JouRNAL OF BACTERIOLOGY, Oct. 1973, p. 25-32Copyright 0 1973 American Society for Microbiology

Vol. 116, No. 1Printed in U.S.A.

Isolation and Characterization of TwoProtease-Producing Mutants from

Staphylococcus aureusANN-CHRISTINE RYDEN, MARTIN LINDBERG, AND LENNART PHILIPSON

Department of Microbiology, The Wallenberg Laboratory, Uppsala University, Uppsala, Sweden

Received for publication 18 June 1973

Two mutants with increased protease production were isolated after ni-trosoguanidine treatment of Staphylococcus aureus 8325N. The wild typeproduces low amounts of extracellular proteolytic activity. The enzyme was

inducible and could only be detected if casein or preferably skim milk powderwas used as inducer. The optimal pH, salt concentration, and media for enzymeproduction were determined. The mutants differed from the wild type in severalphenotypic characters. The pattern of extracellular deoxyribonuclease andalkaline phosphatase differed between the mutants and the wild type. Severalcarbohydrates such as lactose, galactose, and mannitol were not utilized by themutants, probably owing to a block in the uptake. Glucose could, however, beutilized by the mutants. Reversion frequency to wild type with regard tocarbohydrate utilization was spontaneously high, and all revertants regained theparental pattern irrespective of the carbohydrate used for selection. The resultssuggest that a single locus may control the excretion of extracellular enzymes andcarbohydrate uptake in S. aureus.

The proteolytic enzymes of Staphylococcusaureus have not been studied in detail com-pared to other extracellular proteins. Extracel-lular proteases have been described, but fewenzymes have been purified. Excretion of pro-teolytic activity has been correlated with pro-duction of other enzymes such as lysozyme andcoagulase (9). The proteolytic activity of severalstrains growing on casein prepared from differ-ent sources has been determined (13). Robinsonet al. (17, 18) isolated two proteolytic compo-nents of which one was at first considered to beidentical to alpha hemolysin. Tirunarayananand Lundblad (20) made a partial purificationof proteases from strain Walker. They alsoemphasized the importance of pH control dur-ing growth to obtain maximal yield of proteo-lytic activity and other enzymes (22). Wester-berg et al. (23)- distinguished three fractionswith proteolytic activity by isoelectric focusingof extracellular material from S. aureus strainV8. A protease from strain V8 was recentlypurified by Drapeau et al. (4), and its molecularweight, amino acid composition, and substratespecificity were determined. This paper de-scribes two mutants with high extracellularprotease production, isolated from strain8325N. The wild type produces very low

25

amounts of a proteolytic enzyme. The increasein proteolytic activity appears to be associatedwith profound phenotypic changes, suggesting asingle locus of S. aureus controlling the excre-tion of extracellular products, as previouslysuggested for phage-resistant (3) and proteinA-deficient mutants (8).

MATERIALS AND METHODSStrains S. aureus 8325N and derivatives were used

throughout this study. The wild type was obtainedfrom M. H. Richmond, Department of Bacteriology,the Medical School, Bristol.

Isolation of protease-producing mutants. Thecells were grown in Trypticase soy broth (TSB) to anabsorbancy at 524 nm (A,,4) of 0.75, centrifuged, andwashed with sterile saline. After resuspension to theoriginal volume in TSB medium, sterile N-methyl-N'-nitro-N'-nitrosoguanidine (NTG) was added to a finalconcentration of 100 pug/ml. The bacteria were in-cubated for 30 min at 37 C, and the cells werecentrifuged, washed, and diluted 50-fold in TSBmedium without NTG. Growth was continued over-night for phenotypic expression.

High protease producers were selected on milk agarplates, containing nutrient agar with 1.5% skim milkpowder. Proteolytic activity was revealed by clearzones around the colonies.

Cells were also mutagenized with ethylmethylsul-fonate (EMS). A 0.8-ml amount ofEMS was added toa 4-ml overnight culture in TSB medium. The mix-

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

RYDtN, LINDBERG, AND PHILIPSON

ture was incubated for 10 min at 37 C. The reactionwas stopped by dilution with 16 ml of sterile saline.The cells were centrifuged, and washed twice withsterile saline, and suspended in 5 ml TSB medium.No phenotypic expression was carried out.

