INFECTION AND IMMUNITY, May 1992, p. 1820-18250019-9567/92/051820-06$02.00/0Copyright C 1992, American Society for Microbiology

Antigenic Role of Stress-Induced Catalase of Salmonellatyphimurium in Cell-Mediated Immunity

KEIKO KAGAYA, YOZO MIYAKAWA, KOJI WATANABE, AND YOSHIMURA FUKAZAWA*Department of Microbiology, Yamanashi Medical College, Yamanashi 409-38, Japan

Received 3 September 1991/Accepted 14 February 1992

The ability of the H202-induced catalase of Salmonella typhimurium to induce cell-mediated immunityagainst S. typhimurium infection in mice was examined. When exponentially growing cells of S. typhimuriumwere treated with 20 ,uM H202, the cells resisted killing by 1 mM H202 and showed the induction of a new

species of catalase in addition to the constitutively produced one. Two molecules of catalases in S. typhimuriumwere isolated from mutant strains: H202-induced catalase (catalase II, 320 kDa), from a regulatorygene-deficient oxyRI mutant, and constitutive catalase (catalase I, 350 kDa), from a katG gene-deleted mutant.When mice were inoculated with a sublethal dose of live cells, an intensive protective immunity (100%b survivalat 3 weeks) after challenge with a virulent strain associated with the delayed-type footpad hypersensitivity(DTH) reactions to both catalase I and catalase II was induced. Conversely, mice immunized withformalin-killed virulent S. typhimurium did not elicit protective immunity or DTH to either catalase. Whenmice were immunized with catalase I or catalase II, an enhanced protection (to a certain extent: 50% survivalat 3 weeks) was induced in mice immunized with catalase II associated with DTH which did not cross-react withcatalase I but not in those given catalase I. These results suggest that H202-induced stress proteins, includingcatalase II, are the dominant antigens for cell-mediated immunity in live cells of S. typhimurium and that a

burst of such stress proteins in live salmonellae in phagocytes is responsible for the induction of cell-mediatedimmunity that is largely involved in the protection of susceptible mice against Salmonella infection.

Studies on typhoid immunity have been most commonlyconducted by using Salmonella typhimurium infection inmice as a model (6, 18). It is generally accepted that hostdefense against systemic salmonellosis involves two types ofimmunity, namely, humoral and cell-mediated immunity(CMI) (1, 6, 18, 25, 30). However, there are serious discrep-ancies in the individual roles of such immunities on hostdefense against systemic Salmonella infections (18), al-though the contribution of humoral immunity or CMI toprotection reportedly varies among mouse strains (11), i.e.,while protection of inherently susceptible mice against S.typhimurium requires CMI, antibody is shown to protectinherently resistant mice but not inherently susceptiblemice.

Intensive CMI is induced by live avirulent Salmonellastrains (25), whereas humoral immunity is elicited by killedvirulent cells or lipopolysaccharides (1, 10, 21). With respectto CMI, a little understood aspect is the mechanism bywhich live bacteria induce stronger CMI than do killedbacteria.On the other hand, when cells are subjected to environ-

mental stress stimuli, a variety of stress proteins are released(26). An elevated synthesis of these stress proteins ofintracellular bacteria in host phagocytes would increase theiravailability for antigen recognition by the host immunesystem. In fact, an immune response to stress proteins isobserved in infection with Mycobacterium species and someparasites (20, 33).

Enteric bacteria synthesize several stress-induced en-

zymes, such as superoxide dismutase and catalase, that mayprotect bacterial cells from oxidative damage in phagocytes(13). The resistance of S. typhimurium to killing by H202when pretreated with a nonlethal level of H202 in vitro was

* Corresponding author.

attributed to such enzymes induced in bacteria under stress(4).The present study was designed to determine the role of

stress-induced catalase in CMI against S. typhimurium in-fection and to clarify the mechanism of induction of CMIagainst S. typhimurium infection in mice by using an H202-induced catalase (a stress protein). Among the 30 proteins inS. typhimurium induced by H202 (26), catalase was selectedbecause of its defined biological activity. We provide evi-dence that stress-induced catalase rather than the constitu-tive one induces an increased (to a certain extent) protectionassociated with the delayed-type hypersensitivity (DTH)response against S. typhimurium infection in mice.

MATERLILS AND METHODS

Bacterial strains and cultivation. A nalidixic acid-resistantvirulent strain of S. typhimurium SH5170 obtained from T.Arai, Meiji College of Pharmacy, Tokyo, Japan, and a mildlyvirulent strain of LT2 were maintained by bimonthly trans-fers on brain heart infusion agar (Difco Laboratories, De-troit, Mich.) slants. For catalase induction, the organism wasgrown in the minimal medium broth described by Vogel andBonner (32) containing 0.4% glucose as a carbon source

(VBG broth). In other experiments, the bacteria were grownin brain heart infusion broth at 37°C for 18 h with shaking.An oxyRl mutant, TA4100 (4), and a katG gene deletionmutant, TA4113 (4, 26), which were obtained from B. N.Ames, University of California, were used as sources of theH202-induced catalase and the constitutive one, respec-tively. They were cultured with Luria broth (4).

Induction of catalase. Overnight cultures of S. typhimu-rium in VBG broth were inoculated into fresh VBG broth toan initialA650 of 0.02. When cells had reached an A650 of 0.2,H202 was added to a final concentration of 20 ,uM. Afterincubation for 60 min at 37°C with shaking, the culture

1820

Vol. 60, No. 5

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 18

Dec

embe

r 20

21 b

y 12

5.16

3.14

3.16

5.

PROTECTION BY STRESS CATALASE OF S. TYPHIMURIUM 1821

suspensions were directly assayed for catalase activity andfor resistance to H202. The resistance of the bacteria toH202 was determined by the decrease in CFU after a 60-mintreatment with 1 mM H202 at 37°C.

Catalase assay. Catalase activity was determined by thedecrease in the A240 of H202 (2). The catalase activity of awhole-cell suspension was assayed by adding 100 ,ul of 340mM H202 in 0.05 M potassium phosphate buffer (PB), pH7.0, to 2.9 ml of cell suspension. The time required for theA240 of a 3-ml suspension to decrease from 0.45 to 0.40(decomposition, 3.45 ,umol) was determined, and the activityrequired to decompose 1 ,umol of H202 per min was desig-nated as 1 U. The relative catalase activity of each fractionfrom the chromatography column was assayed as follows:100 pul of 340 mM H202 was added to each sample in 2.9 mlof 0.05 M PB (pH 7.0), and the sample was observed for 1min at room temperature for a decrease of A240. Catalaselocated on polyacrylamide gels was stained by the method ofHarris and Hopkinson (16), with slight modifications. Inbrief, gels were incubated in 0.03% H202 in distilled waterfor 3 min, washed with distilled water, and then immersed ina freshly prepared 1:1 mixture of 1% potassium ferricyanideand 1% ferric chloride. When the container was gentlyagitated for a few minutes, a yellow band due to catalaseappeared against a blue-green background.

Isolation and purification of catalases. Catalase in S. typhi-munium was isolated and purified as described for peroxi-dases in Escherichia coli (5), with slight modifications.Briefly, S. typhimurium TA4100 and TA4113 cells grown inLuria broth for 18 h under shaking were harvested, washed,suspended in 0.05 M PB (pH 7.0) at a concentration of 0.2g/ml, and sonicated for 4 min at a volume setting of 1(Ultrasonic Disruptor UP-1; Tomy Seiko Co., Tokyo, Ja-pan). After centrifugation, streptomycin sulfate (SigmaChemical Co., St. Louis, Mo.) was added to the supernatantto give a 2.5% solution to remove nucleic acids. The precip-itate was removed by centrifugation, and solid ammoniumsulfate was added to the supernatant. The precipitate formedwith 35% saturation was collected, dialyzed, applied on aDEAE-Sephacel (Pharmacia, Uppsala, Sweden) columnequilibrated with 0.05 M potassium acetate buffer (pH 5.5),and eluted with a gradient of 0.05 to 0.5 M potassium acetatebuffer (pH 5.5). Fractions containing catalase activity werepooled, concentrated, dialyzed, and applied on a column ofSephacryl S-300 (Pharmacia) equilibrated with 0.05 M PB(pH 7.0). After rechromatography on the above column,fractions with high activity were collected, concentrated byultrafiltration, dialyzed against phosphate-buffered saline(PBS), and used for immunization and footpad reaction.Apoferritin (480 kDa; Schwarz/Mann, Orangeburg, N.Y.),bovine liver catalase (245 kDa; Sigma), and immunoglobulinG (160 kDa; Schwarz/Mann) were used as molecular massstandards. The protein content of catalase was determinedby the method of Lowry et al. (24).

