3 7 9 M&U

H^sro

CELL-FREE RECOVERY AND ISOTOPIC IDENTIFICATION

OF CYANIDE DEGRADING ENZYMES FROM

Pseudomonas fluorescens NCIMB 11764

DISSERTATION

Presented to the Graduate Council of the

University of North Texas in Partial

Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

By

Chien-Sao Wang, B.S., M.S.

Denton, Texas

December, 1995

3 7 9 M&U

H^sro

CELL-FREE RECOVERY AND ISOTOPIC IDENTIFICATION

OF CYANIDE DEGRADING ENZYMES FROM

Pseudomonas fluorescens NCIMB 11764

DISSERTATION

Presented to the Graduate Council of the

University of North Texas in Partial

Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

By

Chien-Sao Wang, B.S., M.S.

Denton, Texas

December, 1995

T>M<

Wang, Chien-Sao, Cell-free recovery and isotopic

identification of cvanide degrading enzymes from Pseudomonas

fluorescens NCIMB 11764. Doctor of Philosophy (Molecular

Biology), December, 1995, 94 pp., 15 tables, 10 illustrations,

references, 79 titles.

Cell-free extracts from Pseudomonas fluorescens NCIMB

11764 catalyzed the degradation of cyanide into products that

included C0 2 , formic acid, formamide and ammonia. Cyanide-

degrading activity was localized to cytosolic cell fractions and was

observed at substrate concentrations as high as 100 mM. Two

cyanide degrading activities were identified by: (i) the

determination of reaction products stoichiometries, (ii)

requirements for NADH and oxygen, and (iii) kinetic analysis. The

first activity produced C0 2 and NH3 as reaction products, was

dependent on oxygen and NADH for activity, and displayed an

apparent Km for cyanide of 1.2 mM. The second activity generated

formic acid (and NH3) pfus formamide as reaction products, was

oxygen independent, and had an apparent Km of 12 mM for cyanide.

The first enzymatic activity was identified as cyanide oxygenase

whereas the second activity consists of two enzymes, a cyanide

nitrilase (dihydratase) and putative cyanide hydratase. In addition

to these enzymes, cyanide-grown cells were also induced for

formate dehydrogenase (FDH), providing a means of recycling NADH

utilized by cyanide oxygenase.

Of the various enzymes elaborated by P. fluorescens NCIMB

11764 putative cyanide oxygenase was shown to be physiologically

important for cyanide assimilation. Therefore, further studies were

carried out to identify this enzyme. The stoichiometry of the

reaction indicated that one mole each of oxygen and NADH were

consumed per mole cyanide degraded, which was consistent with

that expected for an oxygenase type enzyme. Isotopic oxygen-18

labelling experiments indicated that one atom of molecular oxygen

was incorporated when 13C-labelled cyanide was converted to 13CC>2

by high-speed supernatants. These findings confirm the involvement

of a cyanide oxygenase in cyanide metabolism and prove that the

enzyme is a monooxygenase. However, one atom of oxygen from

H2 1 8 0 was also incorporated indicating that an additional step

between cyanide and CO2, involving the possible enzymatic

hydrolysis of a transient intermediate, is also involved. Preliminary

work aimed at the purification of cyanide oxidation enzymes

indicated that activity was localized in two fractions following

ultrafiltration, one containing the bulk of protein (fraction F1) and

the other (fraction F2) containing a small dialyzable protein of as

yet unknown nature.

Results from this work show that P. fluorescens NCIMB 11764

is capable of metabolizing cyanide by several different enzymatic

processes. Of these, oxidative conversion to CO2 and NH3 is

physiologically required for cyanide assimilation. Isotopic 1 8 0 -

labelling results now clearly show that a monooxygenase is involved

in initial cyanide degradative attack. Inhibitor studies suggest that

an essential sulfhydryl group may be involved in a mechanism

paralleling that described for nitrilase-type enzymes, but including

oxygen and water as reactants as opposed to water alone.

11

ACKNOWLEDGMENTS

I would like to express my gratitude to Dr. Daniel Kunz for

providing the opportunity to work in his laboratory and sharing his

experience and knowledge with me. I also thank the members of my

doctoral committee, Dr. Gerard O'Donovan for his support and helpful

comments during the study, Dr. Paul Cook, Dr. Manus Donahue and Dr.

Robert Benjamin, for their discussion and review of this manuscript.

A special thanks to Dr. Barney Venables and Dr. Michael

Richmond for their assistance in GC/MS analysis and isotopic

oxygen-18 labelling methods. I owe special thanks to Juan Silva-

Avalos, Olagappan Nagappan and Jui-Lin Chen for their support,

advice and friendship over the past several years.

I am grateful to my wife, Ai-Fang, for her encouragement in

just the right proportions, and finally, I thank my parents,

Hang-Tong and May, for their sustained support and understanding.

111

ABBREVIATIONS

CDA: Cyanide-degrading activity

CNO: Cyanide oxygenase

CND: Cyanide nitrilase (dihydratase)

CNH: Cyanide hydratase

F1: Ultrafiltration (30K molecular weight cut-off) retentate

fraction

F2: Ultrafiltration (30K molecular weight cut-off) filtrate

fraction

FDH: Formate dehydrogenase

HSSs: High-speed supernatants

KCN: Cyanide (CN7HCN in aqueous solution)

NCIMB 11764: National Collection of Marine and Industrial

Bacteria collection number 11764. Torrey, Scotland

IV

TABLES OF CONTENTS

Page

LIST OF TABLES viii

LIST OF ILLUSTRATIONS x

Chapter

1. INTRODUCTION 1

Cyanide in the microbial world Physical properties of cyanide Detoxification of cyanide by microorganisms Microbial assimilation of cyanide as a growth substrate Assimilation of cyanide as a nitrogenous substrate by

Pseudomonas fluorescens NCIMB 11764

2. METHODS 11

Bacterial strains Growth media and cultivation conditions Preparation of cell-extracts Analytical methods Analytical instrumentation Enzyme assays Oxygen uptake measurement Spectrophotometric measurement of NADH oxidation

by HSSs Identification and quantification of 14C-radiolabelled

cyanide metabolites Identification and quantification of 14C-radiolabelled

formate metabolites GC/MS Identification of cyanide conversion metabolite

Methods for measuring oxygen-18 incorporation Resolution of cyanide oxygenase Chemicals

3. RESULTS 27

PART I. RECOVERY OF CYANIDE-DEGRADING ENZYMATIC ACTIVITIES AT THE CELL-FREE LEVEL 27

Centrifugal fractionation of cyanide-degrading activity in cell-free extracts

Kinetics of cyanide conversion by HSS Reaction products and stoichiometries Conversion of formate to C02 by cyanide-induced

cell-free extracts Spectrophotometric identification of cyanide-induced

formate dehydrogenase activity in HSSs Summary of enzymatic activities induced by cyanide Effects of inhibitors on cell-free cyanide degrading

act iv i ty

PART II. REACTION STOICHIOMETRY AND ISOTOPIC IDENTIFICATION OF CYANIDE OXYGENASE 52

Reaction stoichiometry Evidence for the involvement of formate dehydrogenase

in NADH recycling Enzymatic incorporation of isotopic oxygen-18

PART III. PARTIAL RESOLUTION OF CYANIDE OXYGENASE... 72

Recovery of cyanide oxygenase in ammonium sulfate fractions

Requirement of a small dialyzable component for cyanide oxygenase

Characterization of small dialyzable cyanide oxygenase component

v I

4. SUMMARY 80

REFERENCES 86

V I I

LIST OF TABLES

Page Table

1. Reported mechanisms of cyanide degradation by

microorganisms 3

2. Composition of Lennox medium 12

3. Composition of minimal growth medium 13

4. Conversion of cyanide by cell-free fractions of P. fluorescens NCIMB 11764 29

5. Requirements for cell-free conversion of cyanide by

P. fluorescens NCIMB 11764 30

6. Cyanide degrading activity (CDA) kinetic constants 41

7. Products of cyanide metabolism by HSSs of P. fluorescens NCIMB 11764 43

8. Summary of cyanide-induced enzymatic activities in P. fluorescens NCIMB 11764 49

9. Compound test for as effector on cyanide degrading activity 50

10. Reaction stoichiometry for putative cyanide oxygenase measured in HSSs of P. fluorescens 11764 58

11. Effect of formate and NAD+ on putative cyanide oxygenase activity in P. fluorescens NCIMB 11764. . . 60

V I I I

12. Relative ion intensities of isotopic CO2 species detected by mass spectrometry in reaction mixtures supplied P.

fluorescens NCIMB 11764 HSSs and K13CN 68

13. Recovery of cyanide oxygenase in ammonium sulfate fractions 73

14. Recovery of cyanide oxygenase in ultrafiltration fractions 75

15. Effect of F2 treatment on cyanide oxygenase activity . . 79

i v

LIST OF ILLUSTRATIONS

Page Figure

1. Time course of cyanide consumption by high-speed supernatants from P. fluorescens NCIMB 11764 32

2. Effect of increasing protein concentration on cyanide-degrading activity in high-speed supernatants from P. fluorescens NCIMB 11764 34

3. Kinetics of cyanide degradation by high-speed supernatants from P. fluorescens NCIMB 11764 37

4. Lineweaver-Burk plots for enzyme-catalyzed cyanide degradation 40

5. Spectrophotometric detection of cyanide-induced formate dehydrogenase in HSSs from P.fluorescens NCIMB 11764 48

6. Stoichiometry of O2 uptake and cyanide 55

7. Stoichiometry of cyanide-dependent NADH oxidation. . . . 57

8. A metabolic scheme showing the interrelationship of formate dehydrogenase, putative cyanide oxygenase and cyanide nitrilase in cofactor NADH recycling 62

9. Time course of 1 4C0 2 formation by HSSs from P. fluorescens NCIMB 11764 65

10. Effect of F2 concentration on stimulation of cyanide oxygenase of F1 78

x

CHAPTER I

INTRODUCTION

Cyanide in the microbial world

Cyanide arises in the environment by two means. First it is

formed naturally by a number of bacteria, fungi, and plants in a

process known as cyanogenesis (Knowles, 1976), and second, it can

be generated by various industrial practices such as electroplating,

mining and steel processing (Knowles, 1976; Pettet and Ware, 1955;

and Way, 1981). Interest in cyanide research has grown since a

number of industries have sought to use biological methods for

cyanide waste treatment, which may compete effectively with

chemical processes (Knowles, 1976; Knowles and Bunch, 1986). For

this reason and because of its fundamental value, research directed

at understanding the chemical basis of cyanide metabolism at the

enzymatic level was undertaken.