Culture media. Nutrient broth, buffered with 0.05M tris(hydroxymethyl)aminomethane(Tris)-hydro-chloride buffer (pH 7.5) and containing 1.5% sterilemilk powder, was used in all experiments to obtainmaximal protease production. TSB medium alonewas used to follow deoxyribonuclease (DNase) pro-duction. A peptone broth was used to detect alkalinephosphatase. The CHM-medium of Lindberg et al.(14) was used to detect transport of "4C-labeledcarbohydrate.

Cultivation techniques. Cultures were inoculatedwith bacteria from fresh agar plates. Cultures were

grown in 100-ml Erlenmeyer flasks or in 1-liter flaskson a rotary shaker at 37 C.

Cell suspension densities. Absorbancies of bacte-rial suspensions were measured at 524 nm. An A524 of0.2 corresponds to 101 cells per ml in TSB.Enzyme preparations. The cells were centrifuged

at 16,000 x g for 15 min in a Sorvall centrifuge. Theextracellular supernatant fluid was used to measure

enzyme activity.Protease assay. Proteolytic activity was deter-

mined with Hammarsten casein as substrate. Sam-ples (0.5 ml) of culture fluid were incubated at 37 Cwith 0.5 ml of 0.1 M Tris-hydrochloride (pH 8.0) and1.0 ml of 1% casein in 0.1 M Tris-hydrochloride, pH8.0. After 1 h, 3 ml of 5% trichloroacetic acid was

added, and the absorbance at 280 nm was read afterfiltration through Munktell filter no. 0. One unit ofproteolytic activity is defined as the amount ofenzyme giving an increase of 0.01 optical density unitper h at 280 nm under these conditions.

Extracellular DNase assay. ExtracellularDNase activity was tested by a procedure modifiedafter that of Alexander et al. (1). Calf thymus deoxyri-bonucleic acid (DNA) at a concentration of 1.5 mg/mlin 0.025 M glycine buffer (pH 8.6) with 0.002 M CaCl2was denatured for 15 min at 100 C. A 0.2-ml amountof denatured DNA was incubated with 50 gliters ofculture fluid for 30 min at 37 C. A 0.5-ml amount of 7%perchloric acid and 0.5 ml of distilled water were

added to stop the reaction, and the mixture was

centrifuged in the cold. Hydrolyzed DNA was deter-mined by measuring the absorption of the culturefluids at 260 nm. One unit of DNase activity is definedas the amount of enzyme giving an increase of 0.2 opti-cal density unit per h at 280 nm under these condi-tions.

Alkaline phosphatase assay. p-Nitrophenylphos-phate at a concentration of 0.22 mg/ml in distilledwater was used as substrate. A 1.0-ml amount of fluidculture was incubated with 1.5 ml of 0.05 M glycine-sodium hydroxide buffer (pH 10.4) and 1.0 ml ofsubstrate for 30 min at 37 C. Hydrolysis was stoppedby adding 1.0 ml of 10% NaOH. Released p-nitro-phenol was read at 410 nm. One unit of alkalinephosphatase activity is defined as the amount ofenzyme giving an increase of 0.02 per h at 410 nmunder these conditions.

Coagulase assay. Overnight cultures grown inTSB medium were assayed for coagulase. A 0.1-mlamount of culture supernatant fluid was incubatedwith 0.5 ml of rabbit plasma diluted 1:5 with sterilesaline. The mixture was kept at 37 C without shakingfor 2 to 24 h before estimation of the clotting capacity.

Carbohydrate utilization. The ability to utilizedifferent sugars was tested on eosin-methylene blueselective agar plates by the method of McClatchy andRosenblum (15). A large number of bacteria (up to 109cells/ml) can be spread on these plates. The bacteriaable to utilize the sugar will grow as pink colonies in alawn of white nonutilizing cells.

