INFECTION AND IMMUNITY, Mar. 1990, p. 719-7250019-9567/90/030719-07$02.00/0Copyright C 1990, American Society for Microbiology

Synthesis of a Select Group of Proteins by Neisseria gonorrhoeae inResponse to Thermal Stress

MARION L. WOODS 11,1.2* RODERICK BONFIGLIOLI,2 ZELL A. McGEE,1 AND COSTA GEORGOPOULOS3

Center for Infectious Diseases, Diagnostic Microbiology and Immunologyl* and Department of Cellular,Viral and Molecular Biology,3 University of Utah School of Medicine, Salt Lake City, Utah 84132,

and Menzies School of Health Research, Darwin, Northern Territory, Australia 08112

Received 19 June 1989/Accepted 27 November 1989

We report the thermal conditions that induce the heat shock response in Neisseria gonorrhoeae. Underconditions of thermal stress, Neisseria gonorrhoeae synthesizes heat shock proteins (hsps), which differquantitatively from conventionally studied gonococcal proteins. Gonococci accelerate the rate of synthesis ofthe hsps as early as 5 min after the appropriate stimulus is applied, with synthesis continuing for 30 min, as

demonstrated by in vivo labeling experiments with L-[35S]methionine. Two of the gonococcal hsps areimmunologically cross-reactive with the hsps of Escherichia coli, DnaK and GroEL, as demonstrated byWestern blot (immunoblot) analysis. Ten hsps can be identified on two-dimensional autoradiograms of wholegonococci (total protein). Four hsps can be identified on two-dimensional autoradiograms of 1% N-lauroylsarcosine (sodium salt) (Sarkosyl)-insoluble membrane fractions. Two of the hsps from the 1%Sarkosyl-insoluble fraction are found exclusively in this fraction, suggesting that they are membrane proteins.The identification of this group of proteins will facilitate further study of the function of these proteins andprovide insight into the possible role of hsps in disease pathogenesis.

The heat shock response is a cellular response to stressthat is characterized by increased synthesis of certain pro-teins that are normally synthesized at lower rates duringnonstress conditions. The rate of synthesis may increase 10-to 100-fold over the basal levels of synthesis during a shortinterval, i.e., 5 to 10 min (8, 13, 20). In the course of theresponse, synthesis of other proteins may be reduced or,occasionally, stopped (8, 13, 20).The heat shock response has been found in a wide range of

eucaryotic and procaryotic cell types that have been ana-lyzed specifically for this response. The heat shock responsewas described originally in relation to thermal stress, butsubsequently some of the heat shock proteins (hsps) havebeen shown to be inducible by a wide variety of otherstresses, e.g., low pH and oxidizing agents (8, 13, 20). Thefunctions of many of these proteins are not known, but thisstress-induced response is thought to protect against adverseenvironmental conditions. For instance, heat shock re-sponse occurs in mammalian cell culture systems that havebeen exposed to certain gene products of viruses such asadenovirus and herpes simplex virus (21, 22). Some hsps aresurface exposed. A recent report indicates that a cell surfacetumor-specific transplantation antigen is a heat shock-relatedprotein (25). hsps are necessary for DNA replication of theEscherichia coli bacteriophage lambda (9). The hsp GroELof E. coli has recently been shown to be a "chaperone"protein that may facilitate membrane translocation of bothsecreted proteins, such as ,-lactamase, and outer membraneproteins, such as pro-OmpA (4, 12).The relationship of hsps of pathogens to the disease

process is less clear. A recent report has indicated that the65,000-molecular-weight protein of Mycobacterium tubercu-losis, which stimulates antibody production during naturalinfection and is associated with the induction of adjuvantarthritis in rats, is an hsp (27, 28).The parasitic protozoans Trypanosoma brucei and Leish-

* Corresponding author.

mania major, when shifted from 25°C to 37°C under in vitroconditions, which mimics thermal conditions encounteredby the parasites during transmission from invertebrate hostof mammalian host, exhibit the heat shock response duringthe differentiation process to the mammal-adapted form.These hsps may play a role in differentiation of theseparasites in vivo (26).

Theoretically, a similar response might occur in gonococciand other pathogens prior to and during environmentalstresses of breaching host defenses to establish infection tobecome facultative intracellular parasites (17, 18). Thermalstress is but one of many inducers of the heat shockresponse, which results in preferential overproduction ofthis class of proteins. In the studies reported here, thermalstress was chosen as the type of stress to induce the heatshock response in the gonococcus because it can be pre-cisely controlled and reproduced.

