FEMS MicrobiologyLetters 95 ( 1992195-98 ~': 1992 Federation of European Microbioh~gicalSocieties 11378-11197/92/$115.1X1 Published by Elsevier
FEMSLE 1|4979
Nitrosating activity in Escherichia coli C l a i r e M. O ' D o n n c l l a n d Clive E d w a r d s lh,partment t~l"(h,m'tic.~ and Microhudo,~9', The Unit cr,~ity. I.:t erp~d, UK
Received 211March I~lt~2 Accepted 5 Ma.~ It~i2 Key words: Nitrosation; Escherichia coli: Nitrite rcductase: Aerobic growth; Anaerobic growth
I. S U M M A R Y Nitrosation activity was measured in Escherichh~ coli isolates and a range of nitrite rcductase ( n i t ) mutants. Activity was only detected in intact cells and could be inhibited by a n u m b e r of treatments such as sonication and osmotic shock. Aerobically-grown cells had highest nitrosation activity compared to oxygen-limited ones. Inclusion of nitrite in growth media induced high activities of nitrite reductase and for some isolates, nitrosation. Analysis of nir mutants identified two which were unable to nitrosate. This result suggested that N A D H - d e p e n d e n t nitrite reductase was implicated either directly or indirectly in nitrosation.
This has led to the proposal that such species may have a role in some bladder and gastric cancers [3-5]. Bacterial reduction of nitrate to nitrite or increased availability of nitrites is thought to increase the catalysis of available amines to N-nitroso compounds. Such a proposal merits studies for understanding the nitrosation process, especially because of the elevated levels of nitrates and nitrites now detected in some potable waters [4]. Previous work by us has revealed that a n u m b e r of bacterial isolates from the hypoacidic stomach showed a range of nitrosating activities. The rate of catalysis in some enteric isolates increased when nitrite was included in growth media [6]. This paper examines factors that affect nitrosation in Escherichht coil and, in particular, examines the role of nitrite rcductase in nitrosation.
2. I N T R O D U C T I O N Carcinogenic N-nitroso compounds can be produced in vitro by mary bacteria by nitrosation of secondary amines to form nitrosamines [1.2].
('orre.~lnmth'ncc to: C'. Edwards. Department of Genetics and Microbiology. The University. Liverpool. L69 3BX. UK.
3. M A T E R I A L S A N D M E T H O D S 3.1. Organisms attd ctdtural conditions All the strains of l~cllericltia coil used for this work arc listed in Tables l and 2. E. coil I and 2 arc clinical isolates obtained by gastric endoscopy as described by us previously [6]. Originally named
t)fi E. co// BI and B2, we have renamed them as E.
col{ 1 and E. col{ 2 here to avoid confusion. For studies of nitrosation bacteria were grown with shaking at 37°C in a modified tryptone soya broth (Oxoid) in which Lab M yeast extract (0.3% w / v ) replaced soya bean extract. Oxygen-limited growth took place in 500-ml bottles filled with the above medium and incubated without shaking. Cultures were also occasionally grown in the above medium supplemented with KNO 3 to provide a final nitrite concentration of 5 raM. Bacteria were harvested by centrifugation at 4°C and 15000 × g for 10 min after which the pellets were washed once and resuspended in 0.1 M potassium phosphate buffer, pH 7.
3.2. Measurement of nitrosathtg actirities The nitrosation reaction was carried out using the method of Suzuki and Mitsuoka [7] using morpholine as the secondapy amine and the nitrosomorpholine produced was measured using a Thermal Energy Analyser as described previously [6].
3.3. Assay of nitrite reductase act{city Nitrite reductase activity was measured by the method of Cole and Ward [8]. Whole-cell suspensions (approximately 10 mg total protein) were assayed in a total volume of 5 ml that contained 40 mM glucose, 50 mM potassium phosphate, pH 7.11 and 1 mM nitrite. After incubation at 30°C for 5 rain, nitrite concentrations were measured color{metrically in 0.2-ml samples removed at 2-rain
intervals, each of which was mixed with 8.8 ml 1% w / v sulphanilamide in I M HCI and I ml of 0.02% N-I-naphthyl ethylene diamine dihydrochloride. After 20 min at 18-20°C, the absorbance at 540 nm was determined. This was proportional to the amount of nitrite present over the range 1)-1 mM and could be determined by reference to a calibration curve.
