Chelation Effects on Azotobacter Cells and Cysts

Transcription

1 JOURNAL of BACrERIOLOGY, Jan., 1966 Copyright 1966 American Society for Microbiology Vol. 91, No. 1 Printed in U.S.A. Chelation Effects on Azotobacter Cells and Cysts MILLICENT C. GOLDSCHMIDT' AND ORVILLE WYSS Department ofmicrobiology, University of Texas, Austin, Texas Received for publication 27 August 1965 ABSTRACT GOLDSCHMIDT, MILLICENT C. (University of Texas, Austin), AND ORVILLE WYSS. Chelation effects on Azotobacter cells and cysts. J. Bacteriol. 91: Ethylenediaminetetraacetate (EDTA) is very toxic to Azotobacter in the presence of nitrogen compounds that form complexes with it. This appears to be due to stronger chelation of certain metal ions by the complex. When such complexes of EDTA and nitrogen compound are absent, Azotobacter cysts can be ruptured by chelation without being killed. The lethal action as well as the cyst rupture is modified by the presence of salts. When some strains of Azotobacter are grown MATERIALS AND METHODS on a butanol-salts medium, the vegetative cells A. vinlandli strain was cultured on a modified Burk's nitrogen-free mineral salts medium (2) round up and eventually form a layered outer coat which is separated from the central body by adjusted to ph 7.6 and containing per liter: K2HP04, an inner space filled with polysaccharide material. 0.8 g; KH2PO4, 0.2 g; MgS04-7H20, 0.2 g; CaS04. During normal germination of cysts, the central 2H20, 0.1 g; FeC12-4H20, g; NaMoO4 2H20, body elongates and pushes through the cyst coat g; and MnCl2-4H20, 0.01 g. To produce cysts, This occurs 4 to 8 hr after cysts are placed in a 0.2% butanol and 2.5% agar were added to this growth medium. The cysts can be artificially basal medium, and 100% encystment resulted in 4 ruptured by the addition of ethylenediaminetetraacetic acid (EDTA). Within 2 days. The vegetative cells used in these experiments were from a 24-hr culture grown on the same medium. min, the cyst Survival was determined by plate counts on agar coat breaks, freeing the still rounded central plates in which was poured a thin underlayer of body. Socolofsky and Wyss (9) found that cysts Burk's salts medium plus 0.5% casein hydrolysate; ruptured with EDTA in the presence of tris- after this hardened, a thick overlayer of the usual (hydroxymethyl)aminomethane (Tris) buffer plating medium of 0.5% sucrose in Burk's salts was failed to grow in Burk's medium, and the addition introduced. The sucrose and butanol were sterilized of various enrichments did not remedy the defect. separately and added aseptically to the rest of the Cysts ruptured in a medium of high osmotic medium. Azotobacter cells or cysts were scraped from the agar, resuspended in the various diluents, pressure, such as 0.5 M sucrose or lactose, maintained a more vigorous oxidation of ethyl alcohol were resuspended at a level of about 107 cells per and washed three times by centrifugation. The cells and butanol in the presence of MgSO4, but the milliliter. The chelating agents and the Tris buffer released central bodies still failed to grow. It now were dissolved in distilled water and the ph was appears that this was not due to an unusual adjusted to 8.0. Cyst rupture was determined by phase-contrast microscopy (9). fragility of the central body, but rather to the unfortunate choice of Tris buffer, to which Azotobacter cells are surprisingly sensitive in the RESULTS presence of EDTA. The buffer is used often with Cell death and cyst rupture. When Repaske's Azotobacter cells, and no, previous evidence of (8) EDTA-Tris mixture (25 mnim Tris and 0.25 toxicity has been noticed. EDTA has also been mm EDTA) was applied to vegetative cells, less used widely in nitrogen-free Azotobacter media. than 0.1 % of the cells survived an exposure of 4 This lethal effect of the EDTA-Tris combination mi at room temperature (Table 1). Yet vegetative cells were not killed when exposed to EDTA was investigated. in water, saline, or Burk's medium. Cysts could ' Present address: Baylor University Medical be ruptured by EDTA in distilled water or in the School, Houston, Tex. presence of Tris. When salts were present, the 120

