence homology

between the subunits of two immunologically and functionallydistinct types of fimbriae of Actinomyces spp. J. Bacteriol.172:2462-2468.

53. Yeung, M. K., and S. R. Fernandez. 1991. Isolation of aneuraminidase gene from Actinomyces viscosus T14V. Appl.Environ. Microbiol. 57:3062-3069.

INFECT. IMMUN.

on June 26, 2020 by guesthttp://iai.asm

.org/D

ownloaded from