4. RESULTS A N D DISCUSSION
4.1. Factors af]'ecting nitrosation The effects of a number of physical and chemical treatments on nitrosation activity of E. col{ B were studied. Heat-killed cells or cells disrupted by son{cation had no nitrosating activity confirming biological as opposed to indirect chemical catalysis. A requirement for intact cells means that nitrosation may be catalysed by a linked enzyme system or rcquircs an intact cytoplasmic membrane. Treatment with the divalent metal chelator EDTA, or osmotically-shocking cells, reduced nitrosating activity to approximately 60711% of untreated control values. This suggests a possible role for periplasmic proteins either directly in catalysis or indirectly, possibly via transport-linked binding proteins. Inclusion of nitrite at 5 mM in the growth medium resulted in a 3-fold increase in nitrosation activity of E. col{ B compared with unamended controls. Decreased oxygen availability was also found to reduce nitrosating activity.
Table 1 Nitrosating activity in E. col{ strains grown aerobicallyor oxygen-limitedwith or without nitrite Nitrite (mM)
Aerobic nitrosation "
Nitrite reductase ~'
O,-limitation Nitrosation 5.2 (3.4) 11.2 (4.5)
E. col{ B
l} 5
8.1 (3.5) 24.9 t2.t))
I).1116(11.111101 {}.1156(0.1141
E. col{ I
0
211.5 (9.2)
I).11tl8 (II.IRI51
1.9 (2. It/)
Nitrite rcductase I).{142({I.{IIS) 11.12{1(11.1141 11.|125(11.11112)
E. col{ 2
5 l) 5
6.9 (8.2) I8.b (8.71 25.5 13.1181
11.040(11.951 11,0116(1).11115) II.IM4(I).11281
4.9 (11.521 6.6 (5. I ) 5,9 ( 1.251
I},11S{I 111,1111 I).111t,}(11.1H14) 11.1171 (I}.11151
Figures in parenthesis are _+SD and data are from at least three separate determination.,.. " Activityexpressedas nmol uitrosomorpholineproduced (rag total cell protein) t. h Activityas /J.mol nitrite reduced (ms total protein) ' rain - i
4.2. The effects of nitrite The effects of nitrite on nitrosating and nitrite reductase activity was studied further in E. coli B and two clinical isolates designated E. coli 1 and 2 respectively. Control experiments showed that nitrite up to 10 mM had no effect on growth rate or yield of bacteria. Bacteria were grown aerobically or oxygen-limited with or without nitrite in the growth medium. The results are shown in Table I. Under shaken (aerobic) or static (oxygen limited) conditions, the presence of nitrite induced higher levels of nitrite reductase compared to the activity of cells grown in its absence. In general, activities for statically grown bacteria in the presence of nitrite showed highest nitrite reductase activity. This observation is probably due to nitrite being utilized as an alternative electron acceptor under these conditions. Interestingly nitrosation activity was always highest in aerobically-grown cultures of all the bacteria. Inclusion of nitrite in growth media markedly increased nitrosation of E. coli B, reduced it for aerobically grown E. coil 1 and stimulated the activity for E. coli 2. No stimulation of nitrosation by nitrite occurred for oxygenlimited E. coil 2. A feature of static growth with or without nitrite appeared to be lower nitrosation activities compared to those seen in aerobically grown cultures. No direct relationship was apparent between nitrite reductase activity and nitrosation. This was confirmed for the two clinical isolates by measuring nitrosation and nitrite reductase activities in cultures grown aerobically in the presence of different nitrite concentrations. The results are shown in Fig. I. Nitrosation by E. coli I increased up to 5 mM but then declined to approximately 8% of the maximum value. Nitrite reductase activity increased with elevated nitrite concentration. In E. coli 2 the trend was marked by a linear loss of nitrosating activity at nitrite concentrations up to approximately 8 mM. In contrast, nitrite reductase activity remained highest at this nitrite concentration. The importance of nitrite in enhancing nitrosation activity in some instances is supported by the work of Calmels et al. [9] who recently re-
O.lO
o, o8
O, 06
0.04
0.0:
0
-'
°"°r
2
O'0211
I
0
2
4
8
/T'
..~
i 4
6
8
Nitrltt' conct'ntration tin,M) Fig. I. Effect of increasing concentrations of nitrite in liquid cultures on nitrite reductase (e) a n d nitrosation ( o ) activity of E. coil I [A) and E. coli 2(B).