2 VOL. 91, 1966 CHELATION EFFECTS ON AZOTOBACTER CELLS AND CYSTS TABLE 1. Survival of Azotobacter vinlandii after 4 min of exposure to 0.25 mmtedta* respiration of 24-hr Azotobacter cultures can be Cysts reduced markedly by aerating in saline, washed Diluent cells Vegetative survival St',ra Sr vegetative cells were shaken for 4 hr before exposure to the EDTA-Tris system. Essentially no Rupture vival killing resulted after the usual 4-min exposure. % % %o But, as shown in Fig. 1, the major effect of aerating in saline was to delay the time for Water Burk's medium... 90% NaCl, 0.15 M Tris, 25 mm NaCl + Trist Tris control (no EDTA) * Vegetative cells from a 24-hr culture or cysts from a 4-day culture were washed twice before suspending in diluent at ph 8.0. t Tris was added to the cells 10 sec before the EDTA. cysts failed to rupture and were still viable. Only in the presence of Tris was there extensive killing. Saline solution (0.15 M NaCi) prevented cyst rupture, as did the buffer salts found in Burk's medium. The saline also interfered slightly with the killing of both vegetative cells and cysts. Dipicolinic acid (DPA). DPA is a chelating agent absent from cysts, but present in bacterial endospores. Like other chelating agents, it will also rupture cysts, but, unlike EDTA, it is not germicidal even in the presence of Tris buffer. Table 2 compares the rupturing and the survival of cysts in DPA and EDTA in the presence of Tris at various ph values. Although most of the cysts were ruptured by either chelating agent, only the EDTA-Tris combination showed extensive killing. Pre-exposure to EDTA or Tris. Repaske (8) found that EDTA permanently effected some change in Azotobacter which could not be altered by washing. That is, once cells had been exposed to EDTA, the chelating agent was no longer required as an adjunct to lysis by lysozyme, despite an intervening washing procedure. We confirmed this with vegetative cells and obtained similar results with the Azotobacter cysts. Only 5% survival was found when cysts, harvested in EDTA, were washed three times with water and then exposed to Tris. On the other hand, when cells were harvested in Tris, washed once in water, and then treated with EDTA, 2% survival was observed. However, two additional intervening washes yielded 72% survival upon exposure to EDTA. The effect of pre-exposure to Tris, unlike that to EDTA, eventually can be washed out. Ion effects. When testing the time course of the EDTA-Tris killing, it was observed that "starved" or metabolically less active cells sometimes gave anomalous results. Since the high endogenous lethality by about 20 to 25 min. When cysts were aerated similarly, a delay in killing resulted, but cysts did not rupture until after essentially all the cells were dead. When either cells or cysts were shaken for several hours in Tris rather than in saline, there was no delaying effect on cyst rupture or on the lethality of EDTA for vegetative cells or cysts. The exposure to saline appears to have surrounded the cells with an ionic barrier which prevented the EDTA-Tris from reacting with the cell until the salt diffused away after the transfer of the cells into Tris. A briefer exposure to 0.15 M saline (the time involved in harvesting and washing) prevents cyst rupture, but does not prevent killing by the EDTA-Tris complex. Hence, the action of EDTA in rupturing cysts can be separated from its lethal effect when combined with Tris. Rupture, but not killing, is inhibited in the presence of 0.15 M NaCl (line 5, Table 1). Samples of treated and untreated cysts were collected from these experiments, quantitatively transferred to graphite cups, and turned over to a competent analytical laboratory for examination for metals. The density of specific spectral lines, obtained by emission spectrography, indicated TABLE 2. Comparison of cyst rupture and survival after 4 min in EDTA and DPA at different ph values* ph DPA EDTA Rupture Survival Rupture Survival t * Cysts were harvested in saline and washed twice in Tris at the indicated ph. Final concentrations: EDTA, 0.25 mm; DPA, 0.25 mm; Tris, 25 mm. t Cysts washed in water and resuspended in water (no Tris).

