Vol. 25, No. 6ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, June 1984, p. 754-7590066-48041841060754-06$02.OO/OCopyright © 1984, American Society for Microbiology

Purification and Characterization of Aminoglycoside-ModifyingEnzymes from Staphylococcus aureus and Staphylococcus

epidermidisKIMIKO UBUKATA,* NAOKO YAMASHITA, AKIRA GOTOH, AND MASATOSHI KONNO

Department of Clinical Pathology, Teikyo University School of Medicine, Itabashi-ku, Tokyo 173, Japan

Received 4 November 1983/Accepted 19 March 1984

Several strains of Staphylococcus aureus and Staphylococcus epidermidis, exhibiting characteristicresistance patterns to aminoglycoside antibiotics, were examined. The aminoglycoside-modifying enzymesfrom these strains were purified by DEAE-Sephadex A-50 chromatography, affinity chromatography, andSephadex G-100 gel filtration. Three enzymes, a 3'-phosphotransferase III (molecular weight, 31,000; pl4.1), a bifunctional enzyme having 6'-acetyltransferase and 2"-phosphotransferase (molecular weight,56,000; p1 4.1) activity, and a 4',4"-adenylytransferase (molecular weight, 34,000; pI 4.7), were isolated fromcrude extracts of the resistant strains. Aminoglycoside-modifying enzymes with identical enzymaticproperties derived from S. aureus and S. epidermidis were also immunologically identical.

There have been a number of reports on the mechanismsof resistance to aminoglycosides (AGs) in clinical isolates ofStaphylococcus aureus and Staphylococcus epidermidis (3,4, 13, 14, 17, 19, 20, 22, 26, 28). However, reports onpurification of the AG-modifying enzymes from staphylo-cocci are few (13, 14, 21). Le Goffic (12) reported plasmid-coded resistance to AGs, including gentamicin (GM), andsuggested that both acetylation of 6'-NH2 and phosphoryla-tion of 2"-OH was caused by a single modifying enzyme.Recently, transfer and homogeneity of GM resistance plas-mids intra- and interspecifically in S. aureus and S. epider-midis were also demonstrated (2, 5, 7, 8, 17, 27).We have previously reported on the epidemiology of

kanamycin (KM) or GM resistance in S. aureus isolatedfrom clinical material in Japan (9). Subsequently, plasmidsmediating these resistances, referred to as pTU512 (10),pTUO53, and pTUO68 (23) were studied by ultracentrifuga-tion and electron microscopy. In addition, the modified sitesof AGs, using the inactivated products of sisomicin (SISO)and amikacin (AMK), were analyzed by nuclear magneticresonance (24).

In this report, we describe the results of further biochemi-cal and immunological examination of AG-modifying en-zymes purified from transductants carrying plasmids mediat-ing AG resistance and from clinical isolates of S. aureus andS. epidermidis exhibiting characteristic patterns of resist-ance to AGs.

MATERIALS AND METHODSBacterial strains. The staphylococci used in this experi-

ment and the MICs of various AGs are shown in Table 1. Astrain of S. aureus, referred to as MS353(pTU512), wasobtained by transduction of a plasmid mediating KM resist-ance from donor strain TK512 to recipient strain MS353.Plasmid-free MS353 was kindly provided by S. Mitsuhashi,Gunma University, Maebashi, Japan. Two other strains of S.aureus, designated MS353(pTUO53) and MS353(pTUO68),were obtained by transduction of plasmids mediating GM

* Corresponding author.

resistance from TK053 or TK068 to MS353, respectively. S.aureus TK729 was a wild-type strain from clinical material.Three clinical isolates of S. epidermidis, TK1265, TK406,and TK1043, which had resistance patterns similar to the S.aureus strains, were also examined. The MICs of AGs weredetermined by the spot method on agar plates containingtwofold dilutions of antibiotics. Overnight broth cultures ofthe strains were used to inoculate these plates with amultipoint replicating apparatus, resulting in a final inoculumof 10 CFU per spot.