ported that rats given nitrosomorpholine plus either nitrite or nitrate together with a nitrosating strain of E. coli exhibited increased intragastric formation of N-nitroso compounds. A similar result has been reported by Rowland et al. [10] who showed that high dietary nitrate intake in man correlated with elevated levels of nitroso compounds in feces. Analysis of nitrosation in a number of nir mutants defective in elements of nitrate, nitrite, acetate and formate metabolism revealed that
98 Table 2 Nitrosating activity of LCBgII0 wild-type E. colt and a number of nitrite reductase defective mutants derived from it. Strain
Mutation Lesion
LCBt~XI Wild-type LCB22 nirR Regulatory gene fi~r NO~ and NO 3 reductases LCB82 nirD NADII-NO; reductase LCB84 nirF NADH-NO, reductase LCB85 hire NADH-NO; reductase LCBI9(I nirG Acetate kinase LCB197 nirlt Regulatory gene for NO, and NO.~ rcductases
Nitrosation
to clarify t h e role o f p r e c u r s o r s u b s t r a t e s in nitrosation by e n t e r i c bacteria.
ACKNOWLEDGEMENTS
15.8 (11.98t I) 11 28.5 (I.461 18.8 (0.7) 1.66 (0.115) 23.2 (2.261
Figures in parenthesis are _+SD and data are typical of at least 3 separate determinations. two o f these, n i r D a n d n i r R , c o n s i s t e n t l y failed to catalyse n i t r o s a t i o n a n d a third m u t a n t n h ' G a n d h a d m u c h r e d u c e d levels o f activity ( T a b l e 2). T h e latter m a y be defective in a n a e r o b i c g l u c o s e m e t a b o l i s m r a t h e r t h a n specifically in nitrite red u c t i o n [11]. T h e a b s e n c e o f n i t r o s a t i n g activity in nirD and nirR suggest that the NADH-depend e n t nitrite r e d u c t a s e is i m p l i c a t e d in n i t r o s a t i o n e i t h e r directly as a nitrite r e d u c i n g e n z y m e or indirectly as a n e l e c t r o n d o n o r for t h e cyt o c h r o m e s o f t h e r e s p i r a t o r y chain. T h e s e r e s u l t s a r e at s o m e v a r i a n c e with t h o s e o f C a l m e l s et al., [2] w h o f o u n d t h a t only m u t a n t n h ' R failed to catalyse n i t r o s a t i o n whilst n i r D w a s similar to t h e p a r e n t a l strain. F u r t h e r work is n o w in p r o g r e s s
W e are grateful to t h e N o r t h W e s t C a n c e r R e s e a r c h F u n d for s u p p o r t .
REFERENCES [I] Calmels. S.. Oshima. II.. Vinenl. P., Guunol. A.M. and Bartsch, tt. 11985) Carcinogen h. 911-915. [2] Calmels. S.. Oshima. II. and Bartsch, H. (1988)J. Gen. Microbiol. 134. 221-226. [3] Mirvish, S.S. (1983)J, Nat. Cancer Inst. 71,629-647. [4] Bockler. R., Meyer, H. and Schlag, P. (1983) J. Cancer Res. Clin. Oncol. 10, 502-66. [5] El-aaser. A,. EI-merzabari, M.. Higg, N. and EI-habet. A. (1982) Tumori 69. 23-28. [6] O'Donnell, C.M.. Edwards. C. and Ware, J. (1988) FEMS Mierobiol. Lett. 51. 193-198. [7] Suzuki, S. and Mitsuoka, T. (1984I In: N-nitroso compounds: occurrence, biological effects and relevance to human cancer, (O'Neill. I.K., yon Borstel, R.C., Miller. C.T., Long, J. and Bartsu. H., Eds.), pp. 275-282. IARC Scientific Publications No. 57, Lyon. [8] Cole. J.A. and Ward. F.B. (19731 J. Gen. Mierobiol. 76, 21-29. [9] Calmels. S.. Bereziat, J.C.. Oshima. t1. and Bartsch. H. (1991) Carcinogen 12. 435-439. [10] Rowland, I.R., Granli. T., Bockman. O.C., Key, P.E. and Massey, R.C. (19911 Carcinogen 12. 1395-14111. [ I I ] Macdonald. H.. Pope, N.R. and Cole, J.A. (1985) J. Gen. Microbiol. 13I. 2771-2782.