Physical properties of cyanide

Hydrogen cyanide is a colorless weak acid (pKa=9.3), which

boils at 26°C, and is therefore readily volatilized from aqueous

solution. Cyanide (CN'/HCN) is highly reactive and complexes

tightly to metals such as nickel, copper, zinc, iron and gold (Sharp,

1976). Cyanide also reacts reversibly with keto groups to form

cyanohydrin derivatives. As a result of its reactivity, many enzymes

are strongly inhibited by cyanide. It is therefore often used as a

metabolic inhibitor, especially of cytochrome oxidases. Although

inhibitory, some biological systems elaborate an electron transport

system that is cyanide tolerant. This is thought to occur by the

formation of alternative cytochrome oxidase enzymes that do not

respond to the effects of cyanide (Knowles, 1976).

Detoxification of cyanide by microorganisms

Several chemical and physical treatment process have been

pilot plant evaluated either alone or in combination for the

degradation of cyanide compounds from waste waters (Mudder and

Whitlock, 1984). These processes require large quantities of

expensive chemicals resulting in high operation costs. Therefore,

research directed towards the evaluation and development of

biological treatment processes that generally produce low toxicity

effluent simply and cost effectively are of interest.

A number of cyanide-degrading microorganisms have been

reported and various enzymatic routes of metabolism have been

proposed (Knowles, 1976; Knowles, 1988). These are shown in Table

1 and discussed as follows.

a. Cvanide hydratase (EC 4.2.1.66, formamide hydrolyase)

Phytopathogenic fungi have been shown to elaborate cyanide

hydratase, which catalyzes the conversion of cyanide to non-toxic

formamide. For example, both spores and mycelia of Stemphylium

loti and Gloeocercospora sorghi induce when cultivated in the

CO E .52 "c 05 o> I . o o L . o E >* n c o "co u co u O) 0 "D CD "D C CO > * o

CO E CO 'c co JC o 0 E "D 0 u o Q_ CD OC

J0 -Q 03 h-

co 0 O c a) u 0

H— 0 oc CO E CO 'c CO O) v_ o o k .

o E

CO 0 CL E CO X Hi

CO c o o OJ CD a:

"O 0 > O > c CD E >* N c LLf

CM CO a> 2? r: o

I ^

oB ^ N C CO

S> o CO

5

-c a S CD

CD O O CD O CO CD

CM X z o o X A

o CM

X

o X

CD CO CO 2 ® -0 CD >s -SI T-CD cvi T> c ^ £ o o \H

CD h* o>

N h* a>

in CD CD

ir* c a> co o * 8 i= ° ° * " . *

u O ® JQ S > u. CO := o >* 03 W DC

"D CD

co ® o 8-o

^ g CO T3

CO +-*

c JO Q.

I E 3 £ W (0

a) •a

Z o co

C*3

o CO

§>£ O |

2 I .2 g E E

0 + 1

CO o

CM C/5

0 co co c co "D o

00 cvi O LD

o r— CD CO o> a

co co

c <D "D 5 0

LL

°3 _ _ CD CD

1 f j 3 C * CO Q CO CO

5 3 CD O •S! O S

o o

•2 •c o c CD -C O .<0

5

CO 0 "o CD Q. CO

C J-cd a> o o co <0 *Q -O ° 9 S q5

- 2 c Uj O Ul

CO CM

X

CD <!> c .E 0 c to * >% o + Z o X GO.

co o c CO > * o

CD c a> c

jco g CD ^ 2 <0 m 5 -C

4-' O C O J. ^ LU co. co Sii

I s -00 O)

CO Q>

CO 'v. V . co X

CO CO o c •CS jo Q> O) CO 3

8-"§ o a:

c o o CO

X 3 Z T3 4- £ X c O o A O

0 CD

z x o -X e

S ^ S ® CD ^ CD ' a ^ i: O Z B

CD CD

CO 0) c CD CO v_ o > D) C

CO o> O)

CO Q) CO u. a> CD 2

CO Q) C CD

CO c § ^ CO o c 52 «o , O O JJ b $ ^

s © a

d | <o SS 'G O jD z •>: 3 to ^ « OQ

CO I Z +

X o o o X

o CM

X C\l +

o X

0 co CO

CD CO co «#-» CO JL_ *o >*

•E =5 « «

§ § ^ >* o o

co co oo CD

CO o 5 o c

08 CO

co X

CD K

CD

O

CO c CD O CO s O

CO X

+ CM o

o A

CM o + z o X

CD CO CO c 0 CD >s X o 0

7 3 * C CO >*

O

4

presence of cyanide suggesting that its induction has a

detoxification role (Fry and Millar, 1972; Fry and munch, 1975; Nazly

and Knowles, 1981; and Nazly et al., 1983).

b. Rhodanase (EC 2.8.1.1, thiosulphate: cyanide sulphur

transferase).

Rhodanase catalyses the conversion of cyanide to thiocyanate

in the presence of thiosulphate. It is widely distributed in

biological systems having been described in mammalian tissues,

plants and microorganisms (Sorbo, 1953; Silver and Kelly, 1976;

Ryan and Tilton, 1977; and Alexander and Volini, 1987). In addition

to cyanide detoxification, this enzyme has also been proposed to

function in the transfer of sulphur atoms in oxidative metabolism

(Westley, 1981).

c. B -Cvanoa lan ine synthase [EC 4.4.1.9, L-cysteine hydrogen

sulphide-lyase (adding HCN)].

Incorporation of cyanide into cysteine or o-acetylserine by p-

cyanoalanine synthase is thought to represent a major mechanism by

which cyanide is detoxified in plants (Hendrickson and Conn, 1968;

Ressler et al., 1969; Ting and Zschoche, 1970). This enzyme has

been detected in fungi (Strobel, 1967; Castric, 1981) and some

bacteria including Escherichia coli (Dunhill and Fowden, 1965),

Cromobacterium. violaceum (Brysk et al., 1969; Rodgers, 1981; and

Macadam and Knowles, 1984), and Enterobacter species (Sakai et

a!., 1981; and Yanese et al., 1982).

d. Nitroaenase (EC 1.18.6.1, reduced ferrodoxin: dinitrogen

oxidoreductase [ATP-hydrolysing] ).

Although not explicitly demonstrated in vivo, organisms that

elaborate nitrogenase have also been proposed to be capable of

cyanide detoxification. This is because it has long been recognized

that this enzyme, in addition to catalyzing the reduction of

molecular nitrogen, can also reduce various substrate analogues of

nitrogen such as cyanide (Li et at., 1982). In this process, cyanide

is thought to be reduced to methane and ammonia (a six electron

process), methylamine (four electrons) and possibly also

formaldehyde and ammonia via hydrolysis of a two electron

intermediate. The reduction of cyanide by nitrogenase has been

described for several organisms including Rhodospirillus gelatinosa

(Harris et ai, 1987).

e. Cvanide nitrilase (Cyanide dihydratase)

Cyanide dihydratase catalyses the conversion of cyanide to

formate and ammonia. This enzyme has been detected in the Gram-

positive bacterium Bacillus pumilus (Meyers et ai, 1991; Meyers

et at., 1993) and the Gram-negative organism Alcaligenes

xylosoxidans subsp. denitrificans (Ingvorsen et ai, 1991).

f. Cvanide oxygenase

The oxidative conversion of cyanide to C02 and ammonia by P.

fluorescens NCIMB 11764 was first reported by Harris and Knowles

(1983a), who proposed that this transformation was mediated by an

oxygenase-type enzyme. The importance of this enzyme to cyanide

conversion represents the major focus of this dissertation.

Microbial assimilation of cyanide as a growth substrate

Since cyanide contains two of the essential elements of life it

might be hypothesized that microorganisms could utilize it for

growth. But the toxicity of cyanide presents a problem when

isolating bacteria capable of using it as a substrate for growth.

Thus, a cyanide concentration high enough to serve as a source of

carbon and energy is likely to be too toxic to allow growth and there

have been no reports of organisms that can use cyanide as a

nutritional source of carbon. But several microorganisms capable of

using cyanide as a nitrogen source have been described. These

include, Aspergillus niger (Ivanoff and Zwetkoff, 1936; Rangaswami

et al., 1963); a unidentified OSram-positive, filamentous bacterium

(Ware and Painter, 1955); Pseudomonas fluorescens NCIMB 11764

(Harris and Knowles, 1983a); a pseudomonad reported by White et

ai (1988); seven pseudomonad species and three Klebsiella species

reported by Kunz and coworkers (Silva-Avalos et al., 1990) and

Alcaligenes xylosoxidans subsp. denitrificans reported by

Ingvorsen et al., 1991.

Assimilation of cyanide as a nitrogenous substrate by

Pseudomonas fluorescens NCIMB 11764

P. fluorescens NCIMB 11764 was chosen as the focus of this

study because of its ability to degrade cyanide and use the nitrogen

component as a source of nutrition. Therefore, this organism was

routinely cultivated in a medium where cyanide (KCN) was supplied

as the sole nitrogen source (Harris and Knowles, 1983a; Kunz et ai,

1992). Previous work indicated that growth on cyanide occurs by

conversion to ammonia, which is presumed to be assimilated via

established pathways involving glutamine synthetase (EC

6.3.1.2)/glutamine synthase (EC 1.4.1.13) and glutamate

dehydrogenase (EC 1.4.1.4) enzyme (Neidhardt et ai, 1990).

However, the enzymatic basis of cyanide conversion to NH3 was

poorly understood. Therefore, a major goal of this dissertation was

to learn more about how cyanide (CN* or HCN) was metabolized at the

enzymatic level.

