SYNTHESIS OF VIRULENCE FACTORS BY PASTEURELLA PESTIS G. M. FUKUI, W. D. LAWTON, AND R. P. MORTLOCK U. S. Army Chemical Corps, Fort Detrick, Frederick, Maryland

Received for publication October 14, 1960

At least four antigens of Pasteurella pestis have been directly associated with the expression of maximal virulence; namely, toxin, fraction I, V and W antigens (Burrows and Bacon, 1958). The synthesis of the latter three antigens has been reported by others to be temperature dependent (Schutze, 1934; Burrows and Bacon, 1956a, b). Virulence of P. pestis was consistently reduced by culturing at 5 C; and of the four antigens, only toxin was produced in quantities detectable by the gel-diffusion techniques (Fukui et al., 1959, 1960). After subsequent incubation of such cultures at 37 C for 6 to 8 hr, however, fraction I and V and W antigens were detected and maximal virulence was restored to the culture The lack of significant increase in either viable or total cell count during the incubation period at 37 C was presumptive evidence that phenotypic rather than genotypic changes accounted for the restoration of virulence. The results of previous studies indicated that the restoration of virulence by incubation at 37 C required an adequate energy source, amino acids, oxygen, and pH maintained between 5.5 and 8.4 (Naylor, Fukui, and McDuff, 1961). The work reported in this paper was undertaken to attempt to define the relation of nucleic acids and protein synthesis to the synthesis of virulence antigens and the temperature dependent restoration of virulence. MATERIAL AND METHODS

The Alexander strain of P. pestis was used for this study. The methods involved in the maintenance, growth, and harvesting of the cultures have been described previously (Fukui et al., 1960; Naylor et al., 1961). Likewise, the antigen, viable cell, and virulence assays have been described in previous reports from our laboratory (Lawton, Fukui, and Surgalla, 1960). Protein was assayed by the method described by Lowry et al. (1951), deoxyribonucleic acid by the method of Burton (1956), and ribonucleic acid bv the method of Ceriotti

656

(1955). Dry weights were determined from optical density measurements at 655 m,u with the Coleman model 9 spectrophotometer using calibration curves prepared for either cells grown at 5 C, or at 5 C with subsequent incubation at 37 C for 6 hr in heart infusion broth (Difco). Chloramphenicol was purchased from ParkeDavis. 5-Fluorodeoxyuridine was kindly supplied by Hoffmann-LaRoche, Inc. 5-Bromodeoxyuridine, 5-hydroxydeoxyuridine, and 5-bromouracil were obtained from the California Corporation for Biochemical Research. Mitomycin C was supplied with the compliments of the Bristol Laboratories. Other amino acid and nucleic acid analogues were purchased from the Nutritional Biochemicals Corporation. RESULTS AND DISCUSSION

Preliminary experiments were performed by adding various antimetabolites separately to suspensions of cells grown at 5 C in heart infusion broth to see if we might be able to discern a correlation of nucleic acid and protein synthesis with the restoration of virulence. The results of numerous experiments have demonstrated that amino acid analogues (DL-p-fluorophenylalanine, and DL-f-phenyllactic acid) and purine or pyrimidine analogues (8-azaguanine and 5-fluorodeoxyuridine), as well as a protein synthesis inhibitor (chloramphenicol), were capable of completely blocking the restoration of virulence (Table 1). Other antimetabolites tested did not inhibit the restoration of virulence; however, this may not be evidence for their inactivity. The concentrations tested may have been inadequate, or they may not have penetrated into the cell under the test conditions. At the levels used, the antimetabolites did not kill the cells during the 6 hr of incubation at 37 C. The relative inhibitory effects of the antimetabolites were calculated from the difference in LDD values resulting from mouse intraperitoneal virulence titrations performed with the culture containing antimetabo-

1961]

657

SYNTHESIS OF VIRULENCE BY P. PESTIS

TABLE 1 Effect of antimetabolites on restoration of virulence Antimetabolite

Concn*

Inhibitiont

sgiml

%

10 .......... Chloramphenicol .. DL-p-fluorophenylalanine ... 2,930 2,688 DL-t3-phenyllactic ............ 8-Azaguanine ............... 500 100 5-Fluorodeoxyuridine ..

5-Bromodeoxyuridine ........ 1,000 5-Hydroxydeoxyuridine ...... 100 ....... ...60 5-Hydroxyuridine. 5-Bromouracil ............... 1,000 1,000 .......... 2,6-Diaminopurine. 10 6-Azauracil .................. 10 6-Azathymine ............... 2 Benzimidazole ............... 2-Amino4-methyl pyrimi10 dine ................... 100 Kinetin ..................... 0.1 Mitomycin C ................

100 100 100 100 100 7 0 7 0 0 0 0 0 0 0 0

Nonlethal concentrations. t Calculated from LDWo values.

*

lites as compared to the LDs value obtained in the control system where no antimetabolite was added. The blocking effect of 8-azaguanine was obtained consistently only when it was tested in a chemically defined medium (Wessman, Miller, and Surgalla, 1958).

TABLE 2 Effect of chloramphenicol on the restoration of virulence of Pasteurella pestis Chloramphen-

icolg

Viable prl

Antigens*

FI

V

Culture grown at 5 C: None | 6.0 | - 5 C Culture incubated 6 None 8.6 Tr + 10.0 4.4 - 2.0 5.3 - 1.0 4.4 - Tr 0.4 6.4 - + 0.1 6.6 Tr +

I

W

LD6o with 95% Confidence Limitst

2,400 (1,600-4,000) hr at 37 C: + 9.0 (5.7-14.0) 5,000 (3,000-8,800) 340 (220-550) (17-38) 24 Tr (12-33) 20 + 9.0 (5.7-14.0) +

* Assayed by gel-diffusion; - = not detected; Tr = trace. t Mouse intraperitoneal virulence titrations.

TABLE 3 Effect of 5-fluorodeoxyuridine on the restoration of virulence 5-Fluor- Viable Cell odeox- Count

yuridine gm (X

per ml)

Antigens*

F|V FI V

Culture grown at 5 C: None 0.8 - 5 C Culture incubated 6 1.0 Tr + None 100 0.9 - _ 1.1 - _ 50 1.2 Tr + 25 12.5 1.2 Tr + 4.5 Tr + 6.2 3.1 1.7 Tr + 1.5 1.0 Tr +

I

W

LD60 with 95% Confidence Limits)t

1,600 (615-4,200) hr at 37 C: + 16 (5.1-50) - 4,500 (1,960-10,400) 150 (41-550) + 100 (37-270) 120 (46-312) + + 88 (52-150) 18 (6.9-47) + 19 (5.4-67) +

The detailed results on the effect of chloramphenicol and 5-fluorodeoxyuridine on the synthesis of "virulence" antigens and restoration of virulence are shown in Tables 2 and 3. The results indicated that 10 Aug per ml of chloramphenicol or 100 ,ug per ml of 5-fluorodeoxyuridine were capable of completely blocking the synthesis of fraction I (Fl) and V and W anti* Antigen assay by gel-diffusion;- = not degens, as well as the restoration of virulence. The slight variations in viable cell counts after 6 hr tected; Tr = trace. of incubation at 37 C from the viable counts t Mouse intraperitoneal virulence titration. obtained before incubation at 37 C were ascribed to incomplete dispersal of cells and the plating directly associated with the synthesis of V methods and not to multiplication or the death antigen and the restoration of virulence. Plate of the cells. counts performed with the cell suspensions did Upon assaying for protein, ribonucleic acid not indicate any significant increase in the (RNA) and deoxyribonucleic acid (DNA) number of viable cells (ca. 109 per ml) during synthesis at 37 C in the absence of antime- incubation at 37 C for 6 hr. The addition of chloramphenicol (10 /Ag/ml) tabolites, net increases in protein and RNA but not in the DNA were detected (Table 4). It completely blocked protein and RNA synthesis, was further demonstrated in this experiment that as well as blocking the synthesis of V antigen the net increase in protein and RNA could be and the restoration of virulence. Although no

658

FUKUI, LAWTON, AND MORTLOCK

[VOL. 81

TABLE 4 Effects of chloramphenicol and 5-fluorodeoxyuridine on the synthesis of cell components and restoration of virulence Protein

Dry Wt (mgml) Tu

DNA

RNA

V

Treatment of 5 C Culture

Nonet Incubated 6 hr at 37 C:t No antimetabolite

1.65

of

g/ml

%

385

23.3

Dywt

gg/mi

% of /if A/lDyw

LDI0* with 95%

Antigen Confidence Limits

Dirywt

540

32.7

180

10.9

-

480

(123-1,872) 2.0

2.30

780

33.9

740

32.2

183

7.9

+

1.56

285

18.3

500

32.1

140

8.9

-

(1.7-5.8) Chloramphenicol (10,ug/ml)

480

(123-1,872) 5-Fluorodeoxyuridine

(100

2.01

535

26.6

472

pg/ml)

23.6

182

9.1

-

700 (333-1,470)

* Mouse intraperitoneal LD5o with 95% confidence limits. t Culture grown at 5 C. I Culture grown at 5 C and incubated at 37 C for 6 hr.

reduction of viable cells was observed when chloramphenicol was added, a slight decrease in dry weight, protein, RNA, and DNA was detected. The inhibition of protein synthesis by chloramphenicol was anticipated in view of many published reports (reviewed by Gale (1959)). The inhibition of RNA synthesis by chloramphenicol was not necessarily unexpected in view of the conflicting reports in the literature concerning this effect. Gale and Folkes (1953) and Wisseman et al. (1954) have reported that chloramphenicol inhibited protein synthesis without affecting nucleic acid synthesis when it was added to bacterial cultures in the log phase of growth, whereas Doudney (1960) reported that the addition of chloramphenicol to a cold temperature synchronized culture of Escherichia coli blocked the synthesis of nucleic acid during a short period of incubation (10 to 20 min) at 37 C. According to Doudney, chloramphenicol could block DNA synthesis when added to cultures just prior to cellular division, whereas the addition of chloramphenicol during a short period of time subsequent to cellular division could block RNA synthesis. Therefore, the basis for the chloramphenicol inhibition of RNA synthesis presented in Table 4 might be similar to the observations reported by Doudney. When 5-fluorodeoxyuridine was added to the test system (100 ,g/ml), protein synthesis was partially suppressed, RNA synthesis was completely blocked, and no increase in DNA was

noted. These effects of 5-fluorodeoxyuridine were originally unexpected, since this analogue had been reported by others (Heidelberger et al., 1957; Bosch, Harbers, and Heidelberger, 1958; Scheiner, Kostelak, and Duschinsky, 1957) to function primarily by blocking the synthesis of DNA by interfering with the methylation of uridine to form thymidine which is an essential component of DNA. However, the effect of 5fluorodeoxyuridine in this system could be explained possibly by assuming that 5-fluorodeoxyuridine was converted to 5-fluorouracil which in turn was capable of blocking the required synthesis of RNA. This implication might be related to the studies by Cohen et al. (1958) who found that 5-fluorodeoxyuridine could be converted by E. coli to 5-fluorouracil deoxyribotide which in turn could be subsequently cleaved to give rise to free 5-fluorouracil. Coupled with the reduction in the synthesis of protein and RNA, no synthesis of V antigen or restoration of virulence was detected in the presence of 5-fluoro-

deoxyuridine. Although mitomycin C had been reported by Shiba et al. (1959) to selectively inhibit the formation of DNA by E. coli, we were not able to show this effect. However, our failure to demonstrate a positive mitomycin C effect may have been because we could not demonstrate synthesis de novo of DNA in the short (6 hr) incubation period at 37 C in which the cells remained in the lag phase of growth.

19611]

SYiNTHESIS OF VIRULENCE BY P. PESTIS

Also it was interesting to find that the analogues for phenylalanine were capable of eliciting a blocking effect against the restoration of virulence. However, the results do not permit us to speculate on its exact mode of action in these preliminary studies. One might speculate that the major function of these analogues was to function as a competitive inhibitor, giving rise to biologically nonfunctional protein containing the analogue. But because of their highly active effect in both chemically defined and complex media, they might also interfere with the nucleic acid metabolism as indicated by Pardee and Prestidge (1956) in E. coli. ACKNOWLEDGMENTS

The authors wish to express their gratitude to R. Powell and John Wiedemann for technical assistance and to M. J. Surgalla for interest and support of this research. SUMMARY

The results of our studies using antimetabolites which have been reported by others to block the synthesis of nucleic acids and proteins, indicated that the deoxyribonucleic acid required for the synthesis of virulence factors was preformed in adequate quantities at 5 C. However, synthesis de novo of ribonucleic acid at 37 C was required for the synthesis of proteins (virulence antigens) which were apparently needed for the expression of maximal virulence by Pasteurella pestis. REFERENCES BOSCH, L., E. HARBERS, AND C. HEIDELBERGER 1958 Studies on fluorinated pyrimidines. V. Effects on nucleic acid metabolism in vitro. Cancer Research, 18, 335-343. BURROWS, T. W. AND G. A. BACON 1956a The basis of virulence in Pasteurella pestis. The development of resistance to phagocytosis in vitro. Brit. J. Exptl. Pathol., 37, 286-299. BURROWS, T. W., AND G. A. BACON 1956b The basis of virulence in Pasteurella pestis. An antigen determining virulence. Brit. J. Exptl. Pathol., 37, 481-493. BURROWS, T. W, AND G. A. BACON 1958 The effects of loss of different virulence determinants on the virulence andimmunogenicity of strains of Pasteurella pestis. Brit. J. Exptl. Pathol., 39, 278-291.

659

BURTON, K. 1956 A study of the conditions and nechanism of diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem. J., 62, 315-323. CERIOTTI, G. 1955 Determination of nucleic acids in animal tissues. J. Biol. Chem., 214, 59-70. COHEN, S. S., J. G. FLAKS, H. D. BARNER, M. R. LOEB, AND J. LICHTENSTEIN 1958 The mode of action of 5-fluorouracil and its derivatives. Proc. Acad. Sci., 44, 1004-1012. DOUDNEY, C. 0. 1960 Inhibition of nuclei acid synthesis by chloramphenicol in synchronized cultures of Escherichia coli. J. Bacteriol., 79, 122-124. FUKUI, G. M., H. B. NAYLOR, W. D. LAWTON, W. A. JANSSEN, AND M. J. SURGALLA 1959 The effect of temperature on growth and virulence of Pasteurella pestis. Bacteriol. Proc., 1959, 100. FUKUI, G. M., W. D. LAWTON, D. A. HAM, W. A. JANSSEN, AND M. J. SURGALLA 1960 The effect of temperature on the synthesis of virulence factors by Pasteuerlla pestis. Monograph on Biochemical Aspects of Microbial Pathogenicity. Ann. N. Y. Acad. Sci., 88, 11461151. GALE, E. F., AND J. P. FOLKES 1953 The assimilation of amino acids by bacteria. 15. Actions of antibiotics on nucleic acid and protein synthesis. Biochem. J., 53, 493-498. GALE, E. F. 1959 Synthesis and organization in the bacterial cell. Ciba lectures in microbial biochemistry. John Wiley & Sons, Inc., New York. HEIDELBERGER, C., N. K. CHAUDHURI, P. DANNEBERG, D. MOOREN, L. GRIESBACH, R. DUSCHINSKY, R. J. SCHNITZER, E. PLEVEN, AND J. SCHEINER 1957 Fluorinated pyrimidines, a new class of tumor-inhibitory compounds. Nature, 179, 663-666. LAWTON, W. D., G. M. FUKUI, AND M. J. SURGALLA 1960 Studies on the antigens of Pasteurella pestis and Pasteurella pseudotuberculosis. J. Immunol., 84, 475-479. LOWRY, 0. H., M. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL 1951 Protein measurement with the Folin phenol reagent. J. Biol.

Chem., 193, 265-275. NAYLOR, H. B., G. M. FUKUI, AND C. R. McDuFF 1961 Effect of temperature on growth and virulence of Pasteurella pestis. Physical and nutritional requirements for restoration of virulence. J. Bacteriol., 81, 649-655. PARDEE, A. B., AND L. S. PRESTIDGE 1956 The

660

FUKUI, LAWTON, AND MORTLOCK

dependence of nucleic acid synthesis on the presence of amino acids in Escherichia coli. J. Bacteriol., 71, 677-683. SCHEINER, J. M., E. KOSTELAK, AND R. DUSCHINSKY 1957 5-fluoropyrimidines as growth inhibitors of microorganisms. Federation Proc., 16, 242. SCHUTZE, H. 1934 The envelope antigen of B. pestis and its antibody. Brit. J. Exptl. Pathol., 15, 200-206. SHIBA, S., A. TERAWAKI, T. TAGUCHI, AND J.

[VOL. 81

KAWAMATA 1959 Selective inhibition of formation of deoxyribonucleic acid in Escherichia coli by mitomycin C. Nature, 183, 10561057. WESSMAN, G. E., D. J. MILLER, AND M. J. SURGALLA 1958 Toxic effect of glucose on virulent Pasteurella pestis in chemically defined media. J. Bacteriol., 76, 368-375. WISSEMAN, C. L., JR., J. E. SMADEL, F. E. HAHN, AND H. E. Hopps 1954 Mode of action of chloramphenicol. J. Bacteriol., 67, 662-679.