FEMSLE 1|4979
Nitrosating activity in Escherichia coli C l a i r e M. O ' D o n n c l l a n d Clive E d w a r d s lh,partment t~l"(h,m'tic.~ and Microhudo,~9', The Unit cr,~ity. I.:t erp~d, UK
Received 211March I~lt~2 Accepted 5 Ma.~ It~i2 Key words: Nitrosation; Escherichia coli: Nitrite rcductase: Aerobic growth; Anaerobic growth
I. S U M M A R Y Nitrosation activity was measured in Escherichh~ coli isolates and a range of nitrite rcductase ( n i t ) mutants. Activity was only detected in intact cells and could be inhibited by a n u m b e r of treatments such as sonication and osmotic shock. Aerobically-grown cells had highest nitrosation activity compared to oxygen-limited ones. Inclusion of nitrite in growth media induced high activities of nitrite reductase and for some isolates, nitrosation. Analysis of nir mutants identified two which were unable to nitrosate. This result suggested that N A D H - d e p e n d e n t nitrite reductase was implicated either directly or indirectly in nitrosation.
This has led to the proposal that such species may have a role in some bladder and gastric cancers [3-5]. Bacterial reduction of nitrate to nitrite or increased availability of nitrites is thought to increase the catalysis of available amines to N-nitroso compounds. Such a proposal merits studies for understanding the nitrosation process, especially because of the elevated levels of nitrates and nitrites now detected in some potable waters [4]. Previous work by us has revealed that a n u m b e r of bacterial isolates from the hypoacidic stomach showed a range of nitrosating activities. The rate of catalysis in some enteric isolates increased when nitrite was included in growth media [6]. This paper examines factors that affect nitrosation in Escherichht coil and, in particular, examines the role of nitrite rcductase in nitrosation.
2. I N T R O D U C T I O N Carcinogenic N-nitroso compounds can be produced in vitro by mary bacteria by nitrosation of secondary amines to form nitrosamines [1.2].
('orre.~lnmth'ncc to: C'. Edwards. Department of Genetics and Microbiology. The University. Liverpool. L69 3BX. UK.
3. M A T E R I A L S A N D M E T H O D S 3.1. Organisms attd ctdtural conditions All the strains of l~cllericltia coil used for this work arc listed in Tables l and 2. E. coil I and 2 arc clinical isolates obtained by gastric endoscopy as described by us previously [6]. Originally named
t)fi E. co// BI and B2, we have renamed them as E.
col{ 1 and E. col{ 2 here to avoid confusion. For studies of nitrosation bacteria were grown with shaking at 37°C in a modified tryptone soya broth (Oxoid) in which Lab M yeast extract (0.3% w / v ) replaced soya bean extract. Oxygen-limited growth took place in 500-ml bottles filled with the above medium and incubated without shaking. Cultures were also occasionally grown in the above medium supplemented with KNO 3 to provide a final nitrite concentration of 5 raM. Bacteria were harvested by centrifugation at 4°C and 15000 × g for 10 min after which the pellets were washed once and resuspended in 0.1 M potassium phosphate buffer, pH 7.
3.2. Measurement of nitrosathtg actirities The nitrosation reaction was carried out using the method of Suzuki and Mitsuoka [7] using morpholine as the secondapy amine and the nitrosomorpholine produced was measured using a Thermal Energy Analyser as described previously [6].
3.3. Assay of nitrite reductase act{city Nitrite reductase activity was measured by the method of Cole and Ward [8]. Whole-cell suspensions (approximately 10 mg total protein) were assayed in a total volume of 5 ml that contained 40 mM glucose, 50 mM potassium phosphate, pH 7.11 and 1 mM nitrite. After incubation at 30°C for 5 rain, nitrite concentrations were measured color{metrically in 0.2-ml samples removed at 2-rain
intervals, each of which was mixed with 8.8 ml 1% w / v sulphanilamide in I M HCI and I ml of 0.02% N-I-naphthyl ethylene diamine dihydrochloride. After 20 min at 18-20°C, the absorbance at 540 nm was determined. This was proportional to the amount of nitrite present over the range 1)-1 mM and could be determined by reference to a calibration curve.