3 122 GOLDSCHMIDT AND WYSS J. BACTERIOL. > a: (I, z w 0 w MINUTES EXPOSED TO EDTA-TRIS FIG. 1. Survival of Azotobacter vegetative cells in a mixture of mm EDTA and 25 mm Tris buffer at ph 8.0. (1) Cells shaken 4 hr in 0.15 M NaCl before exposure. (2) Freshly harvested cells washed with water. that there was a selective removal of metals by EDTA from Azotobacter cysts that was influenced by the salt concentrations and by the presence of Tris. When cysts in Tris were exposed to EDTA, very large amounts of Ca+, Mg+, and Mn+ were removed (the cysts rupture, and 99% are dead in 4 min). With the EDTA-water system, in which the cysts ruptured but almost 100% of the central bodies survived, only a small amount of Mg++ was removed. No metal removal was detected when cysts were exposed for 4 min to an EDTA-0.15 M NaCl system, and no cyst rupture or killing followed such treatment. When cysts were shaken 4 hr in saline and then exposed to the EDTA-Tris system for 4 min, the Mg++ concentration did not change, but Mn and Ca++ were removed, although to a lesser extent than without saline pretreatments; no cysts ruptured, but 98% were killed. According to Martell and Calvin (7), Mn+ and Ca+ form more stable complexes with EDTA than does Mg++. The Ca++ complex is about twice as stable, and the Mn- complex about seven times more stable, than Mg++. The stability of metal-edta chelate complexes is decreased by even dilute salt solutions (3). Hence, in the presence of NaCl, the weaker complexes should dissociate first. Tris substitutes. Chaberek and Martell (3) showed how ammonia, amino acids, and other amines form homologues with EDTA that are more stable with regard to metallo-complexes than is EDTA alone. Since ammonia and the amines also form complexes with metals, the additional smaller chelate rings which are formed within the parent structure by these donors tend to stabilize the structures of the larger chelate rings. For example, the stability constant of Ca with EDTA is 3.5, compared with 10.5 with an EDTA-(CH2)n homologue. If the mechanism of the EDTA-Tris toxicity involves formation of a homologue which can more readily remove and withhold essential metals from the cell, other nontoxic nitrogen compounds should also be toxic in the presence of EDTA. As can be seen in Table 3, compounds structurally similar to Tris or compounds (such as the primary amines) known to form homologues with EDTA, when used in place of Tris, are toxic for Azotobacter. Ammonia is very effective, as was observed by Chu (Ph.D. Thesis, Univ. of Texas, Austin, 1958) while studying the mineral nutrition of Azotobacter agilis. She found that the organism thrived on N2 in the presence of high concentrations of EDTA, but failed to grow with EDTA in the medium when ammonia was supplied as a source of nitrogen. No other TABLE 3. Survival of Azotobacter cysts and vegetative cells after 4 min of exposure to 0.25 mm EDTA in aqueous solutions of various compounds at ph 8.0 Compound* Survival Vegetative cells None 65.0 Glycerol Tris 0.08 NH4Cl 0.34 Homoserine 0.2 Ethylamine 0.8 Cysts None 85.0 Tris 0.01 Tris (a) 3.5 NH4Cl 1.6 Ethylamine (a) 4.6 Ethanolamine (a) Phenylpropylamine (a) 0.13 Heptylamine (a) 0.34 Benzylamine (a) 1.4 Benzyl alcohol 84.0 Urea 7.8 Alanine 14.0 Asparagine 9.0 Cystine HC Betaine-HCl 2.7 NaNO3 68 * Final concentration of the compounds was 25 mm, except (a), which was 6 mm.