Purification of the enzymes. Bacterial strains were grownin 3 liters of tryptic soy broth (Difco Laboratories, Detroit,Mich.) at 37°C with continuous shaking until late logarithmicgrowth. The cells were harvested by centrifugation andwashed twice in PMK buffer [0.1 M phosphate buffer, 60 mMKCl, 10 mM Mg(CH3COO)2 - 4H20, 6 mM ,B-mercaptoeth-anol, 10% glycerol (pH 6.8)]. The bacterial pellet wassuspended in 40 ml ofPMK buffer and continuously sonicat-ed with an ultrasonicator (9 KHz; Kubota 200M) at 4°C for60 min. The disrupted cell suspension was centrifuged for 10min, and the supernatant was collected. Sonication wasrepeated once, and the supernatant was pooled. The result-ing solution (ca. 70 ml) was centrifuged at 100,000 x g for 60min. The supernatant was collected and designated S-100.

This fraction was applied to a column (1.8 by 35 cm)packed with DEAE-Sephadex A-50. Adsorbed protein waseluted with a linear gradient of NaCl (0 to 1 M) in PMKbuffer after washing out with the same buffer. Fractionshaving enzymatic activity were pooled and dialyzed againstPMK buffer to remove the salt. A column (0.9 by 10 cm) waspacked with SISO-Sepharose 4B previously prepared bycoupling SISO to cyanogen bromide-activated Sepharose 4B(25). The sample was applied to the column and eluted with alinear gradient of KCl (0.06 to 1 M) in PMK buffer. Fractionshaving enzymatic activity were pooled and concentrated bya Diaflo apparatus with a PM-10 membrane (Amicon Corp.,Lexington, Mass.). The sample was applied to gel filtrationon a column (1.8 by 40 cm) packed with Sephadex G-100.Fractions having enzymatic activity were pooled and reap-plied to a column (0.9 by 10 cm) packed with DEAE-Sephadex A-50. The enzyme was eluted with a linear gradi-ent of NaCl (0 to 0.6 M) in PMK buffer. Finally, fractions

754

on July 11, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

AG-MODIFYING ENZYMES FROM STAPHYLOCOCCI 755

TABLE 1. MICs of various AGs for S. aureus and S. epidermidis strains

Species and strain MIC (,ug/ml)(plasmid) KM LVDM GM DKB TOB SISO AMK NTL HBK

S. aureusMS353 0.4 0.8 0.1 0.2 0.1 0.1 0.4 0.2 0.1MS353(pTU512) >100 >100 0.2 0.4 0.1 0.2 0.8 0.4 0.4MS353(pTUO53) >100 >100 >100 >100 100 >100 50 50 3.1MS353(pTUO68) >100 3.1 >100 <100 100 >100 12.5 50 3.1TK729 >100 3.1 0.2 3.1 >100 0.2 25 0.2 0.4

S. epidermidisTK1265 >100 >100 0.1 0.2 0.1 0.2 1.6 0.4 0.1TK406 >100 1.6 100 50 >100 100 25 3.1 0.4TK1043 >100 3.1 0.4 3.1 >100 0.2 6.3 0.2 0.2

having enzymatic activity were pooled, concentrated, and necessary to modify 1 ,umol of the appropriate antibiotic in 1stored at -80°C until used. h at 37°C.Assay of enzymatic activity. Enzymatic activity and sub- Assay of protein. The protein content was determined by