4. RESULTS A N D DISCUSSION
4.1. Factors af]'ecting nitrosation The effects of a number of physical and chemical treatments on nitrosation activity of E. col{ B were studied. Heat-killed cells or cells disrupted by son{cation had no nitrosating activity confirming biological as opposed to indirect chemical catalysis. A requirement for intact cells means that nitrosation may be catalysed by a linked enzyme system or rcquircs an intact cytoplasmic membrane. Treatment with the divalent metal chelator EDTA, or osmotically-shocking cells, reduced nitrosating activity to approximately 60711% of untreated control values. This suggests a possible role for periplasmic proteins either directly in catalysis or indirectly, possibly via transport-linked binding proteins. Inclusion of nitrite at 5 mM in the growth medium resulted in a 3-fold increase in nitrosation activity of E. col{ B compared with unamended controls. Decreased oxygen availability was also found to reduce nitrosating activity.
Table 1 Nitrosating activity in E. col{ strains grown aerobicallyor oxygen-limitedwith or without nitrite Nitrite (mM)
Aerobic nitrosation "
Nitrite reductase ~'
O,-limitation Nitrosation 5.2 (3.4) 11.2 (4.5)
E. col{ B
l} 5
8.1 (3.5) 24.9 t2.t))
I).1116(11.111101 {}.1156(0.1141
E. col{ I
0
211.5 (9.2)
I).11tl8 (II.IRI51
1.9 (2. It/)
Nitrite rcductase I).{142({I.{IIS) 11.12{1(11.1141 11.|125(11.11112)
E. col{ 2
5 l) 5
6.9 (8.2) I8.b (8.71 25.5 13.1181
11.040(11.951 11,0116(1).11115) II.IM4(I).11281
4.9 (11.521 6.6 (5. I ) 5,9 ( 1.251
I},11S{I 111,1111 I).111t,}(11.1H14) 11.1171 (I}.11151
Figures in parenthesis are _+SD and data are from at least three separate determination.,.. " Activityexpressedas nmol uitrosomorpholineproduced (rag total cell protein) t. h Activityas /J.mol nitrite reduced (ms total protein) ' rain - i
4.2. The effects of nitrite The effects of nitrite on nitrosating and nitrite reductase activity was studied further in E. coli B and two clinical isolates designated E. coli 1 and 2 respectively. Control experiments showed that nitrite up to 10 mM had no effect on growth rate or yield of bacteria. Bacteria were grown aerobically or oxygen-limited with or without nitrite in the growth medium. The results are shown in Table I. Under shaken (aerobic) or static (oxygen limited) conditions, the presence of nitrite induced higher levels of nitrite reductase compared to the activity of cells grown in its absence. In general, activities for statically grown bacteria in the presence of nitrite showed highest nitrite reductase activity. This observation is probably due to nitrite being utilized as an alternative electron acceptor under these conditions. Interestingly nitrosation activity was always highest in aerobically-grown cultures of all the bacteria. Inclusion of nitrite in growth media markedly increased nitrosation of E. coli B, reduced it for aerobically grown E. coil 1 and stimulated the activity for E. coli 2. No stimulation of nitrosation by nitrite occurred for oxygenlimited E. coil 2. A feature of static growth with or without nitrite appeared to be lower nitrosation activities compared to those seen in aerobically grown cultures. No direct relationship was apparent between nitrite reductase activity and nitrosation. This was confirmed for the two clinical isolates by measuring nitrosation and nitrite reductase activities in cultures grown aerobically in the presence of different nitrite concentrations. The results are shown in Fig. I. Nitrosation by E. coli I increased up to 5 mM but then declined to approximately 8% of the maximum value. Nitrite reductase activity increased with elevated nitrite concentration. In E. coli 2 the trend was marked by a linear loss of nitrosating activity at nitrite concentrations up to approximately 8 mM. In contrast, nitrite reductase activity remained highest at this nitrite concentration. The importance of nitrite in enhancing nitrosation activity in some instances is supported by the work of Calmels et al. [9] who recently re-
O.lO
o, o8
O, 06
0.04
0.0:
0
-'
°"°r
2
O'0211
I
0
2
4
8
/T'
..~
i 4
6
8
Nitrltt' conct'ntration tin,M) Fig. I. Effect of increasing concentrations of nitrite in liquid cultures on nitrite reductase (e) a n d nitrosation ( o ) activity of E. coil I [A) and E. coli 2(B).