4 VOL. 91, 1966 CHELATION EFFECTS ON AZOTOBACTER CELLS AND CYSTS 123 reports of the biological effects of this system have come to our attention, although McCallum (6a) observed that ammonium ions interfered with quantitative assay procedures inxolving chelation of calcium by EDTA. Reversal of EDTA-Tris inhibition. Repaske (8) was able to inhibit subsequent lysozyme lysis of EDTA-treated A. vinlandii cells by the addition of cobalt and iron salts. Socolofsky and Wyss (9) were unable to reverse the toxic effect of EDTA-Tris by the addition of MgSO4 or by plating the central bodies on a Burk's-casein hydrolysate solid mriedium. We also failed in all attempts to revive cysts or vegetative cells killed by EDTA-Tris in the presence or absence of 0.5 M lactose with such vigorous competitors for the chelate as CoSO4, MgSO4, and FeSO4. This suggests either high stability of the metallo- EDTA-Tris complex, which causes the killing, or very rapid death from the lesions. Divalent cations added to the reaction mixture before addition of the cells can give complete protection. The effects of various concentrations of CaCl2 on cyst rupture and survival are presented in Fig. 2. Apparently the added calcium in the reaction mixture is chelated before the cell metals are attacked, thus exerting a "sparing" action on the organisms. The EDTA-water sys ILL >'I r S - 2_,--0 D D W I 1.0~ ~ ~ ~ ~ ~~ CL~~~~~~~~~~~~~~~~a l0 MOLAR RATIO CALCIUM:EDTA FIG. 2. Effect of increased calcium on killing and rupturing of Azotobacter cysts by 4 min of exposure to 0.25 mm EDTA in water or in 25 mm Tris at ph 8.0. (1) Survival in Tris. (2) Rupture in Tris. (3) Survival in water. (4) Rupture in water. IL ( \ MOLAR RATIO TRIS:EDTA FIG. 3. Killing of Azotobacter cysts by 4-min exposure to I mm EDTA in various concentrations of Tris buffer at ph 8.0. tem is not particularly toxic; Mg+ was the only cation that our spectrographic measurements showed to be removed from the cyst. The cystrupture curve (curve 4) breaks sharply at 1:1 molar ratio of calcium-edta. With the EDTA- Tris system, however, several metals are removed from the cells. There is a 10-fold increase in survival between Ca-EDTA ratios of 0.25 and 0.75 (curve 1). Yet the cyst rupture curve levels off at a molar ratio between 2 and 3 (curve 2). Tris concentration. The effect of the concentration of Tris on survival with 1 mm EDTA is plotted in Fig. 3. There is a threshold in this concentration curve; at molar ratios of Tris- EDTA of less than 0.5 there is no toxicity. As the ratio is increased to 2:1, the killing increases markedly. The toxicity continues to rise even as the ratio approaches 100:1 (the ratio used in most of our experiments). DIscussIoN Azotobacter cysts can be germinated artificially by chelating agents, and the ejected central bodies grow normally unless a combination of EDTA and certain nitrogen compounds is present in the medium. We had previously observed the toxicity for Azotobacter of EDTA when ammonia was the nitrogen source, but not when nitrogen was supplied as nitrate or N2. That Tris and amino acids in combination with EDTA would also be toxic for Azotobacter was surprising to us, since, in our laboratory, Tris is used extensively as a nontoxic buffer, and EDTA is useful in several bacterial media containing a variety of nitrogen compounds. Yet, Azotobacter are killed by the usual concentrations of combinations of these

5 124 GOLDSCHMIDT AND WYSS J. BACTERIOL. substances within 4 min and cannot be revived by the addition of metal ions. The turbidity curves of Socolofsky and Wyss (9) show that the death is not due to a lytic phenomenon somehow uncovered by the Tris. Another chelating agent, dipicolinic acid, exhibits no lethality in the presence of Tris, although it is very effective in rupturing the cyst exine. The presence of NaCl modifies the effects presumably by decreasing the stability of the metallo-chelate complex. It delays the rupture of cysts by decreasing the rate of removal of Mg+ from the structure. It delays the killing of vegetative cells