MATERIALS AND METHODS

Growth conditions and gonococcal strains. Transparent andopaque gonococcal strains F62, MS11, and R10 and clinicalisolates 192A1, VU1221, RDH40, RDH68 (disseminatedgonococcal infection strain, 1-lactamase positive), and RDH88 were initially grown for 16 to 22 h on clear typing medium,"GC + Iso agar" (11) (GC agar [Difco] + 2% [vol/vol]IsoVitaleX [BBL Microbiology Systems]), at 36°C with 3%CO2 in a highly humidified atmosphere containing air. Col-onies were then inoculated into a chemically defined minimalmedium described previously (11), except that it lackedmethionine and was supplemented with a 50 ,uM concentra-tion of each of the amino acids except methionine. Thisminimal medium, after inoculation with gonococci, wasincubated for 6 to 8 h at 35°C with shaking at 180 rpm in air.Gonococci were used in subsequent experiments when in logphase (ca. 5 x 107 bacteria per ml), as determined by colonycount. Gram staining was used to test the purity of cultures.Growth conditions and E. coli strains. E. coli strains

JM103, DH1, HB101, and Q358 were initially grown on LB

719

Vol. 58, No. 3

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 20

Dec

embe

r 20

21 b

y 46

.56.

71.1

29.

720 WOODS ET AL.

agar (15). Gonococcal minimal medium lacking methioninewas used as the liquid growth medium. Log-phase organismsgrown at 35°C with shaking in air were used in subsequentexperiments.

Induction of heat shock response. A range of temperatures,i.e., 42 to 46°C, was initially selected to establish theoptimum temperature for induction of the heat shock re-sponse in gonococci. Control cultures were radioactivelylabeled at 35°C. Thereafter, in individual experiments achosen water bath temperature (44.5°C) was maintainedduring the period of induction and labeling of gonococcalproteins with radioactive methionine as described below.The criterion for optimum response to thermal induction wasincreased production of selected proteins, with decreasedsynthesis of some other gonococcal proteins. An internalreference protein whose level of synthesis was not apprecia-bly altered at the temperatures used served as a control forcomparing rates of synthesis under heat shock and non-heatshock conditions. The difference in the levels of proteinsynthesis was readily apparent upon comparison of autora-diograms of control (35°C) and test (42 to 46°C) lanescontaining equal amounts of gonococcal proteins separatedby sodium dodecyl sulfate-polyacrylamide gel electrophore-sis (SDS-PAGE) and subjected to autoradiography. TheLowry method was used to monitor protein concentrations(14).

E. coli strains were induced to produce hsps at 42.5°C.Controls were labeled at 35°C.

Radioactive labeling of gonococcal proteins and E. coliproteins. New protein synthesis in the gonococcus at 35°Cand during thermal stress was examined by adding L-[35S]methionine (Amersham) at 10 ,uCi/ml of methionine-freeminimal medium (10) during specified time intervals afterexposure to thermal stress. Radioactive labeling of controlspecimens was done at 35°C but in an otherwise identicalfashion during identical time intervals. Unless otherwisestated, gonococcal proteins and E. coli proteins were labeledwith radioactive methionine for 10 min during the timeinterval 5 through 15 min after the imposition of thermalstress. This time interval was chosen because of the knowntime course of the heat shock response with E. coli (20) andwas subsequently established for gonococci. Radioactivelabeling of proteins was stopped by the addition of a>10,000-fold excess of nonradioactive methionine to theminimal medium cultures, otherwise free of nonradioactivemethionine, followed by centrifugation in a microcentrifuge(Eppendorf) at 12,500 x g for 2 min. The supernatant wasremoved, and the gonococci were suspended in an appropri-ate buffer for subsequent steps. Samples of supernatantswere precipitated with trichloroacetic acid (10% [wt/vol]final concentration) and saved.SDS-PAGE and isoelectric focusing. After the cells were

pelleted, they were lysed in SDS sample buffer (10) (10%[vol/vol] glycerol, 5% 2-mercaptoethanol, 3% SDS, 0.0625 MTris chloride [pH 6.8]) and frozen at -20°C until SDS-PAGEwas performed. All samples were boiled for 5 min prior toSDS-PAGE. Gonococcal proteins that were to be resolvedby two-dimensional (2-D) gel electrophoresis (isoelectricfocusing followed by SDS-PAGE) were lysed in 2-D lysisbuffer (23) (5.7 g of urea, 2 ml of 10% Nonidet P-40 [Shell],0.5 ml of 2-mercaptoethanol, 0.4 ml of ampholines [pHrange, 5 to 7] [Pharmacia, Inc., Piscataway, N.J.], 0.1 ml ofampholines [pH range, 3 to 10] [Pharmacia], 1.7 ml ofdistilled H20) and frozen at -20°C until processed forisoelectric focusing. SDS-PAGE (12.5% polyacrylamide)was carried out by the method of Laemmli (10). Isoelectric

focusing was done by the method of O'Farrell (23) for aduration of 8,000 V h. The gels were dried in a gel-dryingapparatus (Bio-Rad Laboratories, Richmond, Calif.) at 60°Cin a vacuum for 2 h. The gels were exposed to Kodak XAR-5film at room temperature for 16 h to 96 h. Protein molecularweight markers were from Sigma Chemical Company, St.Louis, Mo., and Amersham Corp., Arlington Heights, Ill.(Rainbow Markers).