Transport studies. Bacteria were grown in CHMmedium containing 1% galactose to an absorbancy of0.4, corresponding to 2 x 108 cells/ml. The cells werewashed with saline and suspended in 0.1 M Tris-hydrochloride, pH 7.4. "C-Galactose (0.25 mCi; specact 4.7 mCi/mmol) was added to each of two bacterialsuspensions of 10 ml. One was kept at 0 C and theother at 37 C. Samples (1.0 ml) were taken atdifferent times, filtered through membrane filters(0.45 gim; Millipore Corp.), and washed twice with 10ml of cold, sterile saline. The filters were dried at 60 Cfor 1 h. Radioactivity was measured in a toluene-based scintillation mixture (5.3 g of 2,5-diphenylox-azole-1, 4-bis-2-(5-phenyloxazolyl)-benzene mixture[50:3] per liter) with a Beckman LS-233 scintillationcounter.

Chemicals and substrates. Trypticase soy brothwas purchased from Baltimore Biological Laborato-ries, (Cockeysville, Md), and peptone was from DifcoLaboratories (Detroit, Mich.). NTG was obtainedfrom EGA-Chemicals (Germany), and EMS was fromK & K Laboratories Inc., (Plainview, N.Y.). Calfthymus DNA and p-nitrophenolphosphate were fromSigma Chemical Co. (St. Louis, Mo.), and galactose-J_ 14C was from New England Nuclear Chemicals(Germany).

RESULTSProteolytic activity of wild type and prot+

mutants of S. aureus after induction in dif-ferent media. After mutagenization with NTG,about 5% of the survivors showed clear zones onmilk agar. The prot+ (protease positive) cellswere picked and tested on liquid medium withmilk powder. The caseinolytic activity of themutants was about three to ten times higherthan that of the wild type. Two of these mutantswere analyzed in detail and called prot 4 andprot 6. The phage sensitivity pattern was thesame as that of the wild type (80a 53 47 75).Extracellular proteolytic activity could only beinduced by milk or casein in the medium. Noactivity was found extracellularly when themutants were grown in media containing gela-tine or Casamino Acids, nor did commercialsubstrates such as TSB or nutrient broth mediainduce protease production. Thirteen differentcommercial enzymatic digests of casein were

26 J. BACTERIOL.

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

S. AUREUS PROTEASE-PRODUCING MUTANTS

examined as inducers for protease production.The concentrations were varied from 0.001 to1.5%. None of the digests could function as

inducers. Peptonized milk nutrient, however,which is a hydrolysate of skim milk gave aroundhalf the proteolytic activity of whole skim milkpowder.Enzyme production during growth of wild

type and prot+ mutants. The release of pro-tease and DNase into the medium was studiedduring growth in milk-nutrient broth mediumat 37 C. The result is shown in Fig. 1 A and B.The proteolytic activity follows the growthcurves for both strains prot 4 and 6 and reachesa maximum at the beginning of the stationaryphase. DNase production, however, seems todiffer between the two. DNase is always ex-

creted early during growth from strain prot 4and rises rapidly to high activity. The release ofDNase is delayed compared to the proteolyticenzyme from strain prot 6.

E

C4

w

0

4

TIME (HOURS)

.4

w

Ux

?-i

z

a

The wild-type cells will grow normally in thismedium, but there is no detectable amount ofeither protease or DNase. The production ofDNase was also studied with bacteria grown inTSB medium. The enzyme production mainlyfollows the growth curves of strain prot 6 andthe wild-type strain (Fig. 2A-C). Strain prot 4gives the same pattern as for the milk-nutrientbroth medium with a maximum early in thegrowth curve. Strain prot 4 only grows to halfthe cell density observed for strain prot 6 andthe wild-type strain. The stationary phase ofthe cells was reached before an A.2. of 0.4 and0.8, depending on the size and physiologicalconditions of the inoculum. This was also ob-served when strain prot 4 was grown in milk-nutrient broth or peptone. To study whethergrowth factors were limiting, strain prot 4 was

cultivated in double-concentrated TSB me-

dium and TSB medium with addition of yeastextract. Both yeast extract and twice-concen-trated TSB medium stimulated growth, al-though yeast extract was most effective (Fig. 3).To rule out that strain prot 4 did not produceinhibitory factors, thus preventing cell division,strain prot 4 was cultivated in mixed cultureswith the wild-type strain and strain prot 6. Noinhibition was observed. Strain prot 4 also

I

S1

O 2 6 10 14 16 22

TIME (HOURS)

FIG. 1. Protease and DNase production duringgrowth of strain prot 6 (A) and strain prot 4 (B) inmilk-nutrient broth. Symbols: x, absorbance at 524nm; 0, proteolytic activity; A, DNase activity; 0,pH.