by decreasing the rate of removal of other ions. We cannot tell from these experiments whether the lethality results from removal of the ions Ca++ and Mn+ or from other ions that would accompany their exit, but whose presence or absence was not detected by our spectrographer. The stability constants of the metallochelate with the different ions are affected differently by salt when nitrogen compounds are incorporated into the chelate, and this may account for some differential effects. We cannot exclude the possibility that the EDTA-nitrogen complex penetrates into the organisms more readily and thus brings about the lethality. Such toxic effects might be due to modification of the internal ionic environment of the cell. Tris is itself a chelating agent, and the displacement of line 2 in Fig. 2 shows that the excess Tris reduces the free calcium content and its effectiveness in preventing cyst rupture. The lack of toxicity of EDTA itself requires an explanation, since compounds are present in the Azotobacter cells which should form toxic combinations according to the results in Table 3. Figure 3, however, shows that a distinct threshold concentration of Tris is needed before a toxicity is exhibited. It may be assumed that the amount of free amino groups in the cells will be less than that represented by this threshold. Albert (1) reported that EDTA cannot penetrate bacterial cells, and, consequently, it has no antibacterial activity. Azotobacter, in the presence of nitrogen compounds, is an exception, and Gray and Wilkinson (4) found a strain of Pseudomonas aeruginosa and one of Alcaligenes faecalis that, unlike most other gram-negative organisms, died rapidly in the presence of EDTA. Their unwashed suspensions were more sensitive than washed suspensions, and the EDTA-treated cells released intracellular solutes as if the structural integrity of the cell wall had been impaired. To determine the chemical basis for this unusual sensitivity, Gray and Wilkinson (5) analyzed cell walls and determined that the sensitive organisms differed from usual gram-negative strains principally in their sugar components and in their high content of phosphorus. Leive (6) found an impairment of the permeability barrier in Escherichia coli by EDTA, but the treated cells still grew at a normal rate. It is evident that this problem is complex, with the result depending on the preparation of the cells, the content of the suspending media, and the strain of organism. On the latter point, the authors, together with E.P. Goldschmidt, found male strains of E. coli to be much more sensitive to EDTA lethality than female strains; the results of that study are being prepared for publication. ACKNOWLEDGMENtS This investigation was supported by Public Health Service research grant AI from the National Institute of Allergy and Infectious Diseases. The senior author was a postdoctoral fellow on Public Health Service training grant GM-600 from the National Institutes of Health. LrrERATuRE CITED 1. ALBERT, A Selective toxicity. Metuen and Co., Ltd., London. 2. BURK, D., AND H. LINEWEAVER The influence of fixed nitrogen on Azotobacter. J. Bacteriol. 19: CHABEREK, S., AND A. E. MARTELL Organic sequestering agents. John Wiley and Sons, Inc., New York. 4. GRAY, G. W., AND S. G. WILKINSON The action of ethylenediaminetetra-acetic acid on Pseudomonas aeruginosa. J. Appl. Bacteriol. 28: GRAY, G. W., AND S. G. WELKINSON The effect of ethylenediaminetetra-acetic acid on the cell walls of some gram-negative bacteria. J. Gen. Microbiol. 39: LEIvE, L A nonspecific increase in permeability in Escherichia coli produced by EDTA. Proc. Natl. Acad. Sci. U.S. 53: a. MCCALLUM, J. R Analysis for small amounts of calcium, magnesium, barium, and sulphate using phthalein purple. Can. J. Chem. 34: MARTELL, A. E., AND M. CALVIN Chemistry of the metal chelate compounds. Prentice Hall, Inc., New York. 8. REPASKE, R Lysis of gram-negative organisms and the role of Versene. Biochim. Biophys. Acta 30: SOCOLOFSKY, M. D., AND 0. WYSS Cysts of Azotobacter. J. Bacteriol. 81: