INFECTION

AND

IMMUNITY, Jan. 1991, p. 240-246

Vol. 59, No. 1

0019-9567/91/010240-07$02.00/0 Copyright C) 1991, American Society for Microbiology

Legionella pneumophila Inhibits Protein Synthesis in Chinese Hamster Ovary Cells KEVIN T. McCUSKER,lt* BRUCE A. BRAATEN,2 MICHAEL W. CHO,3 AND DAVID A. LOW2 Pulmonary Division, Department of Medicine, Veterans Administration Hospital,' and Division of Cell Biology and Immunology, Department of Pathology,2 and Department of Cellular, Viral and Molecular Biology,3 University of Utah Medical Center, Salt Lake City, Utah 84132 Received 18 July 1990/Accepted 30 October 1990

Legionella pneumophila is a gram-negative facultative intracellular parasite that causes Legionnaires disease. To explore the interactions between L. pneumophila and host cells, we have developed a continuous cell line model of infection. We show that about 80% of Chinese hamster ovary (CHO) cells were associated with L. pneumophila after incubation for 3 h at a multiplicity of infection of 20 bacteria per cell. Within 3 to 4 h of incubation with L. pneumophila, protein synthesis of CHO cells was markedly inhibited, as shown by the reduction of incorporation of radiolabeled amino acids into proteins. L. pneumophila did not inhibit transport of amino acids or cause degradation of newly synthesized proteins in CHO cells. Cytochalasin D blocked internalization of L. pneumophila by CHO cells, yet CHO cell protein synthesis was inhibited. These results indicated that L. pneumophila could inhibit host protein synthesis from the cell exterior. L. pneumophila that had been killed with antibiotics prior to incubation with CHO cells still inhibited protein synthesis, indicating that the inhibition of CHO cell protein synthesis occurred in the absence of de novo protein synthesis by L. pneumophila.

Legionella pneumophila is the gram-negative bacterial pathogen which causes legionellosis in humans (36). L. pneumophila is a facultative intracellular parasite of monocytic phagocytic cells (14). It has been postulated that the survival of L. pneumophila within phagocytic cells may be the result of a bacterial product which inhibits phagolysosomal fusion (14, 15). In addition, the multiple systemic symptoms associated with legionellosis have prompted speculation that L. pneumophila may produce toxins. Friedman and coworkers (10) demonstrated that a 1.2-kDa peptide from Legionella culture supernatant was cytotoxic for Chinese hamster ovary (CHO) cells, as evidenced by a pH change of the medium after prolonged incubation with fractions containing the peptide. In later reports, this peptide from Legionella culture supernates was shown to inhibit respiratory burst in human neutrophils (11). Although L. pneumophila appears to release a small toxic peptide, it is possible that this pathogen expresses additional toxic molecules. To explore this possibility, we have developed a continuous cell line model of Legionella infection that allows synchronous infection of a high percentage of CHO cells at a low multiplicity of infection (MOI). We have utilized this model to show that L. pneumophila inhibits protein synthesis in CHO cells.

extract agar (BCYE) plates (8). L. pneumophila retains

virulence for many passages on BCYE (16). A virulent stock of L. pneumophila was maintained by suspending bacteria in yeast extract broth containing 40% glycerol and freezing 0.10-ml aliquots at -70°C. Purity of Legionella cultures was intermittently assessed by demonstrating that the bacteria grew on BCYE but not on Luria-Bertani plates (22) and by immunostaining.

Continuous cell lines. Wild-type CHO cells (WTB CHO) gift from April Robbins (Genetics and Biochemistry Branch, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Md.), HeLa cells were a gift from Jerry Kaplan (University of Utah), and Vero cells were a gift from Spotswood Spruance (University of Utah). Chinese hamster ovary cells which were auxotrophic mutants for glycine, adenosine, and thymidine were a gift from Stuart Arfin (University of California, Irvine). Vero and HeLa cells were cultured in Dulbecco's modified Eagle's medium with 5% calf serum. CHO cell strains were incubated in medium A, which is minimal essential medium (Auto Mod; Sigma Chemical Co., St. Louis, Mo.) supplemented with proline, glycine, glutamine, adenosine, thymidine, and 5% heatinactivated fetal calf serum (HyClone Laboratories, Inc., Logan, Utah). Experimental infections of continuous cell lines. CHO, Vero, and HeLa cells were seeded at 105 cells per well into 24-well tissue culture plates (Becton Dickinson Labware, Lincoln Park, N.J.) for 36 h. Cells were incubated overnight in antibiotic-free medium prior to exposure to the bacteria. L. pneumophila was suspended in phosphate-buffered saline (PBS) at a density of 109 bacteria per ml and added to monolayers in antibiotic-free tissue culture medium to achieve an MOI of 20:1. After incubation in antibiotic-free medium for 3 h, cell supernatants were removed and the monolayers were washed three times with 1 ml of sterile PBS to remove unattached bacteria. After further incubation in were a

MATERIALS AND METHODS

Bacterial strains and media. L. pneumophila serogroup 1 Philadelphia strain 1 was obtained from the Centers for Disease Control and serially passaged in embryonated hen eggs until a 50% lethal dose for eggs of approximately 10 bacteria was attained (6). L. pneumophila were inoculated from egg yolk suspensions onto buffered charcoal-yeast * Corresponding author. t Present address: Pulmonary Division, University of Utah Medical Center, 50 N. Medical Drive, Salt Lake City, UT 84132.

240

VOL. 59,

1991

L. PNEUMOPHILA INHIBITS CHO CELL PROTEIN SYNTHESIS

tissue culture medium, the percentage of cells infected was determined by immunostaining as described below. Growth curves of L. pneumophila in CHO cells. CHO cells were seeded in 35-mm tissue culture dishes (Becton Dickinson Labware). Monolayers were incubated with L. pneumophila at an MOI of 2:1 for 3 h and then washed and maintained in antibiotic-free medium A. At each time, supernatant fluid was removed, diluted in sterile PBS, and inoculated onto BCYE plates. CHO cells were lysed with 1 ml of sterile distilled water to release intracellular L. pneumophila, and aliquots of the cell lysates were immediately diluted in distilled water and inoculated onto BCYE agar plates. L. pneumophila was shown to remain viable but did not multiply in sterile distilled water for 30 min. In antibioticfree medium A, the bacteria remained viable for 8 h but did not multiply (data not shown). Peroxidase-conjugated antibody staining. Peroxidase-conjugated antibody staining was used to examine cell monolayers in 24-well plates. Cells were washed three times with PBS and then fixed for 30 min with 3% Formalin in PBS. Following fixation, the cells were washed with PBS (1 ml per well, three times), permeabilized with 0.5 ml of 100% methanol per well for 10 min, and then washed again with PBS (1 ml per well, three times). Rabbit antiserum against Legionella serogroups 1 through 6 (Accurate Chemical and Scientific Corporation, Westbury, N.Y.) was diluted 1:100 in PBS, and 200 ,ul was added to each monolayer for 45 min at 37°C. The cells were washed three times with 1 ml of PBS, and 200 IlI of goat anti-rabbit immunoglobulin conjugated to horseradish peroxidase (Bio-Rad Laboratories, Richmond, Calif.), diluted 1:100 in PBS, was added and incubated with the cells at 37°C for 45 min. Monolayers were washed with PBS, exposed to diaminobenzidine (0.5 mg/ml in PBS) and H202 (0.001%, vol/vol) in PBS for 30 min, and then washed with PBS prior to examination with an inverted microscope. Fluorescence microscopy. Intracellular bacteria were distinguished from extracellular bacteria by using immunofluorescence staining of cells cultured on 10-mm glass coverslips (13, 31). Following experimental infections, CHO cell monolayers were washed with PBS and incubated at 37°C for 30 min with rabbit anti-L. pneumophila antiserum diluted 1:100 (vollvol) in PBS. All subsequent incubations with antibodies were at 37°C. After the monolayers were rinsed with PBS, they were incubated with rhodamine-conjugated goat antirabbit antibody (1:300, vol/vol, in PBS) for 30 min and then washed and fixed with 3.7% Formalin in PBS for 15 min at room temperature. Following permeabilization of monolayers with 100% methanol for 5 min, cells were washed with PBS and then incubated with rabbit anti-Legionella antiserum for 30 min. Monolayers were then washed with PBS and incubated with fluorescein-conjugated goat anti-rabbit antibody for 30 min. After a final wash in PBS, fluorescence was visualized by using a Zeiss fluorescence microscope with appropriate filters to distinguish the rhodamine and fluorescein conjugates. Protein determinations. Protein concentration was measured with Bradford reagent (Pierce Chemical Company, Rockford, Ill.) by using bovine serum albumin as the standard. Aliquots were obtained after lysis of the CHO cell monolayers with 0.1% sodium dodecyl sulfate (SDS) or 0.1% Triton X-100 in water. Radiolabeling experiments. Radiolabeling studies were done by adding 0.5 to 1.0 ,uCi of [35S]methionine or [3H]leucine (ICN Radiochemicals, Irvine, Calif.) to each well in 0.5 ml of medium A and incubating for 1 h at 37°C. The medium was removed, and the monolayers were washed

241

three times with 1 ml of PBS and lysed by addition of 0.5 ml of 0.1% SDS in water. A 0.1-ml aliquot was removed for protein measurement. A 0.35-ml aliquot was brought to a final concentration of 20% trichloroacetic acid, frozen at -20°C for 1 h, and then thawed, and the resultant precipitates were collected on fiberglass filters (Boehringer Mannheim Biochemicals, Indianapolis, Ind.) by using a Hoefer filter apparatus (Hoefer Scientific Instruments, San Francisco, Calif.). The filters were placed in 10 ml of Optifluor (Packard Instrument Company, Inc., Downers Grove, Ill.), and radioactivity was measured by using a Hewlett Packard scintillation counter (Packard Instrument Company, Inc.). L. pneumophia culture conditions and cell toxicity. To determine the effect of culture density on L. pneumophila toxicity to CHO cells, bacteria were inoculated onto BCYE solid medium (100-mm dishes; Becton Dickinson Labware) at different initial concentrations. Aliquots (100 ,l) of L. pneumophila were thawed and spread on a 2-cm2 area on BCYE plates and then cultured for 16 h. The bacteria were then scraped from the plates and suspended in PBS. Quantitation of CFU was estimated by measuring the absorbance (optical density at 550 nm) and by colony counting after plating dilutions of the bacteria. Initially, 3 x 108, 3 x 109, and 3 x 1010 CFU of L. pneumophila were inoculated and spread over the entire surface of separate BCYE plates with a sterile Dacron swab and then incubated at 37°C for 16 h. Bacteria from each plate were suspended in sterile PBS, and the total number of bacteria per plate was estimated by measuring absorbance and by colony counting. Bacteria from each plate were diluted in PBS and incubated with CHO cells (MOI of 20: 1) for protein synthesis measurements as described above. Degradation of proteins in CHO cells infected with L. pneumophila. CHO cells were incubated with 1 RCi of [35S]methionine per ml in medium A for 2 h to label proteins. Following removal of the radiolabel by washing, the monolayers were incubated with L. pneumophila at a MOI of 20:1 for 3 h. The monolayers were washed and then incubated in medium A for 1 h. Radioactivity was determined after lysis with SDS (0.1%) and trichloroacetic acid (TCA) precipitation, as described above. Treatment of L. pneumophila with washing, trypsin, or heat. L. pneumophila was suspended in PBS and adjusted to an optical density at 550 nm of 0.2. One milliliter of the suspended L. pneumophila was added to a 0.2 pM nitrocellulose filter, and 30 ml of sterile PBS was forced over the filter. Washed L. pneumophila was removed from the filter by gentle agitation in sterile PBS, and the optical density at 550 nm was measured. These L. pneumophila cells were incubated with CHO cells at an MOI of 20:1, and the collected effluent from the wash was incubated with CHO cells for 4 h before protein synthesis was measured. Trypsin treatment was effected by suspension of L. pneumophila in PBS with 0.25% trypsin (GIBCO Laboratories, Life Technologies, Inc., Grand Island, N.Y.) for 15 min. After addition of an equimolar amount of soybean trypsin inhibitor (Sigma Chemical Co.), CHO cells were infected at an MOI of 20:1. Controls were uninfected cells and cells infected with L. pneumophila which were neither washed nor trypsin treated. CHO cell protein synthesis was then measured as described above. L. pneumophila was suspended in PBS at an optical density at 550 nm of 0.8, placed in sterile capped polypropylene microcentrifuge tubes, and immersed in boiling water for 2 min. The tube was removed and cooled to room temperature with tap water. After this treatment, the bacte-

242

McCUSKER ET AL.

ria were incubated with CHO cells (MOI of 20:1) as described above. Cytochalasin D treatment. Uptake of bacteria was inhibited with cytochalasin D by using a modification of the method of Elliot and Winn (7). CHO cell monolayers were incubated with medium A containing 0.5 mg of cytochalasin D (Sigma Chemical Co.) per ml for 16 h before the addition of L. pneumophila at an MOI of 20:1 for 3 h. Cytochalasin D (0.5 mg/ml) was included in the PBS and in medium A for subsequent washes and incubation. The CHO cells were incubated for 1 to 2 h in medium A before protein synthesis was measured or immunolocalization studies were performed. Chromium uptake and release. Release of 51Cr and intracellular concentrations of 51Cr were compared in control cells and L. pneumophila-infected cells. CHO cells were incubated overnight with medium A containing 1 ,uCi of 51Cr per ml. Cells were washed with PBS and then incubated with L. pneumophila at an MOI of 20:1 in medium A for 3 h. The monolayers were washed to remove unattached bacteria and incubated in medium A without antibiotics for 1 h. Radioactivity was measured in aliquots of supernatants to determine spontaneous release of label from cells and in aliquots from cells lysed with 0.1% SDS. Methionine uptake. Transport of methionine into cells was measured by using a modification of the method of Foster and Pardee (9). CHO cells were infected with L. pneumophila at an MOI of 30:1 and then washed and incubated in PBS for 1 h at 37°C. Cells were then incubated in PBS containing [35S]methionine (1 ,uCi/ml). After 5 and 15 min of incubation, cells were chilled by placing the tissue culture plates on an ice bath. The culture supernates were removed, and cell monolayers were rapidly washed three times with 1 ml of PBS (4°C). After the cells were lysed with 0.5 N NaOH, aliquots were removed for assay of radioactivity and for determination of protein concentration. Antibiotic killing of L. pneumophila. L. pneumophila was cultured on BCYE and suspended at a concentration of approximately 109 bacteria per ml in PBS or in PBS containing 100 ,ug of gentamicin (Sigma Chemical Co.) per ml and 500 ,ug of erythromycin (Sigma Chemical Co.) per ml. Protein synthesis was measured by adding [35S]methionine (1 ,uCi/ml) and incubating for 1 h. Aliquots were removed at 15-min intervals, and bacteria were separated from extracellular [35S]methionine by spinning for 2 min at 8,000 x g in a microcentrifuge. The pellets were washed by suspending them in 1 ml of PBS followed by centrifugation in the microcentrifuge. The bacteria were solubilized with 0.5 ml of 0.1% SDS and heated at 95°C for 2 min. A 0. 1-ml aliquot was removed for protein determination, and a 0.35-ml aliquot was adjusted to 20% TCA and assayed for radioactivity as described above. Viability of antibiotic-treated L. pneumophila was assessed by using bacteria which had been incubated in PBS or in PBS with gentamicin (100 ,ug/ml) and erythromycin (500 ,ug/ml) for 1 h. Bacteria were diluted and incubated on BCYE for 4 days before determination of colony numbers. RESULTS Initially, we compared the ability of L. pneumophila to adhere to different continuous cell lines. L. pneumophila was incubated for 16 h on BCYE solid medium, suspended in PBS, and then incubated for 3 h with cell monolayers in antibiotic-free tissue culture medium at an MOI of 20:1. After unattached bacteria were removed by washing, 80 to

INFECT. IMMUN.

7

10

c

6

10

E m

I

12

I

I

24 36 Time (hours)

48

FIG. 1. Intracellular replication of L. pneumophila in CHO cells. Bacteria were incubated with CHO cells at an MOI of 2:1 for 3 h, unattached bacteria were removed by washing with PBS, and cells were incubated in antibiotic-free tissue culture medium. At the times indicated (U), cells were lysed with sterile distilled water and aliquots were plated on BCYE. Results shown are the means of two determinations.

90% of the CHO cells were associated with at least one L. pneumophila bacterium, as evidenced by peroxidase antibody staining. In contrast, only 10% of HeLa and 35 to 45% of Vero cells were associated with L. pneumophila. Therefore CHO cell lines were used for the experiments described below. We next determined if CHO cells could support the growth of L. pneumophila. L. pneumophila was incubated with CHO cells at a low MOI (2:1) since a higher MOI (20:1) caused disruption of CHO cell monolayers 8 to 10 h after incubation (data not shown). Immunostaining showed that approximately 5% of the CHO cells were associated with L. pneumophila after 3 h of incubation (data not shown) at an MOI of 2:1. There appeared to be a lag phase in Legionella growth of about 6 to 9 h (Fig. 1). Between 12 and 24 h, Legionella numbers increased logarithmically, with a doubling time of about 2 h. Quantitation of L. pneumophila in the tissue culture medium indicated that after 24 h of incubation, significant numbers of bacteria were released and many CHO cells were lysed. Since tissue culture medium and CHO cell conditioned culture medium did not support the growth of L. pneumophila, these results indicated that L. pneumophila multiplied within CHO cells. CHO cells incubated with L. pneumophila at an MOI of 20:1 exhibited cytopathic changes after 8 to 10 h (data not shown). One mechanism by which bacteria can cause cytotoxicity is by inhibiting host cell macromolecular synthesis. To test this hypothesis, we measured the rate of incorporation of radiolabeled amino acids into TCA-precipitable material in both infected and uninfected CHO cells. After extracellular L. pneumophila was removed from the infected CHO cells by washing, host cell protein synthesis was found to be inhibited by 70 to 80%. Protein synthesis in infected cells did not appear to recover (Fig. 2). Similar results were found by using [3H]leucine. These results indicated that infection of CHO cells with L. pneumophila resulted in a marked reduction in the ability of CHO cells to incorporate radiolabeled amino acids into proteins.

L. PNEUMOPHILA INHIBITS CHO CELL PROTEIN SYNTHESIS

VOL. 59, 1991

243

1200 -

1000 -

100

800 0e

,, 8 ,c

600 400 -

&

200-

i 1 II

.

2

4

.

6

8

10

Time (hours) FIG. 2. Inhibition of CHO cell protein synthesis by L. pneumophila. [35S]methionine incorporation into TCA-precipitable material (1-h pulses) was measured at times indicated after a 3-h incubation of bacteria with CHO cells as described in Materials and Methods. Symbols: O, control cells; A, cells incubated with L. pneumophila. Error bars represent standard error of the mean.

To determine if host cell amino acid transport was inhibited by L. pneumophila infection, we examined methionine uptake in both infected and uninfected CHO cells as described in Materials and Methods. Our results showed that over 15 min, the rate of methionine uptake was similar for both infected and uninfected CHO cells (Fig. 3). Under these same experimental conditions, CHO cell protein synthesis was blocked by 80% in infected cells. Thus, CHO cells infected with L. pneumophila appeared to transport methionine at the same rate as uninfected cells. Assays of CHO cell permeability were performed by incubating CHO cells overnight with 51Cr, followed by rinsing the monolayers with PBS and incubating them in medium A with L. pneumophila for 3 h. The extracellular bacteria were removed by washing with PBS and cells incubated for 1 h in medium A. Quantitation of cell-associated and extracellular 51Cr showed 13% spontaneous release of 51Cr from uninfected control cells and 10% release from cells infected with L. pneumophila. These results indicated that L. pneumophila did not disrupt the integrity of CHO cell membranes. The toxic effect of L. pneumophila on CHO cell protein synthesis was dependent on the conditions used for growth of the bacteria. L. pneumophila incubated at starting concentrations of 3 x 108, 3 x 109, and 3 x 1010 CFU per plate yielded approximately 7 x 109, 5 x 1010, and 1.2 x 1011 CFU

3000

-

2000

-

0,

U 1000-

C

5

10

15

Time (minutes)

FIG. 3. Effect of L. pneumophila infection on transport of [35S]methionine into CHO cells. [35S]methionine uptake after 5- and 15-min incubations was measured in uninfected and infected CHO cells (see Materials and Methods). Symbols: A, L. pneumophila; O, uninfected cells. Error bars show standard deviation from the mean.

>

C

:t

o

50

tR

O I

7xl0

5x10

1.2xl0

Number of Legionella/plate

FIG. 4. Dependence of inhibition of CHO cell protein synthesis by L. pneumophila on bacterial culture conditions. Bacteria seeded at different initial concentrations were incubated for 16 h at 37°C, suspended in PBS, and incubated with CHO cells at an MOI of 20:1. Only bacteria from plates with approximately 101" CFU per plate caused significant inhibition of protein synthesis in CHO cells. Error bars show standard deviation from the mean.

per plate, respectively, after incubation for 16 h at 37°C. L. pneumophila from each of these plates was incubated with CHO cells at an MOI of 20:1. Only bacteria from the plate with the highest initial inoculum concentration of 3 x 1010 CFU per plate caused significant inhibition of CHO cell protein synthesis (Fig. 4). Antibody staining of CHO cells incubated with L. pneumophila from plates with lower initial inoculum concentrations showed that less than 10% of the CHO cells were associated with bacteria (data not shown). These results suggest that the conditions used for culture of L. pneumophila affect the ability of the bacteria to inhibit protein synthesis in CHO cells. It has recently been shown that L. pneumophila secretes proteases capable of degrading a wide variety of protein substrates (1, 3-5). While the results presented above indicated that host protein synthesis was reduced by Legionella infection, it was possible that Legionella proteolytic activity degraded host peptides, resulting in a net decrease of accumulation of [35S]methionine into TCA-precipitable material. This possibility was examined by incubating CHO cells with [35S]methionine before adding L. pneumophila. Four hours after infection, L. pneumophila-infected cells and uninfected cells contained similar quantities of [35S]methionine-labeled TCA-precipitable material (Fig. 5). In separate wells, pulselabeling for 1 h with [35S]methionine 4 h after infection showed that protein synthesis of infected cells was reduced by 80% compared with uninfected controls. These findings argue against the possibility that significant degradation of nascent protein was occurring in L. pneumophila-infected CHO cells. We next determined whether internalization of L. pneumophila by CHO cells was required for the toxic effect on protein synthesis. Internalization of L. pneumophila was blocked by incubation of CHO cells with cytochalasin D prior to addition of the bacteria. This treatment was effective in inhibiting Legionella uptake by at least 95%, as evidenced by immunolocalization using fluorescent-antibody staining (data not shown). Treatment with cytochalasin D did not appear to affect the ability of L. pneumophila to inhibit CHO cell protein synthesis (Fig. 6). These observations suggested that L. pneumophila could exert toxic activity against CHO cell protein synthesis in the absence of internalization.

McCUSKER ET AL.

244

INFECT. IMMUN.

A

2

60000

L

o.

5000C

a

3000C

i

200C

to

control Legionella

4000C

0.

on

X Q C.)

Gentamicin and erythromycin

1000c

Uc

20 40 60 80 100

Time (minutes)

Infected FIG. 5. Effect of L. pneumophila on CHO cell protein degradation. CHO cells were incubated with [35S]methionine for 2 h prior to infection with L. pneumophila. The amount of radiolabeled methionine in TCA-precipitable proteins was measured as described in Materials and Methods. Open column, uninfected cells; stippled column, infected CHO cells. Control

B 300.

c

200

I

0.

b

L. pneumophila inhibited protein synthesis after extensive washing of the bacteria with PBS prior to infection. The effluent from the washed bacteria did not cause reduction of CHO cell protein synthesis. Treatment of L. pneumophila with 0.25% trypsin for 15 min at 37°C prior to infection of CHO cells had no effect on protein synthesis inhibition (data not shown). Moreover, bacteria which had been heated to 950C for 2 min prior to incubation with CHO cells did not cause inhibition of protein synthesis. These results indicated that the toxic Legionella component which inhibited CHO cell protein synthesis did not appear to be secreted in an active form, that it was trypsin resistant, and that it was heat sensitive. The CHO cell protein synthesis inhibition described above could be due to a molecule produced by L. pneumophila after contact with CHO cells. To examine this possibility, we determined whether treatment of L. pneumophila with antibiotics prior to incubation with CHO cells would alter the toxic activity on cell protein synthesis. L. pneumophila was treated with high levels of gentamicin and erythromycin in PBS for 1 h. After this treatment, Legionella protein synthesis was completely inhibited (Fig. 7A) and there was about a 5-log reduction in viable L. pneumophila. Antibiotic-killed L. pneumophila inhibited protein synthesis of CHO cells by 80%, as did bacteria incubated in PBS without antibiotics 1000

800 .S

0o

CONTROL 2:1

5:1 50:1 M.O.I.

50:1

FIG. 7. Inhibition of CHO cell protein synthesis by antibiotickilled L. pneumophila. (A) L. pneumophila was suspended in PBS and treated with gentamicin (100 ,ug/ml) and erythromycin (500 ,g/ml) for 1 h prior to incubation with CHO cells. This treatment completely inhibited bacterial protein synthesis. (B) Antibiotickilled L. pneumophila was incubated with CHO cells at an MOI of 50:1 for 3 h. Bacteria incubated in PBS without antibiotics were added to CHO cells at MOI ranging from 2:1 to 50:1. Protein synthesis was measured as described in Materials and Methods. Open column, uninfected cells; solid column, cells infected with antibiotic-killed L. pneumophila; stippled columns, CHO cells incubated with PBS-treated L. pneumophila.

(Fig. 7). Immunofluorescence showed that antibiotic-killed L. pneumophila was associated with and internalized by CHO cells. If 5% of the Legionella cells were still viable after this antibiotic treatment, the effective MOI would be approximately 2 bacteria per cell. At this MOI CHO cell protein synthesis was reduced by only 15% (Fig. 7). These results indicated that L. pneumophila retains its toxic activity even after being killed with antibiotics and strongly suggested that ongoing Legionella protein synthesis was not required for toxicity.

-

600-

O>

400-

&

200 -

IU

100-

CONTROLS

INFECTED W1TH CYTO D

INFECTED WITHOUT CYTO D

FIG. 6. Inhibition of protein synthesis of cytochalasin-treated CHO cells by L. pneumophila. Open column, uninfected CHO cells; black column, CHO cells infected with L. pneumophila in presence of cytochalasin D; stippled column, CHO cells not exposed to cytochalasin D before infection with L. pneumophila. Error bars show standard deviation from the mean.

DISCUSSION Our results indicate that L. pneumophila causes inhibition of protein synthesis in CHO cells. The inhibition did not appear to result from diminished amino acid transport (Fig. 3) or from disruption of CHO cell membranes, as measured by 51Cr release. Inhibition of CHO cell protein synthesis occurred under conditions in which internalization of bacteria was blocked by cytochalasin D treatment, suggesting that extracellular L. pneumophila could inhibit protein synthesis of CHO cells. However, L. pneumophila did not appear to release an active toxic factor(s) into the culture medium. Inhibition of protein synthesis was not observed after L. pneumophila had been boiled for 2 min, suggesting that the toxic factor is not endotoxin, which is heat stable (20). Interactions between L. pneumophila and continuous cell

VOL. 59, 1991

L. PNEUMOPHILA INHIBITS CHO CELL PROTEIN SYNTHESIS

lines such as Vero, HeLa, MRC-5, HEp-2, U937, and HL-60 have been reported (19, 25, 28, 37). However, the percentage of cells infected after the initial incubation period with L. pneumophila was not reported. In these studies, peak numbers of L. pneumophila were reached after long periods of incubation (2 to 5 days), which suggests that initially only a small fraction of cells were infected with bacteria. In our studies, the efficient infection of a high percentage of cells in the monolayer suggests that specialized mechanisms of adherence of L. pneumophila to CHO cells may exist. Uptake of L. pneumophila by human monocytes is facilitated by engagement of complement receptors (27). Perhaps uptake in CHO cells occurs through a structure analogous to the complement receptor. We are currently investigating this possibility. The findings presented here should facilitate further studies of the cell biology of the host-parasite relationships in L. pneumophila infections. The CHO cell model represents a valuable tool for further studies of basic mechanisms of L. pneumophila infection. We demonstrate that a high percentage of CHO cells in monolayers are infected synchronously after a 3-h incubation with L. pneumophila. Synchronous infections allow investigation of host cell responses to infection with this intracellular parasite. Mutant CHO cell lines have been used to investigate the cell biology of other intracellular parasites (30) and to examine mechanisms of bacterial toxins. For example, CHO cells which are deficient in endosomal acidification have been used to demonstrate that diphtheria toxin requires access to an acidified compartment to be activated (32). We cannot directly extrapolate our findings with CHO cells to monocytes or macrophages, which are primary sites of infection in humans. We have recently found that L. pneumophila inhibits protein synthesis in U937 cells and human monocytes (21a), which suggests that the CHO cell is a valid model for further study. In addition, recent reports of L. pneumophila infections of soft tissue and prosthetic valves suggest that these bacteria may invade nonphagocytic cells (2, 35). The conditions used for Legionella growth affected the ability of the bacteria to inhibit protein synthesis in CHO cells. Only bacteria obtained from high-density cultures (approximately 1011 CFU per plate) caused significant inhibition of CHO cell protein synthesis (Fig. 4). Previous investigations have shown that expression of diphtheria toxin (26) and invasiveness of Salmonella spp. are dependent on the conditions used for bacterial growth (18). Possibly at higher concentrations of bacteria, nutrient depletion induces expression of the factor(s) required for inhibition of protein synthesis. Alternatively, the toxic factor(s) may be expressed only during a specific phase of growth, as occurs in exotoxin expression by Staphylococcus aureus during stationary phase (29). To test this hypothesis, we will need to examine L. pneumophila obtained from broth cultures at different growth phases. L. pneumophila inoculated onto BCYE plates at lower initial concentrations did not inhibit protein synthesis. Antibody staining showed that a much lower percentage of bacteria from these cultures adhered to CHO cells (data not shown). Thus, it is possible that bacterial growth conditions influence the expression of an adhesive molecule required for attachment to CHO cell surfaces. Our results indicated that inhibition of CHO cell protein synthesis occurred even after extensive washing of L. pneumophila. Further, the eluant from washed L. pneumophila did not inhibit protein synthesis in CHO cells. Thus, the

245

factor causing inhibition of CHO cell protein synthesis reported here does not appear to be soluble in tissue culture medium or PBS. In contrast, the 1.2-kDa peptide previously shown to have cytotoxic effects on CHO cells (10) and to inhibit respiratory burst in neutrophils (11) was present in Legionella culture supernates. Most bacterial toxins that inhibit protein synthesis, such as diphtheria toxin, Pseudomonas exotoxin A, and colicin E, are soluble peptides which enter the cytosol of the affected cell (23, 33). The inhibition of CHO cell protein synthesis by L. pneumophila reported here, however, appears to be tightly associated with bacterial cells. Clostridium perfringens enterotoxin causes inhibition of protein synthesis, RNA synthesis, and DNA synthesis in Vero cells by inducing leakage of low-molecular-weight molecules as shown by sensitive assays of cell permeability utilizing 86Rb and [3H]uridine (17, 21). The enterotoxin appears to remain tightly associated with the Vero cell membrane, distinguishing its mechanism of action from soluble toxins. We are currently attempting to isolate the factor which inhibits CHO cell protein synthesis from L. pneumophila as a first step in understanding its mechanism of action. Intracellular parasites depend on continued host cell function for nutrients (24). Since L. pneumophila utilizes amino acids rather than carbohydrates for its energy source (34), it is possible that inhibition of protein synthesis of the host cell would increase availability of nutrients to the bacteria. Intracellular growth of Chlamydia sp. is facilitated when host protein synthesis is inhibited by cycloheximide (12). Of course, severe inhibition of protein synthesis by L. pneumophila will eventually kill host cells. Further studies are in progress in our laboratory to determine whether L. pneumophila inhibits protein synthesis in human macrophages. ACKNOWLEDGMENTS We thank Jerry Kaplan, Don Summers, and John Hoidal for helpful discussions and suggestions. We also thank Helen Wilkinson, Stuart Arfin, Spotswood Spruance, and April Robbins for bacterial strains and mammalian cell lines used for this study. Robert Moesinger provided excellent technical assistance. David Low was supported by a research career development award from the National Institutes of Health (grant #AI 0881), and Kevin McCusker was supported by a Biomedical Research Support Grant (#2S07RR05428-29).

REFERENCES 1. Baskervifle, A., J. W. Conlan, L. A. E. Ashworth, and A. B. Dowsett. 1986. Pulmonary damage caused by a protease from Legionella pneumophila. Br. J. Exp. Pathol. 67:527-536. 2. Brabender, W., D. R. Hinthomn, M. Asher, N. J. Lindsey, and C. Liu. 1983. Legionella pneumophila wound infection. JAMA 250:3091-3092. 3. Conlan, J. W., A. Williams, and L. A. E. Ashworth. 1988. In vivo production of a tissue-destructive protease by Legionella pneumophila in the lungs of experimentally infected guinea pigs. J. Gen. Microbiol. 132:143-149. 4. Conlan, J. W., A. Williams, and L. A. E. Ashworth. 1988. Inactivation of human a-l-antitrypsin by a tissue-destructive protease of Legionella pneumophila. J. Gen. Microbiol. 134: 481-487. 5. Dreyfus, L. A., and B. H. Iglewski. 1986. Purification and characterization of an extracellular protease of Legionella pneumophila. Infect. Immun. 51:736-743. 6. Elliott, J. A., and W. Johnson. 1982. Virulence conversion of Legionella pneumophila serogroup 1 by passage in guinea pigs and embryonated eggs. Infect. Immun. 35:943-946. 7. Elliott, J. A., and W. C. Winn, Jr. 1986. Treatment of alveolar macrophages with cytochalasin D inhibits uptake and subse-

246

McCUSKER ET AL.

quent growth of Legionella pneumophila. Infect. Immun. 51: 31-36. 8. Feeley, J. C., R. J. Gibson, G. W. Gorman, N. C. Langford, J. K. Rasheed, D. C. Mackel, and W. B. Baine. 1979. Charcoalyeast extract agar: primary isolation medium for Legionella pneumophila. J. Clin. Microbiol. 10:437-441. 9. Foster, D. O., and A. B. Pardee. 1969. Transport of amino acids by confluent and nonconfluent 3T3 and polyoma virus-transformed 3T3 cells growing on glass cover slips. J. Biol. Chem. 244:2675-2681. 10. Friedman, R. L., B. H. Iglewski, and R. D. Miller. 1980. Identification of a cytotoxin produced by Legionella pneumophila. Infect. Immun. 29:271-274. 11. Friedman, R. L., J. E. Lochner, R. H. Bigley, and B. H. Iglewski. 1982. The effects of Legionella pneumophila toxin on oxidative processes and bacterial killing of human polymorphonuclear leukocytes. J. Infect. Dis. 146(3):328-334. 12. Hatch, T. P. 1975. Competition between Chlamydia psittaci and L cells for host isoleucine pools: a limiting factor in chlamydial multiplication. Infect. Immun. 12:211-220. 13. Heesemann, J., and R. Laufs. 1985. Double immunofluorescence microscopic technique for accurate differentiation of extracellularly and intracellularly located bacteria in cell culture. J. Clin. Microbiol. 22:168-175. 14. Horwitz, M. A. 1983. Formation of a novel phagosome by the Legionnaires' disease bacterium (Legionella pneumophila) in human monocytes. J, Exp. Med. 158:1319-1331. 15. Horwitz, M. A. 1984. Phagocytosis of the Legionnaires' disease bacterium (Legionella pneumophila) occurs by a novel mechanism: engulfment within a pseudopod coil. Cell 36(1):27-33. 16. Horwitz, M. A. 1987. Characterization of avirulent mutant Legionella pneumophila that survive but do not multiply within human monocytes. J. Exp. Med. 166:1310-1328. 17. Hulkower, K. I., A. P. Wnek, and B. A. McClane. 1989. Evidence that alterations in small molecule permeability are involved in the Clostridium perfringens type A enterotoxininduced inhibition of macromolecular synthesis in Vero cells. J. Cell. Physiol. 140:498-504. 18. Lee, C. A., and S. Falkow. 1990. The ability of Salmonella to enter mammalian cells is affected by bacterial growth state. Proc. Natl. Acad. Sci. USA 87:4304-4308. 19. Marra, A., M. A. Horwitz, and H. A. Shuman. 1990. The HL-60 model for the interaction of human macrophages with the Legionnaires' disease bacterium. J. Immunol. 144:2738-2744. 20. McCartney, A. C., and A. C. Wardlaw. 1985. Endotoxic activities of lipopolysaccharides. In D.E.S. Stewart-Tull and M. Davies (ed.), Immunology of the bacterial cell envelope. John Wiley & Sons, Inc., New York. 21. McClane, B. A., A. P. Wnek, K. I. Hulkower, and P. C. Hanna. 1988. Divalent cation involvement in the action of Clostridium perfringens type A enterotoxin. J. Biol. Chem. 263:2423-2435.

INFECT. IMMUN. 21a.McCusker, K. T., and D. Low. Unpublished data. 22. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 23. Moss, J., and M. Vaughan. 1988. Cholera toxin and E. coli enterotoxins and their mechanism of action. In M. C. Hardegree and A. T. Tu (ed.), Bacterial toxins: handbook of natural toxins. Marcel Dekker, Inc., New York. 24. Moulder, J. W. 1985. Comparative biology of intracellular parasitism. Microbiol. Rev. 49:298-337. 25. Oldham, L. J., and F. G. Rodgers. 1985. Adhesion, penetration and intracellular replication of Legionella pneumophila: an in vitro model of pathogenesis. J. Gen. Microbiol. 131:697-706. 26. Pappenheimer, A. M. 1977. Diphtheria toxin. Annu. Rev. Biochem. 46:69-94. 27. Payne, N. R., and M. A. Horwitz. 1987. Phagocytosis of Legionella pneumophila is mediated by human monocyte complement receptors. J. Exp. Med. 166:1377-1389. 28. Pearlman, E., A. H. Jiwa, N. C. Engleberg, and B. I. Eisenstein. 1988. Growth of Legionella pneumophila in a human macrophage-like (U937) cell line. Microb. Pathog. 5:87-95. 29. Peng, H.-L., R. P. Novick, B. Kreiswirth, J. Kornblum, and P. Schlievert. 1988. Cloning, characterization, and sequencing of an accessory gene regulator (agr) in Staphylococcus aureus. J. Bacteriol. 170:4365-4372. 30. Pfefferkorn, E. R., J. Schwartzman, and L. Kasper. 1983. Toxoplasma gondii: use of mutants to study the host-parasite relationship. CIBA Found. Symp. 99:74-91. 31. Portnoy, D. A., S. Jacks, and D. Hinrichs. 1988. Role of hemolysin for the intracellular growth of Listeria monocytogenes. J. Exp. Med. 167:1459-1471. 32. Robbins, A. R., C. Oliva, J. L. Bateman, S. S. Drag, C. J. Galloway, and I. Mellman. 1984. A single mutation in Chinese hamster ovary cells impairs both golgi and endosomal functions. J. Cell Biol. 99:1296-1308. 33. Stephen, J., and R. A. Pietrowski. 1986. Bacterial toxins. Aspects of microbiology, vol. 2, 2nd ed. American Society for Microbiology, Washington, D.C. 34. Tesh, M. J., S. A. Morse, and R. D. Miller. 1983. Intermediary metabolism in Legionella pneumophila: utilization of amino acids and other compounds as energy sources. J. Bacteriol.

154:1104-1109. 35. Tompkins, L. S., B. J. Roessler, S. C. Redd, L. E. Markowitz, and M. L. Cohen. 1988. Legionella prosthetic-valve endocarditis. N. Engl. J. Med. 318:530-534. 36. Winn, W. C., Jr. 1988. Legionnaires disease: historical perspective. Clin. Microbiol. Rev. 1:60-81. 37. Wong, M. C., E. P. Ewing, Jr., C. S. Callaway, and W. L. Peacock, Jr. 1980. Intracellular multiplication of Legionella pneumophila in cultured human embryonic lung fibroblasts. Infect. Immun. 28:1014-1018.