strate studies were done by the methods of Haas and the method of Bradford (1), using bovine serum albumin as aDowding (6), using [3H]acetyl coenzyme A (5.6 Ci/mmol), standard, and by UV absorbance at 280 nm.[8- 4C]ATP (1.0 mCilmmol), or [y-32P]ATP (3,000 Ci/mmol) Estimation of molecular weight. The molecular weight of(Radiochemical Centre, Amersham, Englapd) as the donor. the purified enzyme was estimated from the results ofKM, lividomycin (LVDM), GM, dibekacin (DKB), tobramy- Sephadex 0-100 gel filtration and from sodium dodecylcin (TOB), SISO, AMK, netilmicin (NTL), and habekacin sulfate-polyacrylamide gel electrophoresis as described by(HBK) were kindly provided by their respective manufactur- Laemmli and Favre (11). Serum albunmin (molecular weight,ers. A bioassay method with Bacillus subtilis ATCC 6633 as 68,000), ovalbumin (molecular weight, 45,000), ot-chymo-a test organism was usually employed for enzymatic assay trypsinogen A (molecular weight, 25,100), and cytochrome cduring the purification procedure. The antibiotics used to (molecular weight, 12,500) were used as standard proteins.follow the purification of each enzyme were as follows: KM Electrofocusing. The isoelectric point of the enzyme wasas substrate for 3'-phosphotransferase [APH(3')-III], GM for determined by using an electrofocusing apparatus (Atto SJ-the bifunctional enzyme with 6'-acetyltransferase [AAC(6')] 1071). A bed of polyacrylamide gel was used by addingand 2"-phosphotransferase [APH(2")] activity, and TOB for ampholines of pH 3.5 to 9.0. After electrofocusing, the gel4',4"-adenylyltransferase [AAD(4',4")]. The reaction mixture was cut into 5-mm slices and suspended in 0.5 ml of PMKfor phosphorylation and adenylylation consisted of 0.25 ml buffer to elute the enzyne from the gel. Enzymatic activityofPMK buffer, 0.15 ml ofenzyme solution, 0.05 ml of 40mM was measured, and the isoelectric point was estimated fromATP, and 0.05 ml of 0.5 mM antibiotic, in a total volume of the pH value of control slices suspended in water. The0.5 ml. For acetylation, 2 mM acetyl coenzyme A was added enzyme position in the polyacrylamide gel was detected byinstead of ATP. After incubation for 1 h at 37°C, the reaction staining with Coomassie brilliant blue G-250.mixture was stopped by being boiled for 3 min. Preparation of antisera. Rabbit antisera for three enzymes,The degree of purification of the enzymes was expressed APH(3')-III purified from MS353(pTU512), bifunctional en-

in units. One unit was defined as the amount of enzyme zyme with AAC(6') and APH(2") purified from MS353

TABLE 2. Comparison of various AGs as substrates for modifying enzymes from S. aureus and S. epidermidis strains

Species and strain Enzyme Substrate(plasmid) activitya KM LVDM GM DKB TpB SISO AMK NTL HBK

S. aureusMS353(pTU512) ApHb 100 274 0 0 0 0 75 0 0MS353(pTUO53) APHc 84 100 100 80 67 107 53 94 11

AACC 91 0 100 85 55 107 77 90 2MS353 (pTUO68) APHc 64 0 100 79 70 105 39 86 9

AACC 95 0 100 92 46 101 80 85 2TK729 AADd 91 78 7 71 100 1 70 1 1

S. epidermidisTK1265 APHIb 100 313 0 0 0 0 93 0 0TK406 APHC 58 2 100 91 63 86 41 85 14

AACC 335 0 100 123 248 94 75 16 3AADd 75 46 4 37 100 0 50 0 0

TK1043 AADd 99 57 6 68 100 0 80 2 2a APH, Phosphotransferase; AAC, acetyltransferase; AAD, adenylytransferase.b Results are expressed relative to KM (100%/).c Results are expressed relative to GM (100%o).d Results are expressed relative to TOB (100%o).

VOL. 25, 1984

on July 11, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

ANTIMICROB. AGENTS CHEMOTHER.

_I

15 20 25Fraction number

FIG. 1. DEAE-Sephadex A-50 chromatography of the bifunctional enzyme with AAC(6') and APH(2") activity purified from S. aureus

MS353(pTUO68). The fraction volume was 2.5 ml. Enzymatic activity in each fraction was determined by the method of Haas and Dowding(4). The reaction mixture for phosphorylation consisted of 10 ,ul of 1 mM GM, 10 pl (0.5 ,uCi) of [y-32P]ATP, 50 ,ul of 0.1 M PMK buffer, and 30,ul of enzyme solution. For acetylation, 10 ,ul (0.5 p.Ci) of [3H]acetyl coenzyme A was used instead of [y-32P]ATP. The reaction mixture was

incubated at 37°C for 60 min. Thereafter, 20-,ul samples were spotted onto 4-cm2 phosphocellulose paper, which was washed five times withdistilled water and dried. Radioactivity was counted by a liquid scintillation system. Symbols: 0, phosphorylating activity; 0, acetylatingactivity.

(pTUO68), and AAD(4',4") purified from TK729, were pre-

pared by the method of Matsuhashi et al. (15). Each inocula-tion consisted of 100 jig of protein of the purified enzyme

mixed with an equal amount of complete Freund adjuvant(Difco). The inoculation was performed subcutaneously intothe backs of rabbits (2.5 kg, female) four times at 10-dayintervals. Having confirmed the rise in the level of antibod-ies, we bled the rabbits 7 days after the last inoculation.Immunological studies with the antibody and each enzymewere performed by double immunodiffusion assays by themethod of Ouchterlony (18).

RESULTSSubstrate profiles for the AG-modifying enzymes. Table 2

shows the substrate profiles for the AG-modifying enzymesprepared from each strain. In the S-100 fraction from S.aureus MS353(pTU512), only phosphotransferase activitywas found, and this enzyme appeared to be APH(3')-IIIfrom its substrate profile. With regard to S. qureus

MS353(pTUO53) and MS353(pTUO68), the presence of bothAAC(6') and APH(2") activity was confirmed in the enzy-

matic preparations. With the exception of HBK, a newlytested antibiotic which was found to be poorly modified, allthe other antibiotics listed in Table 2 proved to be excellentsubstrates for many of these enzymes. In addition to these

activities, APH(3')-III activity was found in MS353(pTUO53)(24). In the S-100 fraction from S. aureus TK729, an aden-ylyltransferase activity was observed. The most efficientsubstrate for this enzyme was found to be TOB, followed byKM, LVDM, DKB, and AMK; the extent of modification ofthese antibiotics was at least 70% that seen with TOB. Onthe contrary, GM, SISO, NTL, and HBK were poorlymodified. From these results, it was suggested that thisenzyme was AAD(4',4").A phosphotransferase activity was found in the S-100

fraction from S. epidermidis TK1265 as the only modifyingenzyme. Its substrate range indicated that the enzymemodifies the 3'-OH of the target antibiotics. Three enzymaticactivities were detected in the S-100 fraction from S. epider-midis TK406, that is, phosphotransferase, acetyltransferase,and adenylyltransferase activities. Based on the substrateprofiles, the enzymes were further characterized as

AAC(6'), APH(2"), and AAD(4',4"), respectively. The onlymodifying enzyme present in the S-100 fraction from S.epidermidis TK1043 was an adenylyltransferase whose sub-strate profile corresponded to that of AAD(4',4"), identicalto the one that occurred in S. aureus TK729.

Purification of enzyme. The modifying enzymes extractedfrom the strains listed in Table 1 were purified after the finalstep of DEAE-Sephadex A-50 chromatography. The elution

TABLE 3. Purification of bifunctional enzyme with AAC(6') and APH(2") activity from S. aureus MS353(pTUO68)Sp act (U!

Step Vol (ml) l~Protein Activity Recovery mg of FoldStep Vol (ml) ~~~(mg) (U)a (% protein)

S-100 fraction 50.0 268.5 42.7 100 0.16 1.0DEAE-Sephadex A-50 60.5 62.3 13.5 32 0.22 1.4Affinity chromatography 22.0 6.0 7.3 17 1.2 7.6Sephadex G-100 gel filtration 7.5 1.0 3.8 9 3.6 23DEAE-Sephadex A-50 7.5 0.5 3.0 7 6.0 38

a One unit of enzyme is defined as the amount required to catalyze the phosphorylation of 1 p.mol of GM per 1 h.

105Ec Q40co" .Q3

s. 0.210-20.1

< 0

16Tx

12E8 >

. _

40

O

756 UBUKATA ET AL.

on July 11, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

AG-MODIFYING ENZYMES FROM STAPHYLOCOCCI 757

TABLE 4. Purification of AAD(4',4") from S. aureus TK729

Protein Activity Recovery Sp actStep Vol (Ml) (g UaM(U/mg of Foldprotein)

S-100 fraction 80.0 219.1 302.9 100.0 1.4 1.0DEAE-Sephadex A-50 68.0 41.5 157.6 52 3.8 2.8Affinity chromatography 30.0 1.7 120.0 40 69 50Sephadex G-100 gel filtration 17.5 0.6 68.5 23 107 78DEAE-Sephadex A-50 17.5 0.3 58.3 19 201 143

a One unit of enzyme is defined as the amount required to catalyze the adenylylation of 1 p.mol of TOB per 1 h.

pattern of the enzyme from MS353(pTUO68) is shown in Fig.1. AAC (6') and APH(2") activities were observed as a singlepeak when GM was the substrate, and the position corre-sponded to one of the protein peaks.The specific activity and the recovery rate of two en-

zymes, AAC(6') and APH(2"), and AAD(4',4"), through thepurification steps are summarized in Tables 3 and 4. In thepurification of AAC(6') and APH(2") from extracts ofMS353(pTU068), specific activity after DEAE-Sephadexchromatography was 6.0 U/mg. However, the final recoveryrate was low (7% of the initial activity). The specific activityof AAD(4',4") from TK729 increased from 1.4 to 201 U/mgafter DEAE-Sephadex chromatography (Table 4). The re-covery rate was only 19% of the initial activity. The finalrecovery rate of each enzyme from other strains also rangedfrom 15 to 23%.

Determination of molecular weight and isoelectric point.Molecular weight values of the enzymes estimated by the gelfiltration method and sodium dodecyl sulfate-polyacryl-amide gel electrophoresis are shown in Fig. 2 and Table 5,respectively. The values were almost the same by bothprocedures. Enzymes with the same functions derived fromS. aureus and S. epidermidis had identical molecularweights. The molecular weight of APH(3')-III was 31,000and that of AAD(4',4") was 36,000. The bifunctional enzymewith AAC(6') and APH(2") activity had a molecular weightof 54,000.

a

-~~~~ ~ ~ b

m C

A B C D E F G H I J

FIG. 2. Estimation of molecular weights by sodium dodecylsulfate-polyacrylamide gel electrophoresis. Electrophoresis wasperformed at a constant current of 20 mA. The gel was removedfrom the apparatus, stained for 3 h with 0.1% Coomassie brilliantblue R in 50% methanol-10% acetic acid, and diffusion-destainedwith several changes of 5% methanol-10% acetic acid. Bifunctionalenzyme AAC(6') and APH(2") activity (lanes): A, MS353(pTUO68);B, MS353(pTUO53); C, TK406. AAD(4',4") activity (lanes): D,TK406; E, TK729; F, TK1043. APH(3')-III activity (lanes): G,MS353(pTU512); H, MS353(pTUO53); I, TK1265. Lane J containsthe standard proteins (a) human albumin (molecular weight, 68,000),(b) ovalbumin (molecular weight, 45,000), (c) a-chymotrypsinogenA (molecular weight, 25,100).

A summary of the different AG-modifying enzymes fromthe strains tested in this study as well as their molecularweights and pl values are shown in Table 5. The pl ofAPH(3')-III and that of the bifunctional enzyme withAAC(6') and APH(2") activity was 4.1, whereas that ofAAD(4',4") was 4.7.

Immunological studies of AG-modifying enzymes. Immuno-logical similarities between AG-modifying enzymes from S.aureus and S. epidermidis were studied by using antisera forAPH(3')-III, AAC(6') and APH(2"), and AAD(4',4"). Anti-serum obtained against APH(3')-III was placed in the centerwell as shown in Fig. 3A. The precipitin line was observednot only between the center well and that containingAPH(3')-III purified from S. aureus MS353(pTU512), whichhad been used in the preparation of the antiserum, but alsobetween the enzymes of APH(3')-III isolated from S. aureusMS353(pTUO53) and S. epidermidis TK1265. When enzymesisolated from other strains with different activities weretested, no precipitin line was produced. The results of theexperiment in which the center well contained antiserumagainst AAC(6') and APH(2") activities are shown in Fig. 3B,and a similar experiment with antiserum obtained againstAAD(4',4") is presented in Fig. 3C. In each case, theprecipitin line was observed between the antiserum in thecenter well and the enzymes with identical activities whichwere purified from S. aureus and S. epidermidis.From these results, it may be suggested that AG-modify-

ing enzymes in S. aureus and S. epidermidis are identicalwith respect to immunological specificity.

DISCUSSIONStudies on the mechanisms of resistance to AGs in S.

aureus and S.epidermidis have been reported exclusively inthe United States and Europe (3, 4, 13, 14, 17, 19, 20, 22, 26,

FIG. 3. Immunodiffusion patterns of antisera to AG-modifyingenzymes. (A) Antiserum to APH(3')-Ill from MS353(pTU512), (B)antiserum to bifunctional enzyme AAC(6') and APH(2") fromMS353(pTUO68), and (C) antiserum to AAD(4',4") from TK729.Wells: 1, APH(3')-III from MS353(pTU512); 2, APH(3')-III fromMS353(pTUO53); 3, AAC(6') and APH(2") from MS353(pTUO68); 4,AAD(4',4") from TK729; 5, APH(3')-III from TK1265; 6, AAC(6')and APH(2") from MS353(pTUO53); 7, AAC(6') and APH(2") fromTK406; 8, AAD(4',4") from TK1043; 9, AAD(4',4") from TK406.

VOL. 25, 1984

on July 11, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 5. Molecular weight and isoelectric point of AG-modifying enzymes from S. aureus and S. epidermidis strains

Species and strain Mol wt by:(plasmid) SDS-PAGEa Gel mfltration pl Modifying enzyme(s)

S. aureusMS353(pTU512) 31,000 31,000 4.1 APH (3')-III

MS353(pTIJ053) 56,000 54,000 4.1 AAC(6') and APH (2")MS353(pTIJO53) ~31,000 31,000 4.1 APH (3')-I11

MS353(pTUO68) 56,000 54,000 4.1 AAC (6') and APH (2")TK729 34,000 36,000 4.7 AAD (4',4")

S. epidermidisTK1265 31,000 31,000 4.1 APH (3')-III

TK406 56,000 54,000 4.1 AAC (6') and APH (2")34,000 36,000 4.7 AAD (4',4")

TK1043 34,000 36,000 4.7 AAD (4',4")

a SDS-PAGE, Sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

28). In Japan between 1979 and 1980, we reported epidemio-logical studies on clinical isolates of S. aureus resistant tomultiple AGs, including GM (9). When strains were testedfor susceptibility to AGs, none was susceptible to GM andresistant to TOB. At that time, S-100 fractions prepared fromseveral GM-resistant strains might have had AAC and APHactivities. Further analysis of modified products of SISO andAMK by carbon magnetic resonance and phosphorus mag-netic resonance spectra demonstrated that the 6'-NH2 hadbeen acetylated and the 2"-OH had been phosphorylatedsimultaneously (24). These results are in accordance with aprevious report (12) which suggested that a single enzymemight be catalyzing the modification of 6'-NH2 and 2"-OH.The degree of purification of the enzyme in that study wasnot sufficient to prove that a single enzyme was responsiblefor the two functions. Therefore, we reexamined the techni-cal procedures employed in the purification of the enzymeand used affinity chromatography to attain a higher degree ofpurification.

In these studies, transductant strains reported previously(10, 23) and some new wild-type strains of S. aureus and S.epidermidis that were resistant to AGs were used. The latterincluded strains that were susceptible to GM and resistant toTOB and AMK which have begun to appear in Japan. Asdescribed above, purification of the enzyme was almostsufficient when affinity chromatography was applied in com-bination with DEAE-Sephadex A-50 and Sephadex G-100gel filtration. The enzymes thus obtained from each straincan be classified into three types. One is APH(3')-III whosemajor substrates are KM, AMK, and LVDM. Another is abifunctional enzyme with AAC(6') and APH(2') activitymodifying GM, SISO, TOB, AMK, and DKB. The third typeis an AAD(4',4") which modifies KM, TOB, AMK, andDKB. Frequently, more than two types of enzyme weredetected in extracts from the wild-type strains.Our most remarkable finding was that the enzymes with

identical functions purified from S. aureus and S. epidermi-dis also had the same molecular weight, pl, and immunologi-cal identity. Namely, APH(3')-III had a molecular weight of31,000 and apI of 4.1, the bifunctional enzyme with AAC(6')and APH(2') had a molecular weight of 54,000 and a pI of4.1, and AAD(4',4") had a molecular weight of 36,000 and apI of 4.7. Le Goffic (12) reported that the molecular weight ofthe bifunctional enzyme with AAC(6') and APH(2") activityfrom S. aureus Palm was estimated to be 28,000 with a pl of5.7, whereas the AAD(4',4") from ApOl had a molecularweight of 22,000 and a pI of 5.1. On the other hand,

Santanam et al. (21) found that the AAD(4',4") from S.epidermidis FK109 had a molecular weight of 46,770 and a pIof 5.0. Comparison of our results with those of these workersclearly indicates that with regard to the bifunctional enzymehaving AAC(6') and APH(2") activity, we have obtained ahigher molecular weight and also a different pI value.We are now examining the enzymatic identity of other

strains resistant to GM or AMK which have been isolatedfrom other hospitals in different geographic areas. A compar-ative analysis demonstrates that these strains have enzymat-ic properties similar to those present in isolates from ourhospital (unpublished data). Cohen et al. (2) demonstratedthe presence of a common large plasmid mediating resist-ance to penicillin and GM in S. aureus and S. epidermidisand reported that this plasmid was transferred intra- andinterspecifically between strains of S. aureus and S. epider-midis (16). Our two plasmids, i.e., pTUO53 mediating resist-ance to AGs, penicillin, and macrolides and pTUO68 mediat-ing only AG resistance, are similarly large plasmids. Themolecular weight of the former was 3.2 x 107, and that of thelatter was 3.6 x 107 (23).The question remains as to whether these plasmids, isolat-

ed in different countries and mediating GM resistance, aresimilar and whether AG-modifying enzymes are immunolog-ically identical.

LITERATURE CITED1. Bradford, M. M. 1976. A rapid and sensitive method for the

quantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72:248-254.

2. Cohen, M. L., E. S. Wong, and S. Falkow. 1982. Common R-plasmids in Staphylococcus aureus and Staphylococcus epider-midis during a nosocomial Staphylococcus aureus outbreak.Antimicrob. Agents Chemother. 21:210-215.

3. Courvalin, P., and J. Davies. 1977. Plasmid-mediated aminogly-coside phosphotransferase of broad substrate range that phos-phorylates amikacin. Antimicrob. Agents Chemother. 11:619-624.

4. Dowding, J, E. 1977. Mechanisms of gentamicin resistance inStaphylococcus aureus. Antimicrob. Agents Chemother. 11:47-50.

5. Forbes, B. A., and D. R. Schaberg. 1983. Transfer of resistanceplasmids from Staphylococcus epidermidis to Staphylococcusaureus: evidence for conjugative exchange of resistance. J.Bacteriol. 153:627-634.

6. Haas, M., and J. E. Dowding. 1975. Aminoglycoside-modifyingenzymes. Methods Enzymol. 43:611-628.

7. Jaffe, H. W., H. M. Sweeney, C. Nathan, R. A. Weinstein, S. A.Kabins, and S. Cohen. 1980. Identity and interspecific transfer

758 UBUKATA ET AL.

on July 11, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

AG-MODIFYING ENZYMES FROM STAPHYLOCOCCI 759

of gentamicin-resistance plasmids in Staphylococcus aureus andStaphylococcus epidermidis. J. Infect. Dis. 141:738-747.

8. Jaffe, H. W., H. M. Sweeney, R. A. Weinstein, S. A. Kabins, C.Nathan, and S. Cohen. 1982. Structural and phenotypic varietiesof gentamicin resistance plasmids in hospital strains of Staphy-lococcus aureus and coagulase-negative staphylococci. Antimi-crob. Agents Chemother. 21:773-779.

9. Konno, M., K. Ubukata, H. Takahashi, Y. Sasaki, and S.Kawakami. 1982. Gentamicin resistance in Staphylococcus aur-eus isolated in Japan. Part 1. Isolation frequency of Staphylo-coccus aureus resistant to gentamicin and correlation of otherantibiotic susceptibilities with bacteriophage type. Chemothera-py (Tokyo) 30:86-95.

10. Kono, M., M. Sasatsu, K. Ubukata, M. Konno, and R. Fujii.1978. New plasmid(pTU512), mediating resistance to penicillin,erythromycin, and kanamycin, from clinical isolates of Staphy-lococcus aureus. Antimicrob. Agents Chemother. 13:691-694.

11. Laemmli, U. K., and M. Favre. 1973. Maturation of the head ofbacteriophage T4. I. DNA packaging events. J. Mol. Biol.80:575-599.

12. Le Goffic, F. 1977. The resistance of S. aureus to aminoglyco-side antibiotics and pristinamycins in France in 1976-1977. Jpn.J. Antibiot. 30(Suppl.):286-291.

13. Le Goffic, F., A. Martel, M. L. Capmau, B. Baca, P. Goebel, H.Chardon, C. J. Soussy, J. Duval, and D. H. Bouanchaud. 1976.New plasmid-mediated nucleotidylation of aminoglycoside anti-biotics in Staphylococcus aureus. Antimicrob. Agents Che-mother. 10:258-264.

14. Le Goffic, F., A. Martel, N. Moreau, M. L. Capmau, C. J.Soussy, and J. Duval. 1977. 2"-O-Phosphorylation of gentamicincomponents by a Staphylococcus aureus strain carrying aplasmid. Antimicrob. Agents Chemother. 12:26-30.

15. Matsuhashi, Y., T. Sawa, T. Takeuchi, and H. Umezawa. 1976.Immunological studies of aminoglycoside 3'-phosphotransfer-ases. J. Antibiot. 29:1127-1128.

16. McDonnell, R. W., H. M. Sweeney, and S. Cohen. 1983. Conju-gational transfer of gentamicin resistance plasmids intra- andinterspecifically in Staphylococcus aureus and Staphylococcusepidermidis. Antimicrob. Agents Chemother. 23:151-160.

17. McGowan, J. E., P. M. Terry, T. S. R. Huang, C. L. Houk, andJ. Davies. 1979. Nosocomial infections with gentamicin-resis-tant Staphylococcus aureus: plasmid analysis as an epidemio-

logic tool. J. Infect. Dis. 140:864-872.18. Ouchterlony, 0. 1949. Antigen-antibody reaction in gels. Acta

Pathol. Microbiol. Scand. 26:507-515.19. Porthouse. A., D. F. J. Brown, R. G. Smith, and T. Rogers. 1976.

Gentamicin resistance in Staphylococcus aureus. Lancet. i:20-21.

20. Santanam, P., and F. H. Kayser. 1976. Tobramycin adenylyl-transferase: a new aminoglycoside inactivating enzyme fromStaphylococcus epidermidis. J. Infect. Dis. 134:S33-S39.

21. Santanam, P., and F. K. Kayser. 1978. Purification and charac-terization of an aminoglycoside inactivating enzyme fromStaphylococcus epidermidis FK109 that nucleotidylates the 4'-and 4"-hydroxyl groups of the aminoglycoside antibiotics. J.Antibiot. 31:343-351.

22. Scott, D. F., D. 0. Wood, G. H. Brownell, M. J. Carter, andG. K. Best. 1978. Aminoglycoside modification by gentamicin-resistant isolates of Staphylococcus aureus. Antimicrob. AgentsChemother. 13:641-644.

23. Ubukata, K., and M. Konno. 1982. Gentamicin resistance inStaphylococcus aureus isolated in Japan. Part 2. Transductionof drug resistance by phage lysate and analysis of plasmid fromgentamicin-resistant Staphylococcus aureus. Chemotherapy(Tokyo) 30:96-103.

24. Ubukata, K., M. Konno, K. Shirahata, and T. Iida. 1982.Gentamicin resistance in Staphylococcus aureus isolated inJapan. Part 3. Mechanisms of gentamicin resistance in Staphylo-coccus aureus. Chemotherapy (Tokyo) 30:546-553.

25. Umezawa, H., H. Yamamoto, M. Yagisawa, S. Kondo, and T.Takeuchi. 1973. Kanamycin phosphotransferase I: mechanismof cross resistance between kanamycin and lividomycin. J.Antibiot. 26:407-411.

26. Vogel, L., C. Nathan, H. M. Sweeney, S. A. Kabins, and S.Cohen. 1978. Infections due to gentamicin-resistant Staphylo-coccus aureus strain in a nursery for neonatal infants. Antimi-crob. Agents Chemother. 13:466-472.

27. Weinstein, R. A., S. A. Kabins, C. Nathan, H. M. Sweeney,H. W. Jaffe, and S. Cohen. 1982. Gentamicin-resistant staphylo-cocci as hospital flora: epidemiology and resistance plasmids. J.Infect. Dis. 145:374-382.

28. Wood, D. O., M. J. Carter, and G. K. Best. 1977. Plasmid-mediated resistance to gentamicin in Staphylococcus aureus.Antimicrob. Agents Chemother. 12:513-517.

VOL. 25, 1984

on July 11, 2018 by guesthttp://aac.asm

.org/D

ownloaded from