Electrophoresis. Native polyacrylamide gel electrophore-sis (PAGE) was performed at pH 8.0 with a 4% gel whichwas stained for catalase activity (16). Sodium dodecyl sulfate(SDS)-PAGE was carried out by the method of Laemmli andFavre (23) with a 10 to 20% gradient gel which was stainedwith Coomassie brilliant blue R (Sigma).

Immunization. Male BALB/c mice (aged 6 weeks) wereobtained from Charles River Japan, Inc., Atsugi, Kanagawa,Japan. Two groups of mice were infected intraperitoneally(i.p.) with 5 x 10 viable cells of S. typhimurium LT2 (0.2550% lethal dose [LD50) grown in brain heart infusion brothor injected i.p. with 1 x 108 formalin-killed cells of SH5170,

the latter given twice with a 2-week interval. These micewere used for the DTH reaction and challenge experiment 3weeks after inoculation of viable cells and 10 days after thelast injection of killed cells, respectively. Mice of the thirdgroup were injected i.p. with 50 ,ug of the catalase I orcatalase II in PBS followed by an i.p. booster injection of 50,ug of each catalase in PBS 2 weeks after the first injection.Immune responses and protection were determined 10 daysafter the booster injection. For controls, other groups ofmice were injected with PBS alone.

Evaluation of protective immunity of mice. A nalidixicacid-resistant virulent strain of S. typhimurium SH5170, ofwhich the LD50 in BALB/c mice following i.p. challenge wasapproximately 10 cells at 3 weeks, was used for the chal-lenge. The control and immunized mice were challenged i.p.with 2 x 102 cells (20 LD50) of S. typhimurium SH5170, bywhich control mice died on day 6 or 7. Protection wasevaluated by bacterium counts in spleens at three differenttimes after challenge and by percent survival and meansurvival days of each of 10 mice during 3 weeks following thechallenge (6). The number of bacteria in the spleens (homog-enized with a Teflon mortar and pestle) of the infected micewere determined on days 3, 5, and 7 after the challenge (6).Viable counts were made by plating the homogenates onnutrient agar (Difco) containing 50 ,ug of nalidixic acid perml, the number of bacteria per spleen was determined, andthe results were expressed as means + standard errors (SE)for three mice per point.DTH assay. DTH was determined by footpad swelling by

using alum-precipitated catalases as antigens (7). Controland immunized mice were injected in the right hind footpadwith 20 pul of alum-precipitated catalase containing 10 p.g ofprotein, and 20 p,l of PBS was injected into the left hindfootpad as a control. The thicknesses of the right and leftfootpads were measured with dial-type calipers (H. KroeplinGmbH, Schlechtern, Germany) 24 h after injection of theantigen, and the thickness of the PBS-injected footpad wassubtracted from that of the catalase-injected footpad for eachmouse. Results were expressed as the mean ± SE of thedifference in footpad thickness for five mice per group.

Statistics. Statistical significance of the data was deter-mined by Student's t test. A P value of less than 0.05 wasconsidered significant.

RESULTS

Induction of catalase in S. typhimurium. When exponen-tially growing cells of S. typhimunum SH5170 in VBG brothwere treated with 20 p.M H202 for 60 min, they wereresistant to killing by 1 mM H202, in contrast to cells thathad not been treated; viabilities of treated and untreatedcells 60 min after the addition of 1 mM H202 to the cultureswere 80 and 0.3%, respectively. The specific activity ofcatalase in H202-treated cells increased during the treat-ment, showing a fivefold-higher activity than that of theuntreated cells (Table 1). PAGE analysis of the catalaseactivity in the H202-treated and untreated cells of strainSH5170 showed a difference in the pattern of the molecularspecies of catalase in these cells. The major catalase activityin the H202-treated cells was detected as a faster-movingband in addition to a slower-moving band. In contrast, in thecontrol cells, only a slower-moving band was detected (Fig.1A and B). S. typhimurium LT2 showed similar results inH202 resistance and PAGE profile of catalase when stimu-lated with 20 p.M H202 (data not shown). These resultsindicate that H202 induces a large amount of catalase, which

VOL. 60, 1992

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 18

Dec

embe

r 20

21 b

y 12

5.16

3.14

3.16

5.

1822 KAGAYA ET AL.

TABLE 1. Catalase induction and adaptation to H202in S. typhimunuma

H20O Catalase activity Mean no. of bacteria/ml (log)H202at time of +SE % Viability

treat- cattienoge_____________ after

ment challenge Before 60 min after challenge(U/A650 U) challenge challenge

- 1.20 ± 0.39 8.84 ± 0.09 6.36 ± 0.35 0.3+ 6.24 ± 1.08 8.82 ± 0.03 8.72 ± 0.05 80.3

a Exponentially growing cells of S. typhimurium SH5170 were incubatedwith or without 20 ,uM H202 for 60 min at 37'C and assayed for catalaseactivity and resistance to challenge with 1 mM H202.

AO. 6 480

Bis a minor product in untreated cells. The slower- andfaster-moving catalases were tentatively named catalase Iand catalase II, respectively.

Purification of catalases. An oxyRi mutant and a katGmutant of S. typhimurium were used for the preparation ofthe antigens. The oxyRl mutant TA4100 was derived fromstrain LT2, in which catalase II activity and eight otherH202-inducible proteins were constitutively overexpressedbecause of the mutation in the oxyR gene encoding theregulator protein for H202-inducible genes (4). The katGmutant TA4113 was deleted for the entire region of the katGgene encoding H202-inducible catalase (4, 26). Cells ofTA4100 and TA4113 strains grown in Luria broth produced94 and 70 U of catalase per A650, respectively, whereasstrains SH5170 and LT2 produced 13 and 19 U of catalaseperA650, respectively. Characteristics of catalase activity ofthese mutants were confirmed by PAGE analysis (Fig. 1Cand D); TA4100 produced a faster-moving catalase (catalaseII), and TA4113 produced only a slower-moving one (cata-lase I). Catalase activity of supematants extracted from eachstrain of S. typhimurium grown in Luria broth for 18 h wasprecipitated with 35% ammonium sulfate, fractionated byDEAE-Sephacel column chromatography with gradient elu-tion, and then applied on a Sephacryl S-300 column. Afterrechromatography on the above column, catalase I appeared

A B C DABCD

_ ~~~~--ln.~~~~~~~~~~~~~~~~~~~~4

FIG. 1. PAGE profile of catalases extracted from S. typhimu-num strains. Sonication supernatants from cells grown in VBGmedium (lanes A and B) or Luria broth (lanes C and D) were directlyused as samples. Lane A, catalases extracted from control SH5170cells (equivalent to 109 cells); lane B, catalases from H202-treatedSH5170 cells (equivalent to 5 x 10' cells); lane C, catalase fromoxyRi mutant TA4100 (equivalent to 6 x 107 cells); lane D, catalasefrom katG mutant TA4113 (equivalent to 5 x 107 cells).

I0N0

0

N

1.2

.1 1°

Fraction No. (2 ml)FIG. 2. Rechromatography of catalases by gel filtration on Seph-

acryl S-300. Fractions of catalase I (A) and catalase II (B) onSephacryl S-300 were rechromatographed on a Sephacryl S-300column equilibrated with 0.05 M PB (pH 7.0). The fractions withactivity were calibrated with the following molecular mass markers:apoferritin (480 kDa), bovine liver catalase (245 kDa), and immuno-globulin G (160 kDa). Solid lines indicate protein contents. Dottedlines indicate catalase activity expressed as the decrease in A240 perminute by 10 p,l of each fraction in 3 ml of 0.05 M PB (pH 7.0). Thefractions with high activity, indicated by horizontal bars, werecollected and used for immunization.

in a higher-molecular-weight region than catalase II; themolecular masses of catalase I and catalase II were esti-mated to be approximately 350 and 320 kDa, respectively(Fig. 2). The fractions of catalase I and catalase II wereshown to be almost homogeneous by SDS-PAGE (Fig. 3).The yields of catalase I and catalase II were approximately78 mg (0.06% of cell wet weight) and 96 mg (0.07%),respectively.Immune response to catalases in mice immunized with killed

or live cells of S. typhimurium. To investigate the possibilityof participation of DTH to catalase II in protection ofimmunized mice, DTH responses to catalase I and catalaseII after immunization of mice with killed or live cells of S.typhimurium were determined in relation to protective im-munity (Table 2). When the mice were inoculated with livecells, a footpad reaction (DTH) to catalase II was exhibited3 weeks after inoculation, indicating that catalase II isgenerated in live S. typhimurium and induces a T-cell subsetthat mediates DTH (TDTH, Thl) in vivo. Although these micealso showed a DTH response to catalase I, the reaction wasweaker than that to catalase II. On the other hand, miceinjected twice with formalin-killed cells did not exhibit afootpad reaction to either catalase. Furthermore, when thesemice were infected with S. typhimurium, an intensive pro-tective immunity was demonstrated in mice immunized withlive cells but not in mice injected twice with formalin-killed

INFECT. IMMUN.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 18

Dec

embe

r 20

21 b

y 12

5.16

3.14

3.16

5.

PROTECTION BY STRESS CATALASE OF S. TYPHIMURIUM 1823

TABLE 3. Cell-mediated immune response and protectionagainst S. typhimunium infection of mice

immunized with catalases

Mouse Footpad swelling' Mean tybimmuni- (0.1 mm) survvalb Mortalityization Catalase I Catalase II (days)

None 0.7 + 0.3 0.7 + 0.6 6.5 + 0.2 100Catalase I 2.7 + 0.6C 0.7 0.6d 7.1 ± 0.2d 100Catalase II 0.7 ± 0.6d 3.8 t 0.4c 16.7 t 1.6e 50

a Mean + SE (n = 5). Assayed as described in footnote a to Table 2.b Mean ± SE (n = 10). Determined as described in footnote b to Table 2.c P < 0.01 versus control mice.d Not significant versus control mice.e p < 0.01 versus control and catalase I-immunized mice.

17 -

FIG. 3. SDS-PAGE analysis of the fractionated catalases. Cata-lase I and catalase II rechromatographed on Sephacryl S-300 wereapplied on a 10 to 20% gradient gel and stained with Coomassiebrilliant blue. Lane A, catalase I; lane B, catalase II. Ovotransferrin(77 kDa), bovine serum albumin (67 kDa), ovalbumin (45 kDa),chymotrypsinogen A (26 kDa), and myoglobin (17 kDa) were used asmolecular mass standards. The monomeric forms of catalases withmolecular weights of around 77,000 were found.

virulent cells (Table 2). These results suggest that Thl tocatalase II induced by immunization with live cells mayparticipate in protection against S. typhimurium infection.

Catalase-induced immune response and protective immu-nity to S. typhimurium infection in mice. When mice wereimmunized with catalase I or catalase II, a significant foot-pad reaction only to homologous catalase was elicited,indicating no cross-reaction between two catalases (Table 3).Blast transformation with corresponding catalases was alsoobserved in spleen cells from immunized mice (data notshown). Mice immunized with the catalases were challengedwith 20 LD50s of S. typhimurium SH5170, and protectionwas determined by mortality and by bacterium counts in thespleens. Mice immunized with catalase II exhibited morethan twofold-longer survival time, and half of the mice

TABLE 2. Cell-mediated immune response to catalases andprotection against S. typhimurium infection in mice

immunized with killed or viable cells of S. typhimunum

Mouse Footpad swelling' Meanimmuni- (0.1 mm) survivalb Mortality"zation Catalase I Catalase II (days) (%)

None 0.6 ± 0.6 0.4 + 0.3 6.2 ± 0.2 100Killed cells 1.2 + 0.6C 0.6 + 0.7c 7.0 + 0.4c 100Live cells 2.7 ± 0.6d 3.5 ± 0.5e >21.Of 0

a Mice (n = 5) were injected with catalase I or II in the right hind footpadand with PBS in the left hind footpad. Data are expressed as the mean ± SEof the difference in thickness between the right and left footpads for eachmouse.

b Mice (n = 10) were observed for death during 3 weeks followingchallenge, and survival was expressed as the mean number of days ± SE.

c Not significant.d p < 0.05.e p < 0.oi.fP < 0.001.

survived 3 weeks after challenge (Table 3). They also exhib-ited a marked decrease in the number of bacteria in compar-ison not only with untreated controls, but also with miceimmunized with catalase I, although the bacterial popula-tions markedly varied among mice at day 7 (Fig. 4). On theother hand, mice immunized with catalase I that exhibitedDTH to the antigen did not show protective immunity; all ofthem died within 8 days and exhibited nearly the samebacterial number in the spleens as did untreated controls at5 days after infection (Table 3; Fig. 4). These results suggest

7

c

a)a)

U)

Cu

L

U1)4-)

a.)

u

-o'4-

0

L.

U1)

E

0

5F

31

1i

3 5 7

Days after infectionFIG. 4. Effects of immunization with catalase I and catalase II on

growth of S. typhimurium in the spleens of mice. Control mice (-)and mice immunized with catalase I (0) or catalase II (A) werechallenged with S. typhimurium SH5170 (20 LD50s), and the num-bers of bacteria in spleens were determined at the days indicated.Each point consists of three mice.

A B

kDa77_ _67m

450.

VOL. 60, 1992

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 18

Dec

embe

r 20

21 b

y 12

5.16

3.14

3.16

5.

1824 KAGAYA ET AL.

that stress-induced rather than constitutive catalase is one ofthe protective antigens in live cells of S. typhimurium.

DISCUSSION

The findings presented here demonstrate that purifiedstress-induced catalase (catalase II, 320 kDa) rather thanconstitutive catalase (catalase I, 350 kDa) induces an in-creased protection to S. typhimunum infection to a certainextent associated with a DTH response in mice.The profiles of native PAGE for these catalases suggest

that catalase I is related to hydroxyperoxidase III (HP-III)with a slower moving band and that catalase II is related toHP-I-II (catalase/peroxidase) with a faster-moving band inthe strain of S. typhimurium LT2 reported on by Christmanet al. (4) and Morgan et al. (26). We used two mutants assources of the catalases. One is an oxyRI mutant, in whichcatalase II activity was constitutively overexpressed be-cause of the mutation in the oxyR gene encoding the regu-lator protein for H202-inducible genes (4). The another is akatG mutant, deleted for the entire region of the katG geneencoding H202-inducible catalase (4, 26). These two mutantsoverproduced each catalase, and they were precipitatedfrom a sonication supernatant with a low concentration ofammonium sulfate (lower than 35%); at that concentrationalmost all other proteins were not precipitated, which wasadvantageous in obtaining highly purified catalases by asimple procedure.

Elucidation of the mechanism by which attenuated livesalmonellas induce intensive protective immunity in mice(whereas killed virulent salmonellas do not) has long beenlacking. However, for our present finding with H202-in-duced catalase, the following possibility is given. When miceare immunized with live bacteria, stress protein-specificT-cell clones of TDTH (Thl) (3) are generated. Such T-cellpopulations specific for stress proteins may not be generatedin mice immunized with killed salmonellas. After challenge,stress proteins are elicited in phagocytes and are predomi-nantly presented by macrophages that stimulate the afore-mentioned primed T-cell clones, leading to efficient macro-phage activation through the release of lymphokines (14, 19).Similarly, in the case of immunization with stress-inducedcatalase II alone, a catalase II-specific primed T-cell clone isstimulated after the challenge, leading to lymphokine releasefor macrophage activation. The lack of protection in miceimmunized with catalase I may be due to a relatively lowfrequency of antigen presentation by macrophages afterchallenge infection, in spite of the presence of primed T-cellclones, thus producing a footpad DTH reaction.Although protection induced by catalase II is not satisfac-

tory, so far as CMI is concerned, we suggest that suchH202-induced proteins, including catalase II, play a role forthe intensive protective immunogenicity of live attenuatedSalmonella strains. The reason for a lower protective immu-nogenicity of catalase II than that of live S. typhimurium isnot clear at present. However, the following possibilities aregiven: (i) a variety of stress proteins other than catalase II inlive bacteria may be required for full induction of CMI, and(ii) the present immunization procedure for a single protein(catalase II) may not be an adequate condition to induce fullCMI with regard to dosage, timing, or route of injection. Aninvestigation to clarify the latter possibility is under way inour laboratory. Although there has been general agreementthat viable vaccines, inducing CMI, will be protectiveagainst S. typhimurium infection in inherently highly suscep-tible mice (11), e.g., BALB/c mice (27), recently it was

claimed that acquired immunity to the S. typhimuriuminfection is manifested by the synergistic actions of circulat-ing antibodies, macrophages with cytophilic antibodies, andDTH (18). If humoral immunity by antibodies producedagainst bacterial cell surface components (22, 29) partlycontributes to full protection against Salmonella infection,even in the susceptible mice, a lack of such humoral immu-nity might be another reason. It seems unlikely that antibodyto catalase II is involved in the protection mechanisms, sincecatalase II is thought to be located in the cell membrane or inthe periplasmic space of the cell wall and catalase II is notthe only virulence factor required, i.e., avirulent strainsproduce catalase II by stimulation with H202 as well.To study the mechanism of immunity to salmonellosis,

there have been a number of experimental systems andmethods varying with the laboratory. In this study we usedthe methods of determining serial bacterium counts inspleens and mean survival day as well as mortality rate.These parameters would be adequate in this study to assessprotection. The growth profiles of bacteria in spleens of miceimmunized with catalase II suggest that the movement of thebacterial population at day 7 of infection may determine theoutcome of survival or death of the immunized mice in ourpresent system. However, the reason for such deviation inmice is not known.Typhoid fever continues to be a major public health

concern in developing countries (9), and several controlledfield trials of newly developed vaccines have been made (15,28). A Vi-positive variant of S. typhi Ty2la was recentlyreported to be used for an improved live oral vaccine (8).This finding together with the present results allows us tospeculate that a conjugate of stress proteins, such as catalaseII (HP-I-type catalase) of S. typhi, which is thought to bepresent in almost all members of the family Enterobac-teriaceae (17), with a Vi antigen may serve as a componentvaccine for S. typhi infection both in inherently susceptibleand in resistant populations (12).

In general, our results suggest that stress proteins play amajor role for the induction of CMI and that H202-inducedcatalase is one such effective antigen. Although heat shockproteins such as groEL are reportedly predominant in theimmune response to infection with intracellular parasites(20, 33), these proteins possess a highly conserved naturethat may induce tolerance or an autoimmune reaction.However, HP-I-type catalase (our catalase II) is revealed tobe a very unusual catalase which shows no obvious homol-ogy to human catalase (17, 31), suggesting that HP-I-typecatalase is a candidate for a safe antigen for humans, at leastas a carrier protein for the above-mentioned conjugatevaccine.

ACKNOWLEDGMENTSWe are grateful to B. N. Ames, University of California, Berke-

ley, Calif., and T. Arai, Meiji College of Pharmacy, Tokyo, Japan,for the gift of their S. typhimurium strains. We thank M. Uchida andS. Ishibashi for technical assistance and R. Tanaka for help inpreparation of the manuscript.

REFERENCES1. Angerman, C. R., and T. K. Eisenstein. 1980. Correlation of the

duration and magnitude of protection against Salmonella infec-tion afforded by various vaccines with antibody titers. Infect.Immun. 27:435-443.

2. Beers, R. F., Jr., and I. W. Sizer. 1952. A spectrophotometricmethod for measuring breakdown of hydrogen peroxide bycatalase. J. Biol. Chem. 195:133-140.

3. Cher, D. J., and T. R. Mosmann. 1987. Two types of murine

INFECT. IMMUN.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 18

Dec

embe

r 20

21 b

y 12

5.16

3.14

3.16

5.

PROTECTION BY STRESS CATALASE OF S. 7YPHIMURIUM 1825

helper T cell clone. II. Delayed-type hypersensitivity is medi-ated by THi clones. J. Immunol. 138:3688-3694.

4. Christman, M. F., R. W. Morgan, F. S. Jacobson, and B. N.Ames. 1985. Positive control of a regulon for defenses againstoxidative stress and some heat-shock proteins in Salmonellatyphimurium. Cell 41:753-762.

5. Claiborne, A., and I. Fridovich. 1979. Purification of the o-dia-nisidine peroxidase from Escherichia coli B. J. Biol. Chem.254:4245-4252.

6. Collins, F. M. 1974. Vaccines and cell-mediated immunity.Bacteriol. Rev. 38:371-402.

7. Crowle, A. J. 1975. Delayed hypersensitivity in the mouse. Adv.Immunol. 18:197-264.

8. Cryz, S. J., Jr., E. Furer, L. S. Baron, K. F. Noon, F. A. Rubin,and D. J. Kopecko. 1989. Construction and characterization of aVi-positive variant of the Salmonella typhi live oral vaccinestrain Ty2la. Infect. Immun. 57:3863-3868.

9. Edelman, R., and M. M. Levine. 1986. Summary of an interna-tional workshop on typhoid fever. Rev. Infect. Dis. 8:329-349.

10. Eisenstein, T. K., and C. R. Angerman. 1978. Immunity toexperimental Salmonella infection: studies on the protectivecapacity and immunogenicity of lipopolysaccharide, acetone-killed cells, and ribosome-rich extracts of Salmonella typhimu-rium in C3H/HeJ and CD-1 mice. J. Immunol. 121:1010-1014.

11. Eisenstein, T. K., L. M. Killar, and B. M. Sultzer. 1984.Immunity to infection with Salmonella typhimurium: mouse-strain differences in vaccine- and serum-mediated protection. J.Infect. Dis. 150:425-435.

12. Fox, J. L. 1988. In search of a better typhoid vaccine: severalimproved vaccines are being tested; the efforts sometimesrequire a dedicated spirit of advanture and an ability to impro-vise. ASM News 54:551-554.

13. Fridovich, I. 1978. The biology of oxygen radicals. Science201:875-880.

14. Fukazawa, Y., K. Kagaya, and Y. Ishibashi. 1983. Effect ofdelayed-type hypersensitivity reaction and transferred lympho-kine on the resistance of mice to Salmonella typhimunuminfection. Infect. Immun. 39:986-989.

15. Germanier, R., and E. Furer. 1975. Isolation and characteriza-tion of galE mutant Ty2la of Salmonella typhi: a candidatestrain for a live, oral typhoid vaccine. J. Infect. Dis. 131:553-558.

16. Harris, H., and D. A. Hopkinson. 1976. Handbook of enzymeelectrophoresis in human genetics. Elsevier/North-Holland,New York.

17. Hopp, T. P., and K. R. Woods. 1981. Prediction of proteinantigenic determinants from amino acid sequences. Proc. Natl.Acad. Sci. USA 78:3824-3828.

18. Hsu, H. S. 1989. Pathogenesis and immunity in murine salmo-nellosis. Microbiol. Rev. 53:390-409.

19. Kagaya, K., K. Watanabe, and Y. Fukazawa. 1989. Capacity ofrecombinant gamma interferon to activate macrophages forSalmonella-killing activity. Infect. Immun. 57:609-615.

20. Kaufmann, S. H. E. 1990. Heat shock proteins and the immuneresponse. Immunol. Today 11:129-136.

21. Killion, J. W., and D. C. Morrison. 1988. Determinants ofimmunity to murine salmonellosis: studies involving immuniza-tion with lipopolysaccharide-lipid A-associated protein com-plexes in C3H/HeJ mice. FEMS Microbiol. Immunol. 47:41-54.

22. Kuusi, N., M. Nurminen, H. Saxen, M. Valtonen, and P. H.Makela. 1979. Immunization with major outer membrane pro-teins in experimental salmonellosis of mice. Infect. Immun.25:857-862.

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

24. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall.1951. Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193:265-275.

25. Mackaness, G. B., R. V. Blanden, and F. M. Collins. 1966.Host-parasite relations in mouse typhoid. J. Exp. Med. 124:573-583.

26. Morgan, R. W., M. F. Christman, F. S. Jacobson, G. Storz, andB. N. Ames. 1986. Hydrogen peroxide-inducible proteins inSalmonella typhimunum overlap with heat shock and otherstress proteins. Proc. Natl. Acad. Sci. USA 83:8059-8063.

27. O'Brien, A. D. 1982. Innate resistance of mice to Salmonellatyphi infection. Infect. Immun. 38:948-952.

28. Robbins, J. D., and J. B. Robbins. 1984. Re-examination of theimmunopathogenic role of the capsular polysaccharide (Vi an-tigen) of Salmonella typhi. J. Infect. Dis. 47:436-449.

29. Svenson, S. B., and A. A. Lindberg. 1981. Artificial Salmonellavaccines: Salmonella typhimurium 0-antigen-specific oligosac-charide-protein conjugates elicit protective antibodies in rabbitsand mice. Infect. Immun. 32:490-496.

30. Szu, S. C., A. L. Stone, J. D. Robbins, R. Schneerson, and J. B.Robbins. 1987. Vi capsular polysaccharide-protein conjugatesfor prevention of typhoid fever. J. Exp. Med. 166:1510-1524.

31. Triggs-Raine, B. L., B. W. Doble, M. R. Mulvey, P. A. Sorby,and P. C. Loewen. 1988. Nucleotide sequence of katG, encodingcatalase HPI of Escherichia coli. J. Bacteriol. 170:4415-4419.

32. Vogel, H. J., and D. M. Bonner. 1956. Acetylornithinase ofEscherichia coli: partial purification and some properties. J.Biol. Chem. 218:97-106.

33. Young, D. B. 1988. Stress-induced proteins and the immuneresponse to leprosy. Microbiol. Sci. 5:143-146.

VOL. 60, 1992

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 18

Dec

embe

r 20

21 b

y 12

5.16

3.14

3.16

5.