E

*0

\wa n

*- t

I-5

I.

CZ MA

z

a3i

I

I, >

S) ag<- :

2 10 14 18 22TIME (HOURS)

I>

i.-

z

50-j

4aI.-

z

303

20 -.4

104z

FIG. 2. DNase production during growth of strainsprot 6 (A), prot 4 (B), and 8325N (C) in TSB.Symbols: x, absorbancy at 524 nm; A, DNase activ-

ity.

VOL. 116, 1973 27

I

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

RYDRN, LINDBERG, AND PHILIPSON

E

C

in 1.2

I-1.0

0.8w

0.6

0.4

0

o0.2z

0

TIME (HOURS)

FIG. 3. Growth of strain prot 4 in different media.Symbols: 0, absorbancy at 524 nm during growth in

TSB; x, absorbancy at 524 nm during growth in 2 x

TSB; U, absorbancy at 524 nm during growth in TSB+ 0.1% yeast extract; *, absorbancy at 524 nm duringgrowth in TSB + 1% yeast extract.

differs morphologically from the wild-typestrain; the cells appear larger and swollen byphase-contrast microscopy. Furthermore, strainprot 4 cells usually stay single compared to theclusters of wild-type cells.

For examination of alkaline phosphatase pro-

duction, the bacteria were cultivated in peptonewhich has a reduced phosphate concentration.Only wild type produced this enzyme (Fig. 4).Thus the prot+ mutants also exhibit a differ-ence in this enzyme activity after mutageniza-tion.

Coagulase of the wild-type, prot 4, and prot 6strains was studied after incubation with rabbitplasma. Coagulation was observed in the tubeswith the wild-type strain and strain prot 6.Even after 24 h, no clotting could be detectedwith strain prot 4.

Influence of pH and buffer concentrationon proteolytic activity during growth. It willbe shown elsewhere that the proteolytic enzymeof strain prot 6 is stable between pH 6 and 9.5(A. -C. Ryden, manuscript in preparation). Theeffect of changing pH in the cultivation mediumwas studied, and the protease activity was

determined. The concentration of Tris bufferwas kept at 0.05 M. There are only slightdifferences in enzyme production between pH7.0 and 7.5 for prot 6 and between pH 7.0 and8.25 for strain prot 4 (Table 1).

The salt concentration was next varied in themedium by changing the Tris buffer from 0.01to 0.1 M. The pH was kept constant at 7.5. Thebuffer concentration did not effect proteaseproduction of the prot+ mutants (Table 2). Thewild type produced no enzyme in 0.1 M bufferalthough the pH was neutral. Enzyme inactiva-tion at low pH did not, therefore, appear toexplain the protease deficiency of the wild type.

Ec0

I-1.2

w 1.00.8

, 0.6C.)

04.0

u

z

m 0.20

Inm.4

0 2 6 10 14 18 22

TIME (HOURS)

I--

5050 Xw

40 n no- F

3 0 X v20 CL

z1 ° -r

4i26

FIG. 4. Alkaline phosphatase production duringgrowth of strains 8325N, prot 4 and prot 6. Symbols:x, absorbancy at 524 nm of strain 8325N; *, absorb-ancy at 524 nm of strain prot 4; *, absorbancy at 524nm of strain prot 6; 0, alkaline phosphatase activityof strain 8325N; A, alkaline phosphatase activity ofstrain prot 4; 0, alkaline phosphatase activity ofstrain prot 6.

TABLE 1. Protease production in milk-nutrient brothat different pHa

ProteolyticStrain Initial pH Final pH activity

(U/ml)

8325 N 7.5 5.2 68.0 6.2 4

prot 4 7.0 7.05 737.25 7.05 767.50 7.45 767.75 7.45 788.0 7.80 588.25 7.90 588.50 8.30 4

prot 6 7.0 7.55 1177.25 7.60 1257.50 7.75 1297.75 8.0 1228.0 8.05 338.25 8.10 228.50 8.25 10

aSkim milk powder was added to a final concentra-tion of 1.5%, and the buffer concentration of Tris was0.05 M.

J. BACTERIOL.28

I:

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

S. AUREUS PROTEASE-PRODUCING MUTANTS

TABLE 2. Protease production in milk-nutrient brothat different concentrations of Tris-hydrochloride

buffer-2

Buffer concn Fia H ProteolyticStrain actvFial pH ty(M) ~~~~~(U/mi)8325 N 0.05 5.20 6

0.10 7.05 2prot 4 0 7.10 84

0.01 7.15 920.025 7.30 960.05 7.45 910.075 7.50 870.10 7.55 71

prot 6 0 7.95 910.01 7.90 920.025 7.85 970.05 7.80 980.075 7.75 920.10 7.80 86

aSkim milk powder was added to a final concentra-tion of 1.5%, and the pH of the Tris buffer was 7.5.

Utilization of carbohydrates by wild-typeand prot+ mutants. The proteolytic activitywas higher when whole skim milk powder wasused rather than casein or an artificial milkmixture containing casein and lactose (Table3). Although the medium was buffered with 0.05M Tris-hydrochloride (pH 7.5), the pH de-creased about two pH units in wild-type cul-tures when lactose was present, due to acidproduction of the utilized lactose. Neitherstrain prot 4 nor strain prot 6 could utilizethis sugar when tested on eosin-methylene blueagar or in fermentation tubes.When glucose was added to the medium, it

was evident that 0.1% glucose increased theproteolytic activity of both strains prot 4 andprot 6, but pH was maintained constant.At 1% glucose, pH decreased in the cultures of

strain prot 6 as for the wild type. Proteaseactivity was also reduced. Strain prot 4, on theother hand, maintained pH and enzyme pro-duction unchanged (Table 4). Glucose was me-tabolized by both prot+ mutants when tested oneosin-methylene blue plates and in fermenta-tion tubes, although the fermentation wasslower for strains prot 4 and prot 6 than for thewild-type strain. Strains prot 4 and prot 6 wereunable to utilize galactose (gal-) and mannitol(mtl-) when grown on eosin-methylene blueagar containing these carbohydrates.Spontaneous reversion from lac- to lac+

and gal- to gal+. The reversions of lac- to lac+and gal- to gal+ were studied on eosin-methylene blue plates. After 3 days of incuba-tion, the large pink colonies in the bacterial

lawn were counted. Table 5 shows the reversionfrequencies. Ten lac+ and 10 gal+ colonies werepicked from both strains prot 4 and prot 6 andassayed for protease production. It was foundthat all the revertants from strain prot 6 utilizinglactose and galactose had the low proteolyticactivity of the wild-type strain. All revertantsfrom strain prot 4 on the other hand still pro-duced 50 to 80% of the original enzyme ac-tivity. All the revertants isolated regainedthe wild-type phenotype with regard to allcarbohydrates, irrespective of the sugar usedfor selection. Phage typing of all revertants gavethe same pattern as for the original mutants.This probably infers that the mutation leading

TABLE 3. Growth and protease production in nutrientbroth (NB) with different components of milk powder

Proteo-

Strain Medium A824 Final activity(U/ml)

8325 N NB + caseina 0.9 7.75 16NB +lactoseb 1.10 5.90 3NB + casein + 1.50 5.50 17

lactoseNB + milkc 1.60 5.50 24

prot 4 NB + casein 0.68 7.40 57NB + lactose 0.45 7.50 8NB + casein + 0.64 7.50 61

lactoseNB + milk 0.69 7.65 107

prot 6 NB + casein 0.82 7.75 43NB + lactose 0.66 7.85 5NB + casein + 0.90 7.80 45

lactoseNB + milk 1.20 7.85 97

a Casein added to a final concentration of 0.54%.bLactose added to a final concentration of 0.77%.c Milk added to a final concentration of 1.5%.

TABLE 4. Protease production in milk-nutrient brothacontaining different concentrations of glucose

ProteolyticStrain Glucose (%) Final pH activity

(U/ml)

8325 0 5.6 240.1 5.2 171.0 5.2 2

prot 4 0 7.5 1040.1 7.4 1731.0 7.1 165

prot 6 0 7.7 1300.1 7.6 1671.0 5.4 24

a The medium was buffered with 0.05 M Tris-HCl,pH 7.5, and contained 1.5% of skim milk powder.

VOL. 116, 1973 29

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

RYDRN, LINDBERG, AND PHILIPSON

TABLE 5. Reversion of strain prot 4 and strain prot 6from lac- to lac+ and gal- to gal+

No. of revert- Reversion fre-

Strain Final ants/ml quency (x 10-6)bacteria/mllac+ gal + lac+ gal+

prot 4 2.8 x 107 154 81 5.5 2.9prot 6 2.9 x 107 203 136 7.0 4.7

to defective carbohydrate utilization is a single-point mutation.Mutagenic reversion of strain prot 4 from

lac- to lac+. Strain prot 4 was treated withNTG and EMS to determine whether it waspossible to obtain lac+ revertants which hadalso lost proteolytic activity. Mutagenizationwith NTG and EMS was carried out as de-scribed but without phenotypic expression. Thereversion frequency to lac+ after NTG and EMStreatments were 7.0 x 10-5 and 2.4 x 10- ,

respectively. This frequency is only 10 timeshigher than that of spontaneous reversion.About 125 reverted colonies were tested on milkagar to screen for cells with and without proteo-lytic activity. Twenty percent of the testedcolonies after NTG treatment gave turbid or nozones at all. The corresponding value after EMSwas 30%. Forty colonies showing turbid zoneswere cultivated in milk-nutrient broth to deter-mine the proteolytic activity quantitatively. Of21 NTG-treated cells, all but two had lost theproteolytic activity after reversion to lac+. Therevertants from EMS treatment showed an-other protease pattern. Only 3 of 19 colonieswere protease negative when tested quantita-tively; the rest of the lac+ colonies still pro-duced protease as strain prot 4. Although NTGmay have caused multiple mutations, the re-sults suggest that the regaining of carbohydrateutilization and the loss of proteolytic activitymay occur together.Uptake of carbohydrates of wild-type and

prot+ mutants. 'IC-labeled galactose (finalconcentration 5.3 x 10-6 mmol/ml) was addedto cells suspended in buffer, and the uptake ofthe carbohydrate was followed at different timeintervals. Strain prot 6 transports no galactoseeither at 0 or at 37 C (Fig. 5). The uptake at 0 C,however, must be slow, since the wild typeshows no increase in "C-incorporation duringthe 2 h studied.

DISCUSSIONInduction of proteolytic enzyme in S. aureus

8325N is at least partially specific since milkpowder and casein could be used as inducers,whereas gelatin and several protein hydrolysatescould not. Free casein only gave 50% of the

proteolytic activity obtained with milk. Thecasein hydrolysate medium used by Tiruna-rayanan (22) or the CCY medium described byArvidson et al. (2) gave no extracellular pro-tease production. Thus, hydrolysate of caseincould not replace the native protein.The defective carbohydrate transport seems

identical to the defect in the Car- mutantdescribed and carefully analyzed by Egan,Morse, and Hengstenberg (5-7, 10-12). Thismutant could not utilize 11 carbohydrates. Theintracellular metabolism of the carbohydrateswas intact, but they could not be transportedinto the cells. The reversion frequency was,however, about 10 times lower than the frequen-cies reported in this paper. Transduction of theCar+ character to Car- cells was successful. TheCar+ revertants and transductants showedwild-type phenotype with respect to all car-bohydrates. Transduction experiments carriedout with strain prot 6 as recipient and phage 80a propagated on strain 8325N (lac+) showedthat the transduction frequencies were not sig-nificantly different from the spontaneous rever-sion. Isolation of spontaneous revertants fromlac+ to lac- with strain 8325N was also at-tempted, but no white colonies could be de-tected among the rapidly growing bacteria uti-

-JI

`I1U

0 20 40 60 80 100 120TIME (MINUTES)

FIG. 5. Uptake of galactose-1-14C by nondividingcells of strains 8325N, and prot 6 at 37 and 0 C.Symbols: A, radioactivity at 37 C of strain 8325N; *,radioactivity at 0 C of strain 8325N; A, radioactivityat 37 C of strain prot 6; 0, radioactivity at 0 C ofstrain prot 6.

30 J. BACTERIOL.

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

S. AUREUS PROTEASE-PRODUCING MUTANTS

lizing the carbohydrate.Glucose has been reported to repress protease

production of staphylococci (22), as of manyother proteases from different bacteria. Thiswas ascribed to catabolic repression, thus lower-ing the intracellular level of cyclic adenosinemonophosphate (AMP) (19). Cyclic AMP mightbe necessary to trigger the ribonucleic acidpolymerase to transcribe a catabolic-sensitiveoperon coding for the enzyme. This idea mightexplain the regulation of potease of strain prot 6since glucose inhibited the enzyme productionwhen added to the medium. This would, how-ever, require that the wild-type strain producesprotease when the cells were grown on caseinwithout glucose, which was not the case. Theenzyme should then be derepressed since thegenetic information for the protease must bepresent. Some other factor must therefore beinvolved in the repression of the protease of thewild type. Since strain prot 4 did not show adecreased proteolytic activity when 1% glucosewas added, the regulation is difficult to explainon the basis of cyclic AMP.A common control mechanism for several

extracellular products was suggested byForsgren et al. (8). They isolated several proteinA-deficient mutants and found that several ofthese mutants also lacked other phenotypiccharacters, such as nuclease, coagulase, a-hemolysin, fibrinolysin, mannitol utilization,and the phage-type pattem. In reversion andtransduction experiments, several of theseproperties were regained together. Similar re-sults were reported by Omenn and Friedman(16) who isolated mutants lacking nuclease,coagulase and 8-hemolysin. The revertants re-gained all the lost characters.

Pleiotrophic changes might also be associatedwith the cell envelope. Chatterjee et al. (3)described a phage-resistant mutant of S. aureusH that differed phenotypically from the wildtype in many respects. The phage-resistantmutant grew more slowly, was larger, andproduced smaller amounts of autolytic enzyme.The reversion experiments resulted in wild-typecells with respect to all these characters. Themutant was previously shown to lack polymericteichoic acid, which suggests some general de-fect in the envelope, leading to the pleiotrophiccharacter. Single locus changes in envelopegenes might thus affect several properties of thecells and suggest another explanation of thepleiotrophic mutants described in this paper.

ACKNOWLEDGMENTS

The capable technical assistance of Inger Olsson-Langeis greatfully acknowledged.

Financial support was received from the Swedish MedicalResearch Council.

LITERATURE CITED1. Alexander, M., L. A. Heppel, and J. Hurwitz. 1961. The

purification and properties of micrococcal nuclease. J.Biol. Chem. 236:3014-3019.

2. Arvidson, S., T. Holme, and T. Wadstrom. 1970. Forma-tion of bacteriolytic enzymes in batch and continuousculture of Staphylococcus aureus. J. Bacteriol.104:227-233.

3. Chatterjee, A. N., D. Mirelman, H. J. Singer, and J. T.Park. 1969. Properties of a novel pleiotropic bacterio-phage-resistant mutant of Staphylococcus aureus H. J.Bacteriol. 100:846-853.

4. Drapeau, G. R., J. Boily, and J. Houmard. 1972. Purifica-tion and properties of an extracellular protease ofStaphylococcus aureus. J. Biol. Chem. 247:6720-6726.

5. Egan, J. B., and M. L. Morse. 1964. Carbohydratetransport in Staphylococcus aureus. I. Genetic andbiochemical analysis of a pleiotropic transport mutant.Biochim. Biophys. Acta 97:310-319.

6. Egan, J. B., and M. L. Morse. 1965. Carbohydratetransport in Staphylococcus aureus. II. Characteriza-tion of the defect of a pleiotropic transport mutant.Biochim. Biophys. Acta 109:172-183.

7. Egan, J. B., and M. L. Morse. 1966. Carbohydratetransport in Staphylococcus aureus. IJI. Studies of thetransport process. Biochim. Biophys. Acta 112:63-73.