ported that rats given nitrosomorpholine plus either nitrite or nitrate together with a nitrosating strain of E. coli exhibited increased intragastric formation of N-nitroso compounds. A similar result has been reported by Rowland et al. [10] who showed that high dietary nitrate intake in man correlated with elevated levels of nitroso compounds in feces. Analysis of nitrosation in a number of nir mutants defective in elements of nitrate, nitrite, acetate and formate metabolism revealed that
98 Table 2 Nitrosating activity of LCBgII0 wild-type E. colt and a number of nitrite reductase defective mutants derived from it. Strain
Mutation Lesion
LCBt~XI Wild-type LCB22 nirR Regulatory gene fi~r NO~ and NO 3 reductases LCB82 nirD NADII-NO; reductase LCB84 nirF NADH-NO, reductase LCB85 hire NADH-NO; reductase LCBI9(I nirG Acetate kinase LCB197 nirlt Regulatory gene for NO, and NO.~ rcductases
Nitrosation
to clarify t h e role o f p r e c u r s o r s u b s t r a t e s in nitrosation by e n t e r i c bacteria.
ACKNOWLEDGEMENTS
15.8 (11.98t I) 11 28.5 (I.461 18.8 (0.7) 1.66 (0.115) 23.2 (2.261
Figures in parenthesis are _+SD and data are typical of at least 3 separate determinations. two o f these, n i r D a n d n i r R , c o n s i s t e n t l y failed to catalyse n i t r o s a t i o n a n d a third m u t a n t n h ' G a n d h a d m u c h r e d u c e d levels o f activity ( T a b l e 2). T h e latter m a y be defective in a n a e r o b i c g l u c o s e m e t a b o l i s m r a t h e r t h a n specifically in nitrite red u c t i o n [11]. T h e a b s e n c e o f n i t r o s a t i n g activity in nirD and nirR suggest that the NADH-depend e n t nitrite r e d u c t a s e is i m p l i c a t e d in n i t r o s a t i o n e i t h e r directly as a nitrite r e d u c i n g e n z y m e or indirectly as a n e l e c t r o n d o n o r for t h e cyt o c h r o m e s o f t h e r e s p i r a t o r y chain. T h e s e r e s u l t s a r e at s o m e v a r i a n c e with t h o s e o f C a l m e l s et al., [2] w h o f o u n d t h a t only m u t a n t n h ' R failed to catalyse n i t r o s a t i o n whilst n i r D w a s similar to t h e p a r e n t a l strain. F u r t h e r work is n o w in p r o g r e s s
W e are grateful to t h e N o r t h W e s t C a n c e r R e s e a r c h F u n d for s u p p o r t .
REFERENCES [I] Calmels. S.. Oshima. II.. Vinenl. P., Guunol. A.M. and Bartsch, tt. 11985) Carcinogen h. 911-915. [2] Calmels. S.. Oshima. II. and Bartsch, H. (1988)J. Gen. Microbiol. 134. 221-226. [3] Mirvish, S.S. (1983)J, Nat. Cancer Inst. 71,629-647. [4] Bockler. R., Meyer, H. and Schlag, P. (1983) J. Cancer Res. Clin. Oncol. 10, 502-66. [5] El-aaser. A,. EI-merzabari, M.. Higg, N. and EI-habet. A. (1982) Tumori 69. 23-28. [6] O'Donnell, C.M.. Edwards. C. and Ware, J. (1988) FEMS Mierobiol. Lett. 51. 193-198. [7] Suzuki, S. and Mitsuoka, T. (1984I In: N-nitroso compounds: occurrence, biological effects and relevance to human cancer, (O'Neill. I.K., yon Borstel, R.C., Miller. C.T., Long, J. and Bartsu. H., Eds.), pp. 275-282. IARC Scientific Publications No. 57, Lyon. [8] Cole. J.A. and Ward. F.B. (19731 J. Gen. Mierobiol. 76, 21-29. [9] Calmels. S.. Bereziat, J.C.. Oshima. t1. and Bartsch. H. (1991) Carcinogen 12. 435-439. [10] Rowland, I.R., Granli. T., Bockman. O.C., Key, P.E. and Massey, R.C. (19911 Carcinogen 12. 1395-14111. [ I I ] Macdonald. H.. Pope, N.R. and Cole, J.A. (1985) J. Gen. Microbiol. 13I. 2771-2782.