Radioactive pulse-labeling of proteins. In order to establishthe time course of the heat shock response in relation tothermal stress, identical sets of 1-ml cultures of log-phasegonococci in minimal medium lacking methionine wereprocessed in the following fashion. One set was maintainedat 35°C during the experiment. The second set was shifted to44.5°C at time zero. One-milliliter portions were radioac-tively labeled for 5 min each with radioactive L-[35S]methionine (50 ,uCi) at 5-min intervals over a timecourse of 50 min. Radioactive labeling of gonococcal pro-teins was stopped by adding 10 volumes of ice-cold acetoneand by placing the portions on ice. This method of stoppingthe radioactive labeling of gonococcal proteins was chosenbecause it instantaneously halts protein synthesis, insolubi-lizes proteins, and is technically easy to perform whenprocessing samples in rapid succession. The samples werecentrifuged at 4,000 x g for 10 min, and the supernatant wasdiscarded. The pellet was subsequently washed twice inacetone. Residual acetone was removed by vacuum evapo-ration, and samples were dissolved in SDS sample buffer andstored at -20°C as described above until examined by SDS-PAGE.

Preparation of Sarkosyl-insoluble membrane fractions.Outer membranes of gonococci were prepared by Sarkosylextraction in the following manner (7, 9). Log-phase gono-cocci in gonococcal minimal medium (125 ml) were radiola-beled with L-[35S]methionine for 10 min at 35 or 44.5°C asdescribed above, followed by the addition of excess unla-beled L-methionine. The gonococci were centrifuged at12,000 x g for 10 min at 4°C. The pellets were thensuspended in 10 ml of 10 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) buffer (pH 7.4) contain-ing 0.1% (vol/vol) protease inhibitor (10 mM phenylmethyl-sulfonyl fluoride [PMSF] in isopropanol). The cells weredisrupted by vortexing with 150-,um-diameter glass beads for2 min on a Vortex genie at the maximum setting. Subse-quently, sonication with a Branson sonifier (setting, 3; dutycycle, 50%; duration, 2 min, on ice) was used to furtherdisrupt the cells, followed by incubation with DNase I (40,ug/ml)-RNase (40 ,ug/ml) in 1 mM MgCl2 at 37°C for 15 min.Centrifugation at 3,000 x g for 10 min at 4°C was performedto remove cell debris, and the supernatant was subjected tocentrifugation (20 min at 100,000 x g, 4°C) to pellet themembrane fraction. The pellet was resuspended in HEPESbuffer-PMSF and subjected to centrifugation (60 min at100,000 x g, 40C). The remaining membrane pellet wasresuspended in HEPES buffer-PMSF to a concentration of1% (wt/vol) N-lauroylsarcosine (Sarkosyl) for 10 min atroom temperature. The Sarkosyl-insoluble fraction was sed-imented at 100,000 x g for 60 min at 40C and washed inHEPES buffer-PMSF. A final centrifugation was performed(100,000 x g for 60 min at 40C). The membrane pellet wasresuspended in HEPES buffer-PMSF and stored at -200Cuntil used. Aliquots of whole, radioactively labeled gono-cocci were saved and frozen prior to membrane preparationsfor subsequent comparison with membrane fractions. Gono-coccal protein concentrations were determined by the

INFECT. IMMUN.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 20

Dec

embe

r 20

21 b

y 46

.56.

71.1

29.

HEAT SHOCK PROTEINS OF GONOCOCCI 721

method of Lowry et al. (14). The method of Bradford (2) wasused to determine the protein concentration of E. coli DH1.

Lactate dehydrogenase analysis was used to assay thepurity of Sarkosyl membrane preparations (3).

Preparation of E. coli proteins and rabbit polyclonal anti-serum. The DnaK and GroEL proteins of E. coli, withmolecular weights of 70,000 and 66,000, respectively, havepreviously been isolated and purified, and polyclonal rabbitantisera have been raised against them (6, 29). Serum wasobtained from rabbits prior to immunization for subsequenttesting in Western blots (immunoblots) as a negative control.

Immunoblotting. Western blots of gonococcal antigenswere prepared by standard techniques (5).

Purification of proteins following 2-D PAGE. For the pur-pose of immunological identification of individual gonococ-cal polypeptides from 2-D polyacrylamide gels, the followingmethod was used. Purification of selected (nonradioactive)gonococcal proteins from 2-D gels was accomplished byexcising the appropriate protein spots from Coomassie blue-stained gels with a Pasteur pipette. The protein of interestwas contained in a 1-mm3 polyacrylamide plug. The proteinswere removed from these plugs by direct insertion of plugsinto individual wells of SDS-polyacrylamide gels. SDS sam-ple buffer was added to the well prior to electrophoresis, andproteins were directly electroeluted from the polyacrylamideplugs into the SDS-polyacrylamide gel.

Immunoprecipitation. Equal amounts of radioactively la-beled protein from heat-shocked and non-heat-shockedgonococcal (RDH68) and E. coli (DH1) proteins were solu-bilized in 0.5% SDS-0.5% 2-mercaptoethanol at 95°C for 5min (1). These extracts were centrifuged for 60 min at 18,000rpm in a Beckman JA-20 rotor at 4°C, and the supernatantswere diluted twofold and adjusted to 1% Triton X-100-1%deoxycholate-100 mM NaCl. Samples were immunoprecip-itated with a 1/10 volume of anti-DnaK serum for 90 min onice. The immunoglobulin-antigen complexes were precip-itated with a 10% suspension of Formalin-fixed Staphylococ-cus aureus (Cowan strain I; Sigma Chemical Co.) andpelleted at 10,000 rpm for 1 min in a JA-20 rotor. The pelletswere washed twice in 50 mM Tris hydrochloride (pH 7.4)-150 mM NaCl-1% sodium deoxycholate-1% Triton X-100-0.1% SDS and heated in SDS buffer at 95°C for 2 min.Following centrifugation at 10,000 rpm in a Beckman JA-20rotor, the supernatants were electrophoresed on SDS-poly-acrylamide gels. Gels were dried and exposed to KodakXAR-5 film as previously described.

RESULTS

Optimal temperature for hsp induction. The optimal tem-perature of the water bath for the induction of the heat shockresponse for the Neisseria gonorrhoeae clinical isolatesRDH40, RDH68, RDH88, and R-10 was found to be 44.5°C.The actual temperature of the culture was 43.5°C. Othergonococcal strains were optimally induced with water bathtemperatures between 44.0 and 44.5°C. It should be notedthat the thermal conditions that are optimal for induction ofthe heat shock response may vary slightly with each strainand have to be determined individually, although the differ-ences appear to be slight. The data presented are mainlythose obtained with strains RDH40 (urethral isolate) andRDH68 (disseminated gonococcal infection isolate from ajoint) but are representative of results obtained with similarexperiments performed with gonococcal strains R10, F62,and MS11 and clinical isolates 192A1, VU1221, and RDH88.The quantitative differences in new protein synthesis be-

1A B

a

b I*b.--WNW 1

c

*

d MF7

2A B

;4

43

s-g 411*~ ..*,

p' S -Im;

_

9

FIG. 1. (Panel 1) Autoradiogram of equal amounts of gonococcalproteins (50 ,ug) in lanes A and B from gonococcal strain RDH40.Lane A, Total protein labeled with radioactive methionine at 35°C;lane B, total protein labeled with radioactive methionine at 44.5°C.(Panel 2) Lane A', 50 ,ug of 1% Sarkosyl-insoluble extracts, labeledat 35°C; lane B', 50 ,ug of 1% Sarkosyl-insoluble extracts, labeled at44.5°C. 4, Proteins synthesized at rates increased over basal levels.Protein molecular weight markers (14C-labeled Rainbow Markers;Amersham) are as follows: a, 200,000; b, 92,500; c, 69,000; d,46,000; e, 30,000; f, 22,500; and g, 14,300.

tween thermally stressed and nonstressed gonococci arevisually apparent when autoradiograms of SDS-polyacryl-amide gels are compared. Supernatants of centrifuged wholegonococci were precipitated with trichloroacetic acid, andno radioactive proteins were demonstrated. Figure 1 is anautoradiogram of an SDS-PAGE gel of gonococcal strainRDH40. Each lane contains 50 ,g of protein, as determinedby the Lowry method. Samples of total protein (panel 1) arein lanes A (35°C) and B (44.5°C). Samples of the 1%Sarkosyl-insoluble fractions (panel 2) are in lanes A' (35°C)and B' (44.5°C). Consistent findings among gonococcalstrains tested, when total protein is considered, are thepresence of increased levels of new protein synthesis of thehsps designated gonococcal hsps 1, 2, 3, and 4, with virtuallyidentical molecular weights. Additionally, 2-D PAGE auto-radiograms of the gonococcal strains are also virtually iden-tical. These four proteins were confirmed on subsequent 2-Dgels (Fig. 2). The 1% Sarkosyl-insoluble preparations ofouter membrane proteins in Fig. 1 (panel 2) demonstrate thepresence of three hsps, namely, gonococcal hsps 3, 4, and11, which can be identified on subsequent 2-D gels (Fig. 2and 3). Consistently found in the heat-shocked lanes of 1%Sarkosyl-insoluble preparations are the two doublets shownin lane B' at molecular weights of 92,000 and 15,000.However, because we are not yet certain of their behavior in

VOL. 58, 1990

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 20

Dec

embe

r 20

21 b

y 46

.56.

71.1

29.

722 WOODS ET AL.

AA

. . .-

-_w

4

7C

_

10 98a

4t -*WZ 3

*,. 6

a

b

a

b

c

11

12 _

e

g

B

_ -3

11 -

12 --v _-O_ 0 .'? - 4 ,,,~~~~'II

I

d* 0

e-4a

,

6 4

7...

*,_ 1A_:.9

FIG. 2. (A) Autoradiogram of 2-D PAGE of 100 ,ug of totalgonococcal proteins from strain RDH40 at 35°C as described inMaterials and Methods. V, hsps synthesized at the basal levels ofprotein synthesis. Molecular size markers a to g are as described inthe legend to Fig. 1. From left to right the pH gradient changes frombasic to acidic for this and subsequent 2-D gels. The protein labeledr is an internal reference protein. (B) Autoradiogram of 2-D PAGEof 100 jig of total gonococcal proteins from strain RDH40, labeledbetween 5 and 15 min following the shift to 44.5°C. V, hsps showingan accelerated rate of protein synthesis during this time interval.

2-D gels (they could represent composite images of severalproteins of nearly identical mobility), we have not assignednumbers to them.

Figure 2, a 2-D autoradiogram of total protein of non-heat-shocked gonococci of strain RDH40 (panel A) andheat-shocked gonococci (panel B), demonstrates the pres-ence of additional hsps that cannot be consistently resolvedby SDS-PAGE in one dimension only. These hsps are

designated gonococcal hsps 1 to 10. The protein designatedas "r" is used as an internal reference comparison control.This protein is found in both total-protein and 1% Sarkosyl-insoluble preparations, and its rate of synthesis is notsignificantly altered at the temperatures used in these exper-iments.

Figure 3, a 2-D autoradiogram of 1% Sarkosyl-insolubleproteins of non-heat-shocked gonococci of strain RDH40(panel A) and heat-shocked gonococci (panel B), demon-strates the consistent presence of four hsps, two of which,i.e., gonococcal hsps 3 and 4, are found in the 2-D gel

FIG. 3. (A) Autoradiogram of 2-D PAGE of 100 tg of 1%Sarkosyl-insoluble outer membrane fraction of gonococcal strainRDH40, at 35°C and labeled as described in Materials and Methods.V, hsps. Molecular size markers a to g are as described in the legendto Fig. 1. (B) Autoradiogram of 2-D PAGE of 100 p.g of 1%Sarkosyl-insoluble outer membrane fraction of RDH40 labeled at44.5°C between 5 and 15 min following the shift to 44.5°C. V, hsps.

autoradiograms of extracts shown in Fig. 2. In addition, twonew hsps are visible in this preparation (Fig. 3), designatedgonococcal hsps 11 and 12, whose presence is not clearlyseen in Fig. 2. Of these four hsps, gonococcal hsp 3 is not as

enriched in amount as the other 1% Sarkosyl-insoluble hspsappear to be. A possible explanation for this will follow inthe Discussion.

In an effort to identify as many gonococcal hsps as

possible, we compared hsps of E. coli with those synthesizedby the gonococcus by using identical gel conditions foi -irstand second dimensions (16). The 2-D mobilities of the -olihsps DnaK, GroEL, GrpE, and GroES were found '!uo bealmost identical to those of the gonococcal hsps 2, 3, 5. mnd6, respectively. This finding suggests that these four protQb:insmay have been conserved between species. To be absolutelycertain of the correspondence of these gonococcal proteinsto those of E. coli, one cannot simply rely on their relativemobilities in one-dimensional and 2-D gels. One availablemethod is that of Western blotting analysis. Such evidencefor the relatedness of two of these proteins follows.

Antigenic relatedness of hsps in different microorganisms.Because hsps are widely conserved in nature, we chose toexamine the antigenic cross-reactivity between two hsps ofE. coli, DnaK and GroEL. Western blot analysis of the

B10 9 8

-f41

943e0. _

* 3

t9--Jw'

r

a

b

b

.

INFECT. IMMUN.

.- j,34w fl.

-INMIP %P

4 OW.

Ir -Alp

p

4mp.

1

0,.-I

r4

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 20

Dec

embe

r 20

21 b

y 46

.56.

71.1

29.

VOL.58,1990HEATSHOCK PROTEINS OF GONOCOCCI 723

A B

a b c d e f 9 a b c d e f g

470K4GroEL70K * _ _

__

FIG. 4. (A) Western blot using rabbit antisera against the DnaKprotein of E. coli, detected with 1251I-labeled staphylococcal proteinA. Extracts from E. coli strains HB101 and Q358 are shown in lanesa and b, respectively, and extracts from N. gonorrhoeae strainsR10, MS11, F62, 192A1, and VU1221 are shown in lanes c to g,respectively. All bacterial extracts shown in these two panels werefrom cultures grown continuously at 35°C. (B) Western blot ofprotein samples corresponding to those used in panel A, butproduced by using rabbit antisera against the GroEL protein of E.coli, detected with 125I-labeled staphylococcal protein A.

proteins from E. coli and from several gonococcal strains(Fig. 4) was performed with immune polyclonal rabbit seraraised against purified DnaK (panel A) and GroEL (panel B)proteins of E. coli (6, 29). All strains used in Fig. 4 weregrown at 35°C and were not heat shocked. Preimmune rabbitantisera were not reactive with the E. coli and gonococcalstrains tested. Western blot analysis revealed that the anti-DnaK antibody reacts with the 70,000-molecular-weightprotein of E. coli and with proteins having the same molec-ular weight from each of the gonococcal strains tested (R10,MS11, F62, 192A1, and VU1221) (panel A). The additionalbands in lane A (panel A) probably represent breakdownproducts of the DnaK protein of E. coli HB101. This wasparticularly pronounced as the sample aged after severalfreeze-thaw cycles. Freshly prepared samples showed asingle antigenically reactive protein at a mobility of 70,000molecular weight. Figure 4B, which uses the same strains asin Fig. 4A, shows a similar degree of antigenic cross-reactivity with GroEl antibody with a protein of approxi-mately 66,000 molecular weight.

Immunoprecipitation with anti-DnaK antibody demon-strated that the antibody was directed against the hsps (Fig.5). The DnaK antibody precipitates a gonococcal proteinfrom lane H (heat shock extract) that has been synthesized ata rate increased over that of basal synthesis (lane G). Theimmunoprecipitation reactions of E. coli DnaK protein un-der heat shock conditions (lane J) and basal conditions (laneI) are shown as controls. No significant immunoprecipitationwas apparent for the 0.2% Sarkosyl-insoluble preparationused for this experiment.Gonococcal hsp 1 was shown to be the protein immuno-

reactive with antibody to the DnaK protein of E. coli whenprotein spots from non-radioactively labeled gonococcalproteins were separated by 2-D PAGE. Individual proteinspots were removed from 2-D gels as described above.Western blots using antibody to DnaK confirmed that gono-

S A B C D E FG H I Ja'_bC- 4fif i4XXX4

e

f

FIG. 5. Autoradiogram of immunoprecipitation using DnaKantisera. Lanes A (350C) and B (44.50C) contain 10 ~±g of RDH68total protein. Lanes C (350C) and D (42.50C) contain 5 jig of E. ccliDH1 total protein. Lanes E (350C) and F (44.50C) contain theimmune precipitate from 50 jig of starting material from a 0.2%Sarkosyl-insoluble fraction of an RDH68 extract. Lanes G (350C)and H (44.50C) contain the immune precipitate from 50 jig of startingmaterial from RDH68 whole-cell extract. Lanes I (350C) and J(42.50C) contain the immune precipitate from 10 p.g of startingmaterial from E. coli DH1 whole-cell extract.

coccal hsp 1 was the immunoreactive protein shown in Fig.4A (data not shown). In an identical fashion, gonococcal hsp3 was shown to be the GroEL equivalent.Time course of the stress-induced protein response. The

results of a typical time course experiment indicated thatsynthesis of gonococcal hsps occurs as early as 5 minfollowing thermal induction and continues for up to 30 minpostinduction. Few newly radioactively labeled proteinswere apparent from 35 to 65 mmn post-thermal induction at44.50C (data not shown).

DISCUSSION

To develop immunologic and other means of interveningin infectious diseases, many investigators have examined themolecular structure of pathogenic microorganisms in anattempt to determine which molecules are associated withvirulence. These types of investigations may provide infor-mation that will be useful in identifying as vaccine candi-dates molecules used by pathogenic microorganisms tocause disease. Most previous laboratory investigations ofgonococci have been done under nonstress growth condi-tions at 370C by using enriched media. Two notable excep-tions to this mode have been reported. When gon'ococci aregrown under conditions of low iron concentrations, produc-tion of membrane-associated iron-binding proteins is in-duced (19). These proteins appear to elicit an antibody

VOL. 58, 1990

."AL, M.

4 m* W

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 20

Dec

embe

r 20

21 b

y 46

.56.

71.1

29.

724 WOODS ET AL.

response during natural infection but do not appear to bepresent when gonococci are grown on enriched iron-supple-mented medium, which is typically used for the growth ofgonococci in vitro. When gonococci are grown under anaer-obic conditions, at least three proteins, Pan 1 to Pan 3, areinduced (7).As demonstrated here, stress of a physical (thermal)

nature induces increased synthesis of hsps in gonococci.Four of these proteins appear to be associated with the 1%Sarkosyl-insoluble membrane fraction, as judged by 2-DPAGE. hsp 3 is not recovered in proportional amounts in the1% Sarkosyl-insoluble preparations, which typically in-crease representation of other known outer membrane pro-teins such as gonococcal protein 1 (data not shown). Recentreports have demonstrated that GroEL, a cytoplasmic hsp,is a chaperone protein that maintains precursor proteins in acompetent conformation for translocation across mem-branes either to be assembled in the outer membrane(OmpA) or to be secreted (P-lactamase) (4, 12). The pres-ence of gonococcal hsp 3 (the GroEL equivalent), which ispresumably a cytoplasmic protein, in the 1% Sarkosyl-insoluble fraction may be the result of its association withproteins transported across the cytoplasmic membrane.That four of these hsps are found in the 1% Sarkosyl-

insoluble fraction indicates that they have the potential tointeract with genital mucosal cell surfaces and with theimmune system. We are currently investigating this possi-bility. The induction of synthesis of this class of proteins inresponse to thermal stress is very rapid, which could beadvantageous to a mucosal pathogen if stresses are sufficientto induce this response in vivo. Gonococcal hsps have notbeen thoroughly characterized, and their function in thegonococcus or role, if any, in disease pathogenesis had notbeen elucidated.

While the thermal conditions (44.0 to 44.5°C) which weredetermined to induce the synthesis of these stress-inducedproteins in gonococci are not particularly clinically relevant,many other stresses of a physical and chemical nature(viruses, oxidizing agents, and reducing agents) have beenshown to induce the same type of protein synthesis responsein a diverse group of cell types (8, 13, 20-22). It is possiblethat the types of stresses that the gonococcus survives onmucosal surfaces, i.e., the contents of polymorphonuclearcells and the acidic vaginal pH, could theoretically induceproduction of a class of proteins that protects it from furtherstress and enhances its survival as a facultative intracellularparasite. That these hsps are important to the survival ofgonococci and other microorganisms is suggested by theirhaving been highly conserved in evolution. The observationthat some hsps appear to have common antigenic determi-nants among bacterial species and strains makes their con-sideration as vaccine candidates of particular interest (28).This study should provide a basis for further investigationinto the physiologic and potentially pathogenic role of gono-coccal hsps during experimental infection in organ cultureand cell culture systems and in natural infection.

ACKNOWLEDGMENTS

M.L.W. was supported in part by training grant 5-T32-GMO7825-04 from the Department of Health and Human Services. Thiswork was funded by Public Health Service grants AI-26285 andAI-21029 from the National Institute of Allergy and InfectiousDiseases and by the Menzies School of Health Research.We thank Laurie Rich for excellent technical assistance. We also

thank Kay Withnall, Julie Gaggin, and Barry Way (Royal DarwinHospital) for help in obtaining clinical specimens.

LITERATURE CITED1. Bardwell, J. C. A., K. Tilly, E. Craig, J. King, M. Zylicz, and C.

Georgopoulos. 1986. The nucleotide sequence of the Escherichiacoli, K12 dnaJ+ gene. J. Biol. Chem. 261:1782-1785.

2. Bradford, M. M. 1976. A rapid and sensitive method forquantitation of microgram quantities of protein using the prin-ciple of protein-dye binding. Anal. Biochem. 72:248-254.

3. Bergmeyer, H. U., E. Bernt, and B. Hess. 1965. Lactate dehy-drogenase, p. 736-746. In H. U. Bergmeyer (ed.), Methods ofenzymatic analysis. Academic Press, Inc., New York.

4. BochKareva, E. S., N. M. Lissom, and A. S. Girshovich. 1988.Transient association of newly synthesized unfolded proteinswith the heat-shock protein. Nature (London) 336:254-257.

5. Burnette, W. N. 1981. "Western blotting": electrophoretictransfer of proteins from sodium dodecyl sulfate-polyacryl-amide gels to unmodified nitrocellulose and radiographic detec-tion with antibody and radioiodinated protein A. Anal. Bio-chem. 112:195-203.

6. Chandrarekhar, G. N., K. Tilly, C. Woolford, R. Hendrix, andC. Georgopoulos. 1986. Purification and properties of the GroESmorphogenetic protein of Escherichia coli. J. Biol. Chem.261:12414-12419.

7. Clark, V. L., L. A. Campbell, D. A. Palermo, T. M. Evans, andK. W. Klimpel. 1987. Induction and repression of outer mem-brane proteins by anaerobic growth of Neisseria gonorrhoeae.Infect. Immun. 55:1359-1364.

8. Craig, E. A. 1985. The heat shock response. Crit. Rev. Bio-chem. 18:239-280.

9. Georgopoulos, C. P. 1977. A new bacterial gene (groPC) whichaffects lambda DNA replication. Mol. Gen. Genet. 151:35-39.

10. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)227:680-685.

11. La Scolea, L. J., Jr., and F. E. Young. 1974. Development of adefined minimal medium for the growth of Neisseria gonor-rhoeae. Appl. Microbiol. 28:70-76.

12. Lecker, S., R. Lill, T. Ziegelhoffer, C. Georgopoulos, P. J.Bassford, Jr., C. A. Kumamoto, and W. Wickner. 1989. Threepure chaperone proteins of Escherichia coli-Sec B, triggerfactor and GroEl-form soluble complexes with precursor pro-teins in vitro. EMBO J. 8:2703-2709.

13. Lindquist, S. 1986. The heat-shock response. Annu. Rev. Bio-chem. 5:1151-1191.

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

15. Maniatis, T., E. I. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

16. Matthews, R. G., and F. C. Neidhardt. 1988. Abnormal induc-tion of heat shock proteins in an Escherichia coli mutantdeficient in adenosylmethionine synthetase activity. J. Bacte-riol. 170:1582-1588.

17. McGee, Z. A., G. L. Gorby, L. R. Barley, C. Barlow, and C. M.Clemens. 1988. Parasite-directed endocytosis, p. 245-252. In M.Horowitz and M. Lovett (ed.), Bacteria-host cell interaction.Alan R. Liss, Inc., New York.

18. McGee, Z. A., A. P. Johnson, and D. Taylor-Robinson. 1981.Pathogenic mechanisms of Neisseria gonorrhoeae: observa-tions on damage to human fallopian tubes in organ culture bygonococci of colony type 1 or type 4. J. Infect. Dis. 143:413-422.

19. Mietzner, T. A., G. H. Luginbuhl, E. C. Sandstrom, and S. A.Morse. 1984. Identification of an iron-regulated 37,000-daltonprotein in the cell envelope of Neisseria gonorrhoeae. Infect.Immun. 45:410-416.

20. Neidhardt, F. C., and R. A. VanBogelen. 1987. Heat shockresponse, p. 1334-1345. In F. C. Neidhardt, J. L. Ingraham,K. B. Low, B. Magasanik, M. Schaecter, and H. E. Umbarger(ed.), Escherichia coli and Salmonella typhimurium: cellularand molecular biology. American Society for Microbiology,Washington, D.C.

21. Nevins, J. R. 1982. Induction of the synthesis of a 70,000 dalton

INFECT. IMMUN.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 20

Dec

embe

r 20

21 b

y 46

.56.

71.1

29.

HEAT SHOCK PROTEINS OF GONOCOCCI

mammalian heat shock protein by the adenovirus ElA gene

product. Cell 29:913-919.22. Notarianni, F. L., and C. M. Preston. 1982. Activation of

cellular stress protein genes by herpes simplex virus tempera-ture-sensitive mutants which overproduce immediate earlypolypeptides. Virology 123:113-122.

23. O'Farrell, P. H. 1975. High resolution two-dimensional electro-phoresis of proteins. J. Biol. Chem. 250:4007-4021.

24. Phillips, T. A., V. Vaughn, P. L. Bloch, and F. C. Neidhardt.1987. Gene-protein index of Escherichia coli K-12, edition 2, p.919-966. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B.Magasanik, M. Schaecter, and H. E. Umbarger (ed.), Esche-richia coli and Salmonella typhimurium: cellular and molecularbiology. American Society for Microbiology, Washington, D.C.

25. Ullrich, S. J., E. A. Robinson, L. W. Law, M. Willingham, andE. A. Apella. 1986. A mouse transplantation antigen is a heat

shock-related protein. Proc. Natl. Acad. Sci. USA 83:3121-3125.

26. Van der Ploeg, L. H. T., S. H. Giannini, and C. R. Cantor. 1985.Heat shock genes: regulatory role for differentiation in parasiticprotozoa. Science 228:1443-1446.

27. Van Eden, W., J. E. R. Thole, R. Vander Zee, A. NoordziJ,J. D. H. Van Embden, E. J. Hansen, and I. R. Cohen. 1988.Cloning of the mycobacterial epitope recognized by T-lympho-cytes in adjuvant arthritis. Nature (London) 331:171-173.

28. Young, R. A., and T. J. Elliott. 1989. Stress proteins, infectionand immune surveillance. Cell 59:5-8.

29. Zylicz, M., J. H. LeBowitz, R. McMacken, and C. Georgopoulos.1983. The dnaK protein of Escherichia coli possesses an AT-Pase and autophosphorylating activity and is essential in an invitro DNA replication system. Proc. Natl. Acad. Sci. USA80:6431-6435.

VOL. 58, 1990 725

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/i

ai o

n 20

Dec

embe

r 20

21 b

y 46

.56.

71.1

29.