8. Forsgren, A., K. Nordstrom, L. Philipson, and J. Sjo-quist. 1971. Protein A mutants of Staphylococcusaureus. J. Bacteriol. 107:245-250.

9. Goldbach, W., and H. Haenel. 1964. Uber Lysozym- undProtease-bildung bei Staphvlokokken. Z. Zentrabl.Bakteriol. Parasitenk. Infektionskr. Hyg. Abt. 1 Orig.195:201-205.

10. Hengstenberg, W., J. B. Egan, and M. L. Morse. 1967.Carbohydrate transport in Staphylococcus aureus. V.The accumulation of phosphorylated carbohydrate de-rivatives, and evidence for a new enzyme-splittinglactose phosphate. Proc. Nat. Acad. Sci. U.S.A.38:274-279.

11. Hengstenberg, W., J. B. Egan, and M. L. Morse. 1968.Carbohydrate transport in Staphylococcus aureus. VI.The nature of the derivatives accumulated. J. Biol.Chem. 243:1881-1885.

12. Hengstenberg, W., W. K. Penberthy, K. L. Hill, and M.L. Morse. 1969. Phosphotransferase system of Staph-ylococcus aureas: its requirement for the accumula-tion and metabolism of galactosides. J. Bacteriol.99:383-388.

13. Lieb, F. L., and B. Brenner. 1961. Ober Proteasen desStaphylokokkus pyogenes aureus. Arch. Hyg.145:161-166.

14. Lindberg, M., J. E. Sj6strom, and T. Johansson. 1972.Transformation of chromosomal and plasmid charac-ters in Staphylococcus aureus. J. Bacteriol.109:844-847.

15. McClatchy, G. K., and E. D. Rosenblum. 1963. Inductionof lactose utilization in Staphylococcus aureus. J.Bacteriol. 86:1211-1215.

16. Omenn, G. S., and J. Friedman. 1970. Isolation ofmutants of Staphylococcus aureus lacking extracellularnuclease activity. J. Bacteriol. 101:921-924.

17. Robinson, J., F. S. Thatcher, and J. Montford. 1960.Studies with staphylococcal toxin. V. Possible iden-tification of alpha hemolysin with a proteolytic en-zyme. Can. J. Microbiol. 6:183-194.

18. Robinson, J., F. S. Thatcher, and J. Montford. 1960.Studies with staphylococcal toxins. VI. Some proper-ties of a non-hemolytic dermonecrotic toxin producedby some strains of Staphylococcus aureus. Can. J.Microbiol. 6:195-201.

VOL. 116, 1973 31

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

RYDRN, LINDBERG, AND PHILIPSON

19. Schwartz, D., and J. R. Beckwith. 1970. Mutants missinga factor necessary for the expression of catabolite-sensi-tive operons in E. coli, p. 417. In J. R. Beckwith andD. Zipser (ed.), The lactose operon. Cold SpringHarbor, New York.

20. Tirunarayanan, M. O., and G. Lundblad. 1966. Investiga-tions on the enzymes and toxins of Staphylococci.Proteolytic enzymes and their purification. Acta Pa-thol. Microbiol. Scand. 68:135-141.

21. Tirunarayanan, M. O., and G. Lundblad. 1968. Investiga-tions on the enzymes and toxins of Staphylococci.

Study of hyaluronidase by the viscosimetric method.Acta Pathol. Microbiol. Scand. 73:211-219.

22. Tirunarayanan, M. O., and G. Lundblad. 1968. Investiga-tions on the enzymes and toxins of Staphylococci.Relationship of pH to growth and production of en-zymes. Acta Pathol. Microbiol. Scand. 74:274-286.

23. Vesterberg, O., T. Wadstr6m, K. Vesterberg, H. Svens-son, and B. Malmgren. 1967. Studies on extracellularproteins from Staphylococcus aureus. I. Separationand characterization of enzymnes and toxins by isoelec-tric focusing. Biochim. Biophys. Acta 133:435-445.

32 J. BACTERIOL.

on October 13, 2020 by guest

http://jb.asm.org/

Dow

nloaded from