Initial studies with P. fluorescens NCIMB 11764 by Harris and

Knowles (1983a) led them to suggest that enzymatic conversion of

cyanide was oxygenase (CNO) mediated. This suggestion was based

on the following observations: (i) washed cell suspensions and cell-

extracts did not catalyze cyanide conversion under anaerobic

conditions, (ii) cyanide turnover by cell-free preparations appeared

to be correlated with simultaneous oxygen uptake and NADH

oxidation, and (iii) C02 appeared to be formed (along with NH3) as a

metabolic product. These investigators proposed that either a

monooxygenase (1) or a dioxygenase (2) as shown in Scheme 1 acted

on cyanide. The former was thought to produce cyanate, thus

requiring an additional enzyme for its breakdown. This enzyme, it

was proposed, might be a cyanase (EC 3.5.5.3). However, subsequent

studies showed that while 11764 elaborated cyanase after growth

8

v+ HCN + 0 2 + NADH + H+ > HOCN + H20 + NAD'

HOCN + H20 > C02 + NH3 (1)

HCN + 0 2 + NADH + H+ > C02 + NH3 + NAD+ (2)

Scheme 1

on cyanate (Kunz and Nagappan, 1989), it was not coinduced along

with cyanide oxygenase when cells were grown on cyanide.

Conversely, growth on cyanate induced cyanase activity but not

cyanide oxygenase. In addition, a mutant strain unable to grow on

cyanate (phenotypic designation Cnt"), was still able to utilize

cyanide (phenotypic designation Cn+) as a nitrogen source (Nagappan,

1992). These results showed that cyanate and the enzyme cyanase

were not involved in cyanide utilization by P. fluoresceins NCIMB

11764 and therefore, it was proposed that the enzyme responsible

for cyanide conversion to C02 (and NH3) was a dioxygenase.

However, additional investigations by Kunz et a/., (1992) showed

that intact cells of NCIMB 11764 produced not only CO2 as a reaction

product, but also formate and formamide. Since formation of the

latter two products was observed under both aerobic and anaerobic

conditions, it was postulated that other enzymes in addition to a

putative oxygenase were capable of attacking cyanide. Scheme 2

shows three different enzymes that were proposed to be involved in

cyanide metabolism in P. fluorescens NCIMB 11764. These include

in addition to a cyanide oxygenase (CNO), a cyanide nitrilase-type

h o o 2 h + n h 3 ^ -

Assimilation

2H20

JL. CND

HCN

Ofe NADH + H+ NAD+

H>0

I™3

hcosh2

hH3 + C02

Assimilation

Scheme 2

enzyme (dihydratase) (CND) (giving formate and ammonia) and a

cyanide hydratase (CNH) generating formamide. Evidence that the

latter transformation was separate from that leading to formate

was based on the fact that formamide was shown not to undergo

further degradation (Kunz et al., 1992).

The hypothesis that this organism might produce a cyanide

nitrilase was consistent with evidence for related enzymes in other

bacteria. For example, Ingvorsen et al., (1991) showed that this

type of enzyme was made in Alcaligenes xylosoxidans subsp.

denitrificans strain DF3. Complete conversion of cyanide to

formate and ammonia under either aerobic or anaerobic conditions

was demonstrated without the apparent involvement of formamide

as an intermediate. A similar enzyme is thought to be responsible

for hydrolysis of cyanide to formate and ammonia in a pseudomonad

isolated from industrial wastewater by White et a/., (1988).

10

Cyanide degradation by this organism was also shown not to require

oxygen. With regards to cyanide hydratase, this enzyme has thus far

been described in phytopathogenic fungi only (e.g., Stemphylium loti

[Fry and Millar, 1972], and Gloeocercospora sorghi [Nazly et a!.,

1983]), making its possible occurrence in NCIMB 11764 novel. In

addition, no organism other than NCIMB 11764 has ever been

described that contains a putative cyanide oxygenase. Therefore,

biochemical studies were initiated to further identify and

characterize these enzymes at the cell-free level. In particular,

this dissertation had the following objectives: (i) recover cyanide

degrading enzyme activities at the cell-free level, (ii) verify the

actual involvement of cyanide oxygenase by isotopic oxygen-18

labelling experiments, and (iii) initiate the purification of the latter

enzyme.

CHAPTER II

METHODS

Bacterial strains

All experiments were conducted with P. fluorescens NCIMB

11764 which was obtained from the National Collection of Marine

and Industrial Bacteria, Torrey, Scotland. This bacterium was

isolated by R. E. Harris and C. J. Knowles in the United Kingdom

(Harris and Knowles, 1983a). For long-term storage this and other

strains were maintained in 10% dimethyl sulphoxide at -70°C. Cells

were routinely subcultured on Lennox agar (see Table 2 for formula)

plates every 4-6 weeks.

Growth media and cultivation conditions

The formula for minimal medium (MM) used for the growth of

cells is shown in Table 3. The procedure used for routine fed-batch

cultivation of cells on cyanide as a nitrogen source (cyanide-grown

cells) was the same as that described by Kunz et a!., 1992. For

this purpose, the inoculum was cultivated in 200 ml of minimal

medium containing glucose (20 mM) as the carbon source and 1 mM

NH4CI as the source of nitrogen. After 48 hours incubation at 30°C,

the entire 200 ml-inoculum was added to 2 liters of glucose

supplied MM (20 mM) with KCN (0.25 mM) serving as the nitrogen

11

12

Table 2. Composition of Lennox medium {Lennox, 1955)

Lennox broth medium

Bacto tryptone 10.0 g

Yeast-extract 5.0 g

NaCI 5.0 g

Distilled water 1,000 ml

pH 6.8, sterilize at 121°C for 20 min.

For Lennox agar, 20 g of Bacto-agar (Difco) was added.

13

Table 3. Composition of minimal growth medium (Kunz et al., 1981)

Buffer phosphate (P10X^

KH2P04 91.0 g

NaOH 16.8 g

Distilled water to complete 1,000 m I

pH 7.0

R-salts m-200Xl

10 % MgS04 • 7H20 400 ml

1 % FeS04 • 7H20 100 ml

concentrated HCI 2 ml

Mix and sterilize at 121°C for 20 min.

Minimal medium compositions

P1X (sterile) 100 ml

Glucose 1 M (sterile) 2 ml

R-salts 200 X (sterile) 0.5 ml

Glucose and R-salts were added aseptically to sterilized P1X. For

minimal agar, 2 % Bacto-agar (Difco) was added.

14

source. After 24 hours incubation a second addition of KCN (0.25

mM) was made followed by additions of 0.125 mM KCN each at 48 and

60 hours (total cultivation time was 72 hours with KCN additions

totaling 0.75 mM). The cells were then harvested by centrifugation

at 12,000 x g at 4°C and washed twice in 50 mM Na2HP04-KH2PC>4

buffer (pH 7.0) (SP buffer) and used immediately, or were stored at -

70°C. Cells induced for cyanide-degrading activity (CDA) were also

acquired through a service contract with the Fermentation Pilot

Plant in the Department of Biochemistry, University of Wisconsin,

Madison, Wl. In this procedure cells induced for cyanide-degrading

activity were obtained by growing cells starved for nitrogen in

stationary phase as described by Kunz et al., (1994). The large-

scale growth of cells was performed in 360 liters of glucose

supplied MM (20 mM) and NH4CI (1 mM) in a 400 liter fermenter.

After 24 hours the cells were induced by several additions of KCN

(0.1 to 0.2 mM) and cells were harvested 24 hours later by

continuous flow centrifugation. The yield was 300 g [wet weight]

cells, which were stored at -70°C until use.

Preparation of cell-extracts

Cell-extracts were prepared by the breakage of cells in a

French Press. For this purpose, frozen cells were suspended in two

volumes of SP buffer and broken at 20,000 psi. A small amount of

deoxyribonuclease (20 ng ml"1) was added to the resultant viscous

liquid and the preparation was incubated for 15 minutes at room

temperature prior to centrifugation at 30,000 x g for 30 minutes.

15

The supernatant retained as "crude-extract", which was then

further separated into a high-speed pellet (membrane) and a high-

speed supernatant (HSS, cytosolic) fraction by ultracentrifugation at

150,000 x g for 90 minutes at 4°C. The pellet was resuspended in

the original volume of phosphate buffer and washed by an additional

high-speed centrifugation before it and other fractions were

assayed for cyanide conversion activity. In general, enzyme activity

in HSS was stabilized by the addition of 1.5 mM EDTA (disodium

ethylene-diamine tetraacetate) to the buffer when extracts were

prepared. NAD+ was removed from HSSs when necessary by

incubation with 0.2 U ml"1 NADase (NAD+ glycohydrolase [EC 3.2.2.5]

[Sigma Chemical Co]) for 1 h at 30°C.

Analytical methods.

Cyanide

Cyanide was determined by the colorimetric method of

Lambert et al., (1975). For this purpose, 10 jxl of sample was added

to a test tube containing 1.14 ml of water and 50 nl of N-

chlorosuccinimide-succinimide reagent, followed by the addition of

50 |il barbituric acid-pyridine reagent. The absorbance was read at

580 nm. A lavender color indicated the presence of cyanide and the

absorbance was converted to cyanide concentration in unknown

samples using standard curve.

Ammonia

Ammonia was determined using the colorimetric method

described by Fawcett and Scott (1960). Reaction mixtures contained

16

in a total volume of 1.125 ml: 0.125 ml of sample, 0.25 ml of

sodium phenate reagent, 0.375 ml of 0.01% of sodium nitroprusside

and 0.375 ml of 0.02 N sodium hypochlorite. After 30 min, the

absorbance was measured at 630 nm at room temperature. A blue

color indicated the presence of ammonia and the absorbance was

converted to ammonia concentration using standard curve.

Formamide

The quantification of both formamide and formic acid in

reaction mixtures was determined after deproteinization of samples

with ZnSC>4. Also, any unconsumed cyanide present in deproteinized

samples that might interfere either with formamide or formic acid

determinations was first removed by flushing samples with a N2-

gas stream for 5 min. Samples were deproteinized by adding 10 |xl

6 M ZnSC>4 to 100 nl of cyanide-free sample. Precipitated protein

was removed by centrifugation in a microcentrifuge. The formamide

present was then determined by the colorimetric method of Fry and

Millar (1972), which involves conversion of formamide to the

corresponding hydroxamate. The sample (0.1 ml) was incubated for

10 min at 60°C with 0.2 ml of a 1:1 mixture of 3.5 N NaOH and 2.3 M

hydroxylamine hydrochloride. One tenth ml of 4 N HCI was added

followed by 0.1 ml of 1.23 M FeC^ reagent prepared in 0.075 N HCI,

and the absorbance read at 540 nm after 5 minutes. The limit of

formamide detection using this procedure was <0.2 mM. When

present at concentrations below this, formamide was determined by

chemical hydrolysis to formic acid, which was then determined

enzymatically (see below). Formamide hydrolysis was achieved by

17

adding 10 \i\ 3 M H 2 S0 4 to 90-100 | l deproteinized sample in a

sealed microcentrifuge tube and allowing the mixture to incubate at

95°C for 2 h. After this the sample was neutralized with 5 M NaOH

and assayed for formic acid.

Formic acid

Formic acid was determined enzymatically using commercial

formate dehydrogenase (Sigma Chemical Co., St. Louis, MO) as

described by Hopner and Knappe, (1974). Reaction mixtures

contained the following in 0.4 ml : 18-20 |imol KH2P04 (pH 7.5 ), 0.4

jxmol NAD+, 0.16 U formate dehydrogenase, and 10-20 jxl of sample

(0.1-0.2 |imol formic acid). Reactions were initiated by the addition

of enzyme and the change in absorbance at 340 nm was measured

over 2-4 min. Since reactions proceeded only to about 62%

completion, an average extinction value (e) for NADH of 3.85 mM*1

cm - 1 was routinely used to calculate the formate concentration.

Protein determination

The protein content in cell-extracts was determined using the

Lowry procedure (Lowry et a!., 1951). This was carried out by

adding 5 ^l of sample to 0.2 ml water. Just before use reagent A

was prepared by mixing one ml of 0.5% CuS04 • 5H20 prepared in 1%

sodium citrate with 50 ml of 2% Na2C03 prepared in 0.1 N NaOH.

One ml of reagent A was then added to samples, which were

incubated for 10 min at room temperature. This was followed by the

addition of 0.1 ml of 1 N Folin-phenol reagent. After 30 min

incubation the absorbance was read at 750 nm and protein

18

concentration was determined from standard curve using bovine

serum albumin as a standard.

Analytical instrumentation

All spectrophotometric analyses were recorded on either an

LKB Ultraspec II or a Perkin-Elmer Lambda-6 uv/vis

spectrophotometer. The pH of all prepared solutions was measured

on a PHM82 model Radiometer pH meter.

Enzyme assays

Ceil-free cyanide-disappearance assay

The routine determination of cyanide degrading activity was

measured in cell-free extracts in sealed vials (2 ml HPLC screw-cap

autosampler vials sealed with silicon rubber/PTFE septa, Rainin

Instrument Co., Woburn, MA). The standard reaction mixture

contained in 0.25 ml: cell-extract (9-10 mg protein ml"1), 2 mM

KCN, 4 mM NADH (where indicated), and SP buffer (pH 7.0). Reactions

were initiated by the injection of KCN and incubation mixtures were

placed at 30°C on a Gyrotory shaker (250 rpm). Ten |il samples were

withdrawn with a syringe at desired intervals for the determination

of cyanide by the method described in Analytical methods. Separate

reaction vials containing cell-extract incubated in the absence of

cyanide were also included as references for colorimetric

determinations. Where indicated, reaction mixtures were made

anaerobic by flushing vials with oxygen-free N2 for 5 minutes prior

to injecting KCN.

19

Cyanide oxygenase assay (oxygen/NADH-dependent

CDA)

The CDA in HSSs measured under aerobic conditions in the

presence of 4 mM NADH with KCN supplied at 2 mM was taken as a

measure of putative cyanide oxygenase (CNO) activity.

Cyanide dihydratase/cyanide hydratase assay

(oxygen/NADH-independent CDA)

The CDA measured at high concentrations of KCN (25 mM) under

anaerobic conditions in the absence of NADH was taken as a measure

of putative cyanide dihydratase/cyanide hydratase activity.

Reaction mixtures were made anaerobic by flushing vials with

oxygen-free N2 for 5 minutes prior to injecting KCN.

Formate dehydrogenase

Formate dehydrogenase was assayed spectrophotometrically in

reaction mixtures containing the following components in 0.4 ml, 50

mM KH2P04 (pH 7.5), 0.1 mM NAD+, and 1 mM sodium formate.

Reactions were initiated by the injection of HSS (10-20 |il, 0.12-

0.24 mg protein), and NADH formation was monitored at 340 nm.

Specific activities were calculated using a molar extinction

coefficient of 6.22 mM'1 cm"1 for NADH.

Oxygen uptake measurement

Enzyme-dependent oxygen uptake was measured with a

Hansatech Clark-type oxygen-electrode (Hansatech Instruments Ltd.,

Norfolk, England). The electrode was connected to a CB1 control box

20

and Cole Parmer Model 8373-20 recorder (Cole-Parmer Instrument

Company, Chicago, IL). The electrode cathode disc was prepared for

operation by placing a cigarette paper spacer (1" square) over the

disc, followed by moistening with saturated KCI. A one-inch square

piece of electrode membrane was then placed over the paper-covered

disc and held in place with spacer and O ring applied so that the

membrane was stretched smoothly over the surface of the electrode

dome. Calibration of the electrode was carried out as follows. The

range switch for the instrument was set to x1, output control to

maximum (fully clockwise), and back-off switch to "cancel". The

control box switch and recorder were then turned on and one ml of

aerated SP buffer placed into the electrode cuvette chamber. The

maximal output reading was adjusted to 90-95% (with 100%

representing buffer saturated with 0 2 ) with the output control

switch, and a few crystals of sodium dithionite (Na2S204) w a s added

to the chamber. As the dithionite combined with dissolved 0 2 the

electrode signal quickly responded to zero. At this time the "back-

off" control was turned on and the output reading adjusted to 10-

15% (of maximum recorder output) using the "coarse" adjustment.

The electrode chamber was then thoroughly washed with H 2 0 and

filled with 1 ml of aerated SP buffer. The recorder came to 90-95%

of full scale deflection indicating that the instrument was ready for

use. Further calibration was achieved by adding 0.1 nmol of catechol

follwed by 10-25 nl of crude catechol-2,3-dioxygenase (C230)

(prepared from P. putida cells carrying the TOL plasmid and induced

for this enzyme by growth on m-toluate [Kunz & Chapman, 1981]),

21

and measuring pen deflection upon addition of enzyme; since 1 mole

of oxygen is consumed per mole substrate degraded by C230, this

provides a convenient means of making sure the instrument is

calibrated appropriately.

Oxygen uptake by HSSs from P. fluorescens NCIMB 11764 was

measured as follows. One ml of SP buffer was added to the

electrode cuvette maintained at 30°C by means of a water-jacketed

circulator and the instrument allowed to stabilize momentarily.

Two tenths ml of HSS (2.4 mg protein) was then added and the rate

of oxygen uptake recorded for 1-3 min. After this, 1 ^imol of NADH

was added and the rate of 0 2 uptake again recorded for 1-2 minutes

or until a steady rate of consumption was observed. This gave a

measure of the rate of NADH oxidation. To determine whether KCN

could further stimulate oxygen consumption, 0.05 - 0.1 jxmol KCN

was added when the rate of NADH oxidation had reached steady state

and both the stimulation in rate and total magnitude of oxygen

consumption further recorded [see Figure 7 for further description

of the order of addition of substrate(s)].

Spectrophotometric measurement of NADH oxidation by

HSSs

The oxidation of NADH by HSSs was further determined

spectrophotometrically. For this purpose, 0.1 jimol NADH (10 fxl of

10 mM) was added to a quartz cuvette that contained 0.4 ml SP

buffer and 0.1 ml of HSS (1.2 mg protein). After determination of

the endogenous rate of NADH oxidation (NADH oxidase), 5 JLLI 5 mM

22

KCN was added. Another 5 fi.1 5 mM KCN was added when cyanide

became depleted. Specific activities were calculated using a molar

extinction coefficient of 6.22 mM"1 cm"1 for NADH.

identification and quantification of 1 4 C - r a d i o l a b e l l e d

cyanide metabolites

Radiolabeled metabolites generated under cell-free conditions

were categorized as either volatile (containing CO2 and HCO3*) or

non-volatile (containing formamide and formic acid) using BaCI2 as

a bicarbonate trapping reagent according to the procedure described

by Fallon et al. (1991) and Kunz et al.{1992). Reactions were

carried out in 25 ml serum-stoppered flasks fitted with a center

well. The main-compartment of the flask contained, 0.4 ml cell

extract (HSS) (9-10 mg protein ml*1), NADH (where indicated) and

SP buffer in a total volume of 0.5 ml. Reactions were initiated by

the injection of labelled (1-2.5 M-Ci; 37-92.5 kBq) and unlabelled KCN

(2 or 25 mM), and when complete, as a determined by simultaneous

colorimetric assays for cyanide, 0.3 ml of 4 N NaOH was injected

into the center well to trap volatile CO2. The flasks were allowed

to incubate for an additional 15 minutes at which time the contents

of the center well and main compartment of the flask were removed

and fractionated with BaCl2 to recover volatile and nonvolatile

products. One quarter ml of the contents from the main

compartment was treated with 0.12 ml of 0.1 N NaOH plus 0.03 ml of

40% BaCI2 for 5 min, and the mixture separated into fractions

designated main-compartment alkaline barium precipitate

23

(containing bicarbonate) and main-compartment alkaline barium

soluble (containing formamide plus formate) by centrifugation. The

pellet (alkaline barium precipitate) was washed by centrifugation

with 0.2 ml of 0.12 N NaOH and the washing combined with the

supernatant (alkaline barium soluble). The pellet was then

resuspended in 0.5 ml of 0.12 N NaOH and both it and the supernatant

sample were added to scintillation fluid. The contents of the center

well (0.3 ml) were treated in exactly the same fashion, generating

fractions designated center-well alkaline barium precipitate

(containing CO2 and HCO3") and center-well alkaline barium soluble

(containing unconsumed cyanide). All radioactive samples were

added to 8 ml of Aquasol scintillation fluid (Du Pont, NEN Research

Products, Boston, MA) and counted on an LS 7000 liquid scintillation

counter (Beckman, Irving, CA).

Identification and quantification of 1 4 C - r a d i o l abel led

formate metabolites

The identification and quantitation of C02 generated from

formic acid was also performed radioisotopically using the same

procedure employed for 14C02 quantitation from cyanide. Reactions

were again performed in sealed flasks and initiated by the injection

of 1-2.5 |xCi (37-92.5 kBq) sodium [14C]formate (0.5 mM). NAD+

(where supplied) was provided at 1 mM.

GC/MS identification of cyanide conversion metabolites

Carbon-containing reaction products were also determined by

24

GC/MS. GC/MS analysis was performed in collaboration with Dr. B.

Venables, Trac Laboratories Inc., Denton, TX., using a Hewlett

Packard 5992 GCMS equipped with a Teknivent Vector-One Data

Acquisition System. 1 3 C0 2 45 m/z species in the headspace of

incubations conducted with K13CN as the substrate was analyzed by

mass spectrometry. This was accomplished by injecting 10 (AI gas

over a polydimethylsiloxane capillary column (Alltech Associates

SE3-30M). The presence of formic acid was also confirmed by

GC/MS. This was accomplished by injecting 10 jal of the reaction

mixture onto an Alltech Associates AT1000 10m megabore column

(injector 110°C, flow 6.5 ml min"1 He, oven programmed at 40°C for

2 min, then ramped to 110°C at 10°C min"1) fitted to a mass

spectrometer. The reaction time was 9 min and mass of 46 (HCOOH)

were detected in incubations conducted with KCN.

Methods for measuring oxygen-18 incorporation

The conditions for measuring oxygen-18 incorporation were

the same as for the assay of the putative cyanide oxygenase except

that K13CN was used as the substrate. Also, to increase the amount

of enzyme available and reduce the extent of labelled CO2 exchange,

some reactions were conducted with concentrated HSS and with

NaHC03 (final 50 mM) added to the reaction mixtures. HSS was

concentrated by lyophilization in a Speed-Vac and the dried powder

resuspended in water or oxygen-18 water. Reactions with labelled

water were conducted in 2 ml-sealed HPLC vials containing HSS or

concentrated HSS (2.4 -11 mg protein), 120-200 nl H 21 8 0 (final 44-

25

77% enrichment), 5 fxl of 100 mM K13CN (final 2 mM) and 10 p.l of

100 mM NADH (final 4 mM) in a final volume 0.25 ml. Experiments

with 1 8 0 2 were carried out in 15 ml capacity glass vessels to which

a break-seal glass ampoule containing 25 ml of 1 8 0 2 (at 1 atm) was

attached. The vessel was flushed with oxygen-free N2 (99.9 atom%)

for 10 minutes before breaking the seal and gases were allowed to

equilibrate throughout the system. Reactions mixtures contained

the same components as those described for H 21 8 0 experiments. In

each case, reactions were initiated by the injection of NADH and the

headspace gas analyzed as described above at the times where

indicated . The relative abundances of m/z species 45 ( 1 3 C 1 6 0 1 6 0) ,

47 (1 3C1 601 80), 49 (1 3C1 801 80), 44 (1 2C1 601 60), 46 ( 1 2 C 1 6 0 1 8 0)

and 48 ( 1 2 C 1 8 0 1 8 0) were quantified from the relative ion intensities

measured at an ionizing voltage of 70 eV. The isotopic enrichments

of 1 8 0 2 (m/z 32) and H 21 8 0 (m/z 20) were determined respectively,

by analyzing the headspace gas and aqueous reaction solution.

Resolution of cyanide oxygenase

Ammonium sulfate precipitation

HSSs were fractionated by ammonium sulfate precipitation

into three fractions: the 0-35% pellet, the 35-70% pellet and the

70% supernatant fraction. The two pellet fractions were

resuspended in SP buffer and all three fractions were desalted by

dialysis against SP buffer (MW cutoff 6000-8000).

U l t r a f i l t r a t i o n

Ultrafiltration was carried out using Centricon-30 (30,000 MW

26

cutoff) centrifugal microconcentrators. HSSs were loaded on

Centricon and centrifuged at 5,000 x g for an hour at 4°C. After

centrifugation, retentate and filtrate fractions were pooled.

Chemicals

Both K12CN (97%) and K13CN (99 atom%) were obtained from

Aldrich Chemical Company (Milwaukee, Wl). K14CN (54 mCi mmol"1,

2.0 GBq mmol"1 ) and H14COONa (48.7 mCi mmol*1, 1.8 GBq mmol"1)

were purchased from Du Pont (NEN Research Products, Boston, MA).

Isotopic 1802 (99 atom%) and H2180 (96 atom%) were supplied by

Isotec Inc. (Miamisburg, OH). Other chemicals and reagents used

were purchased from Fisher (Springfield, NJ), or Sigma (St. Louis,

MO) and were of the highest available commercial grade.

CHAPTER III

RESULTS

PART I. RECOVERY OF CYANIDE-DEGRADING ENZYMATIC ACTIVITIES AT

THE CELL-FREE LEVEL

Studies by Kunz et al., (1992) showed that intact cells grown

on cyanide catalyzed its time-dependent disappearance from

reaction mixtures at concentrations as high as 100 mM. Since the

product distribution varied with the reaction conditions (aerobic vs

anaerobic) and substrate concentration, it was hypothesized that

more than a single enzyme was responsible for cyanide attack. As a

first step toward the identification and characterization of these

enzymes efforts were made simply to show that the cyanide

conversion activities observed for intact cells could be recovered at

the cell-free level. The results that follow represent the outcome

of these efforts.

Centrifugal fractionation of cyanide-degrading activity in

cell-free extracts

Crude extracts were initially fractionated into cytosolic and

membrane fractions by ultracentrifugation as described in METHODS,

and different fractions assayed for CDA. Since initial studies by

27

28

Knowles and colleagues (1983a,b) indicated that cyanide degradation

using cell-free preparations was both oxygen- and NADH-dependent,

these experimental conditions were repeated in an effort to recover

CDA. Table 4 shows that under these conditions the bulk of CDA

measured at relatively low substrate concentration (2 mM) was

located in the HSS fraction. These findings are similar to those

first reported by Harris and Knowles (1983b). However, as shown in

Table 5, activity could still be observed when NADH was omitted

from reaction mixture (21% of that of complete), or under anaerobic

conditions (7% of complete). F:igure 1 shows the rate at which

cyanide was degraded under various reactions. These data again

show the stimulatory effect of oxygen and NADH on cyanide

degrading activity, but also show the lack of complete dependence on

the two for activity. These results supported the idea that P.

fluorescens NCIMB 11764 was capable of producing additional

enzymes besides a putative cyanide oxygenase. The presence of

these additional enzymatic activities which appeared to function

anaerobically was previously not observed in this organism, but was

inferred from the intact cell studies of Kunz et al., (1992). In order

to show that cyanide disappearance under cell-free conditions were

enzyme-catalyzed, rates of disappearance were measured at

different protein concentrations and the results obtained are

illustrated in Figure 2. These data show that regardless of whether

or not activity was measured aerobically (in the presence of NADH)

or anaerobically (in the absence of NADH) CDA was protein-

concentration dependent. It was further shown that in non-enzyme

29

Table 4. Conversion of cyanide by cell-free fractions of P.

fluorescens NCIMB 11764

Cell fraction Cyanide converted3

(nmol min"1 mg [protein]"1)

crude-extract

30,000 X g supernatant 8

high-speed

150,000 X g supernatant 14

high-speed

150,000 X g pellet

a Reaction mixtures contained: 2 mM KCN, 4 mM NADH, 50 mM SP phosphate buffer (pH 7.0), and 1.6 to 4.0 mg protein in a final volume of 0.25 ml. Measured as initial rates of cyanide disappearance.

30

Table 5. Requirements for cell-free conversion of cyanide by P. fluorescens NCIMB 11764

Reaction Cyanide converted

conditions3 nmol min"1 mg"1 (%)

Complete (+ 02 /+ NADH) 14 (100)

+ 02 / -NADH 3 (21)

- 0 2 /+ NADH 1 (7)

- 0 2 / - NADH 1 (7)

a Reactions conducted with HSS and activity measured as the rate of cyanide disappearance as described in METHODS. Anaerobic reaction vials were flushed with N2 for 10 minutes prior to the addition of NADH and KCN.

31

Figure 1. Time course of cyanide consumption by high-speed

supernatants from P. fluorescens NCIMB 11764. Reactions

conducted at 2 mM KCN and 4 mM NADH where indicated under the

following conditions: Q +02/+NADH; Q +02/-NADH; # , - 0 2 /

-NADH.

32

•8

>» u

\ 10 20 30 40 50

Time (min)

33

Figure 2. Effect of increasing protein concentration on cyanide-

degrading activity in high-speed supernatants from P. fluorescens

NCIMB 11764. Activities measured at 10 mM KCN under aerobic

(+02/+NADH) [Q] and anaerobic (-02 / -NADH)[^] conditions.

34

r—>

C I

§ u 0> H3

u

1 2

Protein (mg)

35

controls, or reaction mixtures supplied boiled HSS (100°C for 15

min), <1% of the cyanide supplied was degraded.

Kinetics of cyanide conversion by HSS

The kinetics of cyanide degradation were investigated under

the different conditions in which CDA was observed. This included:

aerobic reaction mixtures with NADH present, aerobic reaction

mixtures with NADH absent, and anaerobic reaction mixtures with

NADH absent. Figure 3 shows a plot of initial rates of cyanide

consumption observed under the above conditions as a function of

substrate concentration. These data show that cell-free CDA could

be observed at KCN concentrations as high as 100 mM, which is one

hundred-fold greater than that previously reported (Harris and

Knowles, 1983b). Again, the highest rate of cyanide consumption

was observed when both oxygen and NADH were supplied, with

maximal activity recorded at 20 mM cyanide: above this the activity

declined probably due to enzyme inhibition. A second activity

observed in the absence of oxygen (and NADH) was also apparent at

cyanide concentrations as high as 100 mM. Under aerobic conditions

with NADH omitted rates of substrate consumption fell between

those recorded aerobically and anaerobically, suggesting that under

these conditions the total CDA represented the contribution of

possibly two enzymes. Under all three conditions the initial rates of

substrate disappearance appeared to followed Michaelis-Menten

kinetics. In order to further differentiate the various activities,

36

Figure 3. Kinetics of cyanide degradation by high-speed

supernatants from P. fluorescens NCIMB 11764. Reaction

conditions: O . +02/+NADH; D , +02/-NADH; -02/-NADH.

37

e 200 \ %

S

/

r/ a 50

K.CN

38

rates of substrate consumption were compared after the initial

rates recorded under anaerobic conditions (with NADH absent) were

subtracted from those recorded aerobically (with NADH present). In

this way enzymes contributing to the total CDA were distinguished

on the basis of favored reaction conditions and the apparent

(denoting the use of non-purified enzyme preparations) kinetic

constants estimated without individually purifying each enzyme.

Figure 4 shows the results when plot of the data was made in a

Lineweaver-Burk form (Cornish-Bowden and Eisenthal, 1974; Dowd

and Riggs, 1965; Eisenthal and Cornish-Bowden, 1974). These data

show that significantly different Km values were obtained under

aerobic and anaerobic conditions (1.2 versus 12 mM, respectively).

From these observations it was concluded that one enzyme in HSSs

had a low Km and was both oxygen- and NADH-dependent (referred to

CDA1), and a second enzyme had a high apparent Km and required

neither oxygen nor NADH (referred to as CDA2). In contrast to the

differences observed in apparent Km, the observed Vmax values for

the two were found to be identical (125 iM min"1; Figure 4). The

kinetic data for the CDAs described in cell-extracts are further

summarized in Table 6.

Reaction products and stoichiometries

The dependence of CDA1 on oxygen and NADH was found to be

analogous to the findings of Harris and Knowles (1983b), who

proposed that cyanide-degradation was oxygenase-mediated.

However, in order to further demonstrate the involvement of such an

39

Figure 4. Lineweaver-Burk plots for enzyme-catalyzed cyanide

degradation under aerobic (O ) and anaerobic ( 0 ) conditions.

40

0.025

0.020 -

S 0.015 -

0.005 -

1.2 -0.8 -0.4 0 0.4 0.8 1.2 1/[KCN] (mM1)

41

Table 6. Cyanide degrading activity (CDA) kinetic constants

Activity Vmax Km

(nmol min"1mg"1) (mM)

CDA1 (+02/+NADH) 12.5 1.2

CDA2 (-02/-NADH) 12.5 12

42

enzyme, and also to identify the nature of additional enzymes (e.g.,

CDA2), the reaction products formed after incubating HSSs with

cyanide at concentrations favoring one or the other activity were

determined. For this purpose, incubations were conducted at 2 and

25 mM KCN (considered optimal, respectively, for the apparent low

[CDA] and high [CDA] Km activities) and the product balances

determined. Table 7 shows that regardless of the reaction

conditions, greater than 90% of the cyanide carbon could be

recovered as C02 , formic acid and formamide. In contrast, the

accumulation of cyanide-nitrogen equivalents as formamide and

ammonia varied substantially (39-93 molar%). Presumably, this is

because ammonia once formed, is further metabolized. This helps to

explain the molar ratios of formamide : ammonia, which under

aerobic (with NADH present) and anaerobic (without NADH)

conditions were only 10:38 and 32:7 (48 and 39% molar recoveries),

respectively. For comparison, at 25 mM KCN, values of 37:56 and

41:49 were recorded.

Significant similarities were obtained when the carbon-

derived product balances obtained under cell-free conditions (Table

7) were compared with those previously reported for intact cells

(Kunz et a!., 1992). In each case, CO2 was identified as the major

reaction product under aerobic conditions as opposed to formic acid

and formamide under anaerobic conditions. Also, CO2 represented

the major reaction product at low substrate concentrations as

opposed to formic acid and formamide at higher concentrations.

These results when taken together with the kinetic data summarized

CO Is-

0Q 2 O z CO c Q) O CO a> o

CO CO CO X >% JQ E CO o CO % E o

"c CO >» o

*4—.

o CO o 3 "D O u

a.

o jD co

2 O *

o E n u-CD Q. O E c

"O <D E

CO <*—• o 3 "D O

C O "«4—» CO JD 3 O c

§

CO c o

c o o c o "<•— Q CO CD 0C

CO "c o E E <

CD "D E CO E V. o LL

O CO o E u. o (JL

CVJ O o

cd • I £

u CD E 3 CO C O O

CVJ 00 CO CO

CO •—

+1 M +1 +1 0 00 CO

CVJ CO w

h-CO a>

00 Tf

r— 00 CVJ

UD 0 r—

h-00

+1 +1 +1 +1 CVJ 0 r -

0 h-CO

CVJ CO

to 0 Tj-

00 CVJ <J>

CO 0 T—

+1 +1 +1 +1 0 CO

m CO

CVJ o> CO

00 T—

to 0 r CVI

+1 +i +1 +1

K 00

00 00 1-

CO 00 0 CO

CVJ

10 CM 0 10 CO 0

CVJ to CM 10 CVJ CVJ

1 Q < 2 .+ "cvj O

T-< Q O

I Q <

CM o

CM < Q o

4 3

« "O © 0) ™ * - »

2 X .2 E 1 o © '"5 ® co •€ ® 5 DC .

I ® 2 S 2 x: Q. T3 3 « O) =

© s x: T c

c 5 — o

© u. CD S

£ § o cd "a ^ 2 ? a §

o ^

- s E

0 CO CD

"D S CD E 3 CO c o o CO

cd u CO c £ o JZ CO CO

5 co 4s1 O) CD CO

CD * 73 ° C v- CO to ^

? E E 2 to c -o o © 0 15 CO ~ CD CO J-> Q--o p .® — co £ © a c

c © •«— •*—'

•D 2 (D Q.

S P "O E c o o O T-

I © o> L. N-' (D s « (/) « I c -rx <D T3 .§ S

uj 8

2 °

I s

— h-<*- LL| O ^

© "O O) Q)

CO *-* c CD E

ai £

CD CO * a)

CD a. rr X O q

0 co v. co a CD CO

«•— IN < TJ ^

W . "O

2 0

O £ Z E o 0 15 £ «= .2 .2 >, w £ g ©

L. —— C CO CD q (D CO c x: o ~ E o >» o

tQ i» <D *•-£ w © .E o .2 E £ > t a ~ © O co " a t j

44

in Table 6 suggest that the oxygen- and NADH-dependent activity

(CDA1) was similar to putative cyanide oxygenase (CNO) as first

described by Harris and Knowles (1983b). This enzyme, because of

its lower apparent Km, is believed to be responsible for conversion

of cyanide to C02 in 87% yield when supplied at low substrate

concentration (2 mM). In contrast, at higher concentrations CDA2 is

believed to catalyze the conversion of cyanide to formate (and NH3)

and formamide. These results are analogous to the findings reported

earlier (Kunz et a/., 1992) in which the same products were

identified when intact cells were supplied high concentrations of

cyanide. The production of these products is consistent with an

enzyme reaction mechanism involving putative cyanide nitrilase

(producing formic acid [and NH3 j) and cyanide hydratase (producing

formamide) enzymes as previously proposed (Scheme 2) (Kunz et al.,

1992).

Conversion of formate to C02 by cyanide-induced cell-free

extracts. The detection of C02 as a reaction product regardless of

whether or not HSSs were induced with cyanide under aerobic or

anaerobic conditions (Table 7) raised the possibility that C02 could

arise by two different means as shown in Scheme 3. One of these, it

was presumed, involved the direct conversion of cyanide to C02 (by

the action of CDA1 [putative cyanide oxygenase]) while the second

involved C0 2 production from formate (since the latter was also

identified as a reaction product). To determine whether formate

could be further metabolized, initial experiments were performed in

45

which HSSs were incubated with radiolabeled formate and the

production of 1 4 C0 2 (and H14C03") was measured as described in

METHODS. As much as 25 ± 2% (mean and mean deviations from two

CQ>+NH3

NH3+HCOOH COz

Scheme 3

separate experiments) of the formate supplied to HSSs was found to

be converted to C02 during a 2.5 h incubation. This increased to 88 ±

3% when reaction mixtures were also supplied NAD+, indicating that

formate could in fact be converted to C02, and that the responsible

enzyme was NAD+-dependent. The conversion in the absence of

added NAD+ was somewhat of a surprise until it was discovered that

pretreating HSSs with NADase prior to incubation decreased the

yield of C02 to 4.8 + 1.7%. This indicated that formate conversion in

the absence of added cofactor could be attributed to endogenous

NAD+ present in HSSs.

Spectrophotometric identification of cyanide-induced

formate dehydrogenase activity in HSSs

The observation that conversion of formate to C02 was NAD+-

dependent suggested that the responsible enzyme was a formate

46

dehydrogenase. This was verified by spectrophotometric assays as

shown in Figure 5. These data show that HSSs catalyzed the

formate-dependent reduction of NAD+ and that this activity was

protein-concentration dependent. Other experiments indicated that

activity was present under either aerobic or anaerobic conditions.

Summary of enzymatic activities induced by cyanide

Table 8 summarizes the various enzymatic activities detected

in P. fluorescens NCIMB 11764 when either grown on, or induced

with cyanide. One of these (CDA1) was tentatively identified as

cyanide oxygenase as already discussed while CDA2 is believed be

composed of one or two enzymes, namely CND and CNH. Since it has

not been possible at this point to differentiate the latter two

activities, CDA2 is simply referred to as "CND/CNH", representing

the possible contribution of two different enzymes. In addition to

CDA1 and CDA2, cyanide also induced the formation of formate

dehydrogenase.

Effects of inhibitors on ceil-free cyanide degrading

ac t iv i ty

In order to obtain information on the possible mechanism of

cyanide degradation, various compounds, as shown in Table 9, were

tested as potential inhibitors (or activators) of cyanide degrading

activity. The results show that of all the compounds listed, only

phenylmercuric acetate, p-hydroxymercuri benzoate, and HgCI2

gave any significant inhibition (35-53%). Since these compounds are

47

Figure 5. Spectrophotometric detection of cyanide-induced formate

dehydrogenase in HSSs from P. fluorescens NCIMB 11764-showing

the effect of increasing protein concentration on formate-dependent

formation of NADH.

48

r r

en

CN

o> <*•» 9 C

B w

E H

( u i a ot>£) a o u B q j o s q y

49

Table 8. Summary of cyanide-induced enzymatic activities in P.

fluorescens NCIMB 11764

Specific activity (nmol min"1 mg"1 [protein])

Cultivation

conditions3

CNO CND/CNH Formate

dehydrogenase

Cyanide-grown

Cyanide-induced

Ammonia-grown

12.6

8.1

<0.5

11.8

4.5

<0.5

14.0

8.8

<0.5

a Cultivation conditions are described in METHODS.

50

Table 9. Compound test for as effector on cyanide degrading activity

Compound Concentration Activi ty3

(mM) (%)

Control 100

HgCI2 0.1 65

p-Hydroxymercuri benzoate 0.0005 62

Phenylmercuric acetate 0.005 47

Fe3+ 0.2 98

Ni2 + 0.2 87

Acetylene 1.5 99

Imidazole 1.5 96

Pyrimidine 1.5 91

Histidine 1.5 86

Antimycin A 1 100

Carbonyl cyanide m-chloro-

phenyl hydrazone 1 100

2,2'-Dipy ridyl 1 100

2,4-Dinitrophenol 1 100

Sodium azide 1 100

a Activity measured as rate of cyanide disappearance in the standard cyanide degradation assay described in METHODS with reagents present at the concentration indicated.

51

known to interact with sulfhydryl groups, these data indicate that

this functional group may play a role in the enzymatic mechanism.

Although none of the compounds tested showed any significant

activation, EDTA (1.5 mM) was found to stabilize cyanide degrading

activity in HSSs; it therefore, was routinely added to buffer used to

prepare cell-extracts.

52

PART II. REACTION STOICHIOMETRY AND ISOTOPIC IDENTIFICATION

OF CYANIDE OXYGENASE

Results which showed that C02 (and NH3) were formed as

reaction products when cyanide was oxidatively metabolized by

HSSs (Table 6) supported the idea that an oxygenase-type enzyme as

originally proposed by Harris and Knowles, (1983b) was responsible

for conversion. We further suspected that since the Km towards

cyanide by CDA1 (believed to contain putative cyanide oxygenase)

compared with CDA2 was much lower (1.2 vs 12 mM) than the former

enzyme, it was important physiologically for cyanide utilization.

This idea was reinforced by parallel work conducted by fellow

graduate student Mr. Jui-Lin Chen who showed conclusively that

cyanide oxygenase activity in a mutant having lost the ability to

grow could not be detected (Kunz. et al., 1994). Thus, because of

the importance of this enzyme for cyanide utilization, one of the

major goals of this thesis was to prove unequivocally that an

enzyme fitting the description of an oxygenase was involved. To

accomplish this, experiments aimed at verifying the reaction

stoichiometry as originally reported by Harris and Knowles (1983b)

were repeated and the molecular source of oxygen atoms

incorporated into cyanide was investigated with isotopically

labelled 1 8 0 2 and H21 80.

53

Reaction stoichiometry

Figure 6 shows the results obtained when oxygen consumption

was measured with an oxygen electrode in the presence and absence

of cyanide (and NADH). These data show a stimulation in the rate of

oxygen consumption by NADH and a further stimulation upon the

addition of cyanide, which continued until all of the available

cyanide was consumed. The initial NADH-dependent rate of oxygen

consumption was believed to be attributed to NADH oxidase activity

which was calculated to have a specific activity of 2 nmol min"1

mg"1 (protein). The cyanide stimulated activity was concluded to be

due to cyanide oxygenase. The calculated molar ratio for oxygen

consumed and cyanide degraded was 1:0.97 + 0.12 (n=7). Related

experiments aimed at measuring the stoichiometry between NADH

oxidation and cyanide consumed as determined spectrophoto-

metrically were also conducted. As shown in Figure 7, cyanide

stimulated the rate of NADH oxidation above that attributed to NADH

oxidase, with the amount of NADH consumed continuing until all of

the available cyanide was again degraded. In this case, the molar

ratio between NADH oxidized (based on e, 6.22 mM"1cm"1) and

cyanide degraded was determined to be 1:1.02 ± 0.06 (n=10). Table

10 summarizes the results obtained for the reaction stoichiometry

which are in general agreement with findings initially reported by

Harris and Knowles (1983b).

54

Figure 6. Stoichiometry of 0 2 uptake and cyanide. The oxygen

electrode at 30°C contained 0.7 ml phosphate buffer (pH7.0) and 0.3

ml HSS {3.6 mg protein). NADH (10 |xl, 100 mM) was injected and the

rate of oxygen uptake by NADH oxidase was determined. KCN (10 ]i\,

10 mM) was then added and the stimulation in oxygen uptake was

recorded.

55

0.05 fimol KCN

0.1 jimol KCN

1 jxmol NADH

300 )il HSS (3.6 mg protein)

56

Figure 7. Stoichiometry of cyanide-dependent NADH oxidation. A

cuvette containing 0.4 ml phosphate buffer and 0.1 ml of HSS (1.2

mg protein) was incubated with 10 ]nl of 10 mM NADH. After

determination of the endogenous rate of NADH oxidation (NADH

oxidase), 5 jnl 5 mM KCN was added. Another 5 |il 5 mM KCN was

added when cyanide became depleted.

57

3 e

4> £

(uiu OK) wueqjosqv

58

Table 10. Reaction stoichiometry for putative cyanide oxygenase

measured in HSSs of P. fluorescens 11764a

KCN 0 2 NADH Estimated

Supplied Consumed Consumed molar ratios

(nmol) (nmol) (nmol) KCN: 02 :NADH

25 NDb 25 ± 1 1 : ND : 1.00

50 48 ± 6 50 ± 4 1 : 0.96 : 1.00

100 97 ± 12 ND 1 : 0.97 : ND

a Cyanide was measured colorirnetrically. Oxygen and NADH consumption were determined polarographically and spectrophtometrically as described in METHODS.

b ND indicates not determined.

59

Evidence for the involvement of formate dehydrogenase in

NADH recycling

The discovery of formate dehydrogenase in P. fluorescens

NCIMB 11764 suggested that the NAD+-driven oxidation of formate

might provide a means of generating the NADH utilized by cyanide

oxygenase. To explore this possibility the effect of NAD4" and

formate on both cyanide oxygenase activity and corresponding molar

conversion yield of cyanide into C02 were tested; the rationale being

that if these compounds were involved in NADH recycling, an effect

on both enzymatic activity and production of C02 might be observed.

Table 11 shows that when reactions were conducted with extracts

that were treated with NADase, both cyanide oxygenase activity and

the corresponding C02 yield were substantially lower than in

comparable reactions in which untreated HSSs were employed.

However, when reactions were supplied with NAD+, both cyanide

degrading activity and C02 yield increased to 67 and 46% of

maximal, and when formate was supplied, these increased to 98 and

100%, respectively. In contrast, the addition of formate alone had

no significant effect on either enzymatic activity or C02 production.

A role for both formate and NAD+ in the recycling of NADH was thus

inferred. A metabolic scheme showing the inter-relationship of

formate dehydrogenase, putative cyanide oxygenase and cyanide

nitrilase in cofactor recycling is shown in Figure 8.

60

Table 11. Effect of formate and NAD+ on putative cyanide oxygenase

activity in P. fluorescens NCIMB 11764

Reaction conditions Specific % Molar yield

activity 1 4 C0 2 from

[nmol min"1 K14CNb

(mg protein)"1]3

NADH 12.6 ± 1.4 (100) 87.1 ± 5.1 (100)

No NADH 2.6 ± 0.6 (21) 25.1 ± 2.8 (29)

NADase-treated HSS 1.2 ± 0.3 (10) 3.6 ± 0.8 (4)

NAD+ 8.5 + 1.2 (67) 39.6 ± 0.6 (46)

Formate 2.4 + 0.4 (19) 18.2 ± 2.4 (21)

NAD+ + formate 12.4 + 1.6 (98) 87.2 + 1.3 (100)

a Calculated from initial rates of cyanide consumption by HSSs from cyanide-grown cells.

b Determined as the sum of volatile C02 and non-volatile HC03~

collected after > 90% of the substrate was consumed.

61

Figure 8. A metabolic scheme showing the inter-relationship of

formate dehydrogenase, putative cyanide oxygenase and cyanide

nitrilase in cofactor NADH recycling.

62

0. HCN

NADH + NAD

CNO

FDH

CND COo + NH-

63

Enzymatic incorporation of isotopic oxygen-18

While the above experiments were all consistent with a

reaction mechanism involving an oxygenase, ultimate proof for this

required that incorporation of molecular oxygen into the substrate

be demonstrated (Hayaishi, 1982). In anticipation of these

experiments a time course for the production of C02 was determined

as shown in Figure 9. This experiment shows that HSS catalyzed the

conversion of radioactive K14CN into CO2 in about an 80% yield in 30

minutes. Enzymatic activity was both oxygen and NADH dependent

although a small amount of conversion was also observed under

anaerobic conditions which is consistent with activities expected

respectively, for CNO and CND/CNH as described in Part I of this

dissertation. The specific activity for the putative cyanide

oxygenase determined by measuring the rate of CO2 production was

estimated to be 16 nmole min"1 mg"1. This agrees with values

obtained by other methods including measurement of rates of

cyanide disappearance, oxygen uptake, or NADH oxidation. Once it

was established that we were dealing with what appeared to be an

oxygenase-type enzyme oxygen-18 incorporation experiments were

performed, the procedures for which are outlined in Methods.

The recovery of labelled oxygen in CO2 was complicated since

it readily exchanges its oxygen atoms with water. At pH 7.0, for

example, CO2 is rapidly hydrated and dehydrated (hydration/

dehydration cycle) (Pocker and Bjorkquist, 1977) as shown in

Scheme 4 and scrambling of oxygen atoms occurs. This helps to

explain the results observed when theisotopic content of 1 3C02

64

Figure 9. Time course of 1 4 C0 2 formation by HSSs from P.

fluorescens NCIMB 11764. Reactions conducted with 2 mM K1 4CN

and 4 mM NADH supplied under aerobic (O) and anaerobic ( £ )

conditions. 1 4 C0 2 was determined as described in METHODS.

65

100 "O 0) fc. 0) > c o o

<M S 8 *

o s <D O) <0

0) o I-0) Q.

10 20

Time (minute)

66

formed from K13CN was determined in initial experiments conducted

for prolonged periods as shown in Table 12. Only 0.1% of the CO2

formed was 180-Iabelled when 1802 was supplied (Experiment 1).

C01 80 + H20 " HC00180' + H+ C00 + H21 80

Scheme 4

This can be explained by the fact that any potentially labelled

species would likely be diluted out as a result of exchange with the

large excess of unlabelled water. Conversely, the same explanation

can be used to explain why up to 46.8 and 19.2%, respectively of m/e

species 47 ( 1 3C 1 60 1 80) and 49 (1 3C1 801 80) were detected in

incubations conducted with H2180 (Experiment 1, Table 12). In this

case, unlabelled 13C02 became labelled because of chemical

exchange with the large excess of labelled H2180 (44%) supplied. It

is important to emphasize that the ratio of m/z 45:47:49 observed in

the latter experiment (0.34:0.47:0.19) is consistent with the

expected ratio (0.31:0.49:0.19) assuming complete chemical

exchange. The latter calculation was arrived at from the following

probability expression:

£(x2 + 2xy + y2) = 1

where x2 = probability that both atoms are derived from 1 6 0

and y2 = probability that both atoms are derived from 1 8 0

at 44% H 21 8 0 enrichment

67

Table 12. Legend

a Reaction mixtures (0.25 ml) contained approximately 50 mM SP

buffer, HSS (or concentrated HSS at the protein concentration

shown), and 50 mM NaHC03 (where indicated). Reaction mixtures

supplied H 21 8 0 contained 125 jxl of 96 atom% H2

1 80, and 100 jil

and 95 JLLI respectively, HSS and four-times concentrated HSS in

Experiment 1 and 2. In Experiment 3, 200 ^l of H 21 8 0 was used to

resuspend the dry powder obtained after lyophilizing HSS, which

was added to reaction vessels already supplied NaHC03. Reactions

were initiated by injection of 2 mM K13CN and 4 mM NADH. b Values obtained for 1 8 0 2 experiments were corrected for m/z 46

and 48 generated from 1 2 C 1 6 0 2 (m/z 44) by non-enzymatic

labelling as a result of chemical exchange with 97-99 %

enrichment 1 8 0 2 .

c o V-» o co 0

0 E o u

o Z

w to * m "O C c

E <o >> w _Q CO

CO TJ X 0) 0 1 £ 0 T-"D co CD 0 "o CD a. CO

2 O

CM o o a GL O

CO c Q) O CO s O

o . .52 0* *+r ~U O Q

CO 0 OL a. __ 3 CO co c <D CO

a> •E 3 c 5 .2 £ CD >

as CD a: cvi

0

co

"D CD xt a 1

<D • CD 73 '

0«.s O -C CO O T- CO

0 o ^ "5. o o ^ 4-< sO O £ CO

c o 4-> CO JD D a c

o>

N

in

u o CO

T5 O u 0 a

CO c o *4-< 73 c o o

> s O E O c

X ° LO

C C • o .E •= ^ 0 C O -J-

I - I

c 0 o u. 0 CL

C 0 E JC o SZ c 0

co +-*

c 0 £ *u 0 Q. X LLI

0

0 to 0 jQ CO

CO .Q D CO

0> o> o>

o o CM h-

O o

o o o>

o CD

o o o

h* O h-CO CO

co o co CO CO CO O) O) O)

o o o CM CO

CO a>

00 to

O) O)

a> 0)

h-a>

CM CO

Cvj o CD

68

CM a!

co CO

CO

o o CM K

00

o CO

CO CO

o

IT)

O CO

O O t-

O) O) CO 00 CO CO ^ ^ CO

r- co in lo in co

o o o CM co

00 in

h-h-

CM CO

0 00

'"ca 1

69

It was therefore concluded from these experiments that because of

the length of incubation it would not be possible to determine the

molecule source of oxygen atoms incorporated.

Several solutions were thought of to overcome the problem of

CO2 exchange. One was to decrease the time between reaction

initiation and analysis of reactions. The second was to increase the

enzyme concentration, and the third was to incorporate bicarbonate

into reaction mixtures. The latter approach it was hypothesized,

would alter the chemical equilibria (hydration/dehydration cycle) to

the point where exchange would effectively be reduced. The latter

approach was based on findings reported over forty years ago by

Mills and Urey (1940) who showed that the rate of CO2 exchange at

25°C in aqueous solution was reduced from 0.22 min*1 ( t h a i f - f i f e > 3

min) to 0.0114 min-1 ( t h a i f - i i f e » 60 m ' n ) when excess bicarbonate

versus CO2 was provided in solution. After initial experiments

showed that the addition of bicarbonate (50 mM) to reaction

mixtures had no effect on enzyme activity, the second set of

labelling experiments shown in Table 12 (in which CO2 was rapidily

analyzed and concentrated enzyme preparations were used

[Experiment 2]) were carried out.

Both m/z species 47 and 49 were again detected in reaction

mixtures supplied H21 80 after 1 minute. This time, however, the

ratio of isotopic species 45:47:49 (0.54:0.33:0.13) was different

from that expected on a theoretical basis assuming complete

exchange at equilibrium (0.28:0.50:0.22, based on 47% isotopic

enrichment). This finding implied that some biological labelling

70

from water had occurred. Parallel experiments with 1 802 suggested

that oxygen incorporation had also occurred. In this case, 10% label

was recovered in m/z species 47 ( 1 3 C 1 6 0 1 8 0) , which represented a

hundred-fold increase over that observed in long-term incubations

(Experiment 1). Thus, these experiments convincingly indicated that

labelling from molecular oxygen had occurred. The data with H2180

were less convincing because it was difficult to distinguish

between enzymatic and non-enzymatic (chemical exchange) labelling

because of the relatively low enrichment of H21sO (47%). Therefore,

experiments were repeated one more time (Experiment 3) with even

more concentrated enzyme preparations and the isotopic content of

CO2 determined after 20 seconds. Also, H2180 in experiment 3 was

enriched to 77%. The results with 1 802 were consistent with those

observed for Experiment 2, with approximately 10% of the labelled

CO2 recovered in m/z 47 species and no m/z 49. Interestingly, very

similar results were now also obtained with H2180, with m/z 47

representing almost 50% of labelled species and no 49 being

detected. These results provided strong evidence that in addition to

the incorporation of a single atom of oxygen from 1 802, one atom

was also incorporated from water. A further comparison of the

ratio for m/z 45:47;49 with those observed in Experiment 1 and 2 (in

which case significant non-enzymatic labelling due to chemical

exchange had occurred because of longer reaction times and the

absence of bicarbonate) revealed a significant difference. In this

case the ratio was 50:50:0, further indicating that m/z 47

(13c16o180) had been formed by enzymatic labelling.

71

The above findings suggested that one atom each of oxygen

from molecular oxygen and water were incorporated into cyanide.

We were concerned somewhat with the rather low recoveries of m/z

47 when 1 802 was supplied, since one might hypothesize that the

45:47:49 ratio should be close to 3:97:0 (at 97% enrichment).

However, the explanation for this may be related to the reaction

mechanism, a proposal for which is shown below. For example, if

initial incorporation from molecular oxygen occursto give an

NADH + H+ + 18O2 NAD+ + H2180 H2

16O NH3

V V V 'J HCN ^ ^ • X-180 ^ ^ • C180160

Scheme 5

unstable intermediate that is further hydrolyzed (by the large

excess of unlabelled water provided in 1 802 experiments), then the

amount of label capable of being recovered in CO2 would be expected

to be quite low as was observed.

72

PART III. PARTIAL RESOLUTION OF CYANIDE OXYGENASE

The identification of a cyanide oxygenase and putative cyanide

dihydratase/cyanide hydratase enzymes in cell-extracts was

consistent with findings reported previously (Kunz et al., 1992),

indicating that cyanide is metabolized by more than a single

mechanism in P. fluorescens NCIMB 11764. Because of the central

importance of CNO in cyanide assimilation, preliminary efforts were

made to purify this enzyme.

Recovery of cyanide oxygenase in ammonium sulfate

f r a c t i o n s

Initial studies reported by Harris and Knowles (1983b)

indicated that cyanide oxygenase activity was present in combined

0-35% and 35-70% ammonium sulfate fractions. These experiments

were repeated and similar results were obtained (Table 13); some

activity (30% of that observed for HSS) was also recovered in the

35-70% fraction. However, over 50% of the initial specific activity

detected in HSS could not be recovered in the combined 0-35% and

35-70% ammonium sulfate fractions. A similar loss in activity was

observed when HSS was subjected to simple dialysis (Table 13),

suggesting that an essential component was removed during

desalting of fractions by dialysis. In examining the data initially

reported by Harris and Knowles (1983b) a similar loss in activity

was noted, however, at the time no explanation for this loss was

CO c o o CO

0 CO *•— D CO

E 3 *c o E E 03

CD CO CO c CD O) >1 X o 0) T3 "E CO >* o

0 > o o <1> OC

CO T— JD JQ CO h-

O) E t

JQ .E >* E

"O 0) "O CO

o CO

o <D a CO

O) a> "O 1 z o

"o E c

CO c E 0 •3 a> CO E c (D >s O CO c CO 0 CO 0 co c -0 "5 4—»

1— co

0 "D k» CL

c s "55 c

c o <«-* 0 CO

7 3

o O o

O o CO ^

r- 1 0

o V 00 to

to o V

CD N

CO 10

tO o

00 0>

00 CO CM

CO

o 00 00 00

8 X

nO m CO

0s

o N-I

to CO

XP + 0 s

O ^ to 10 R «? CO A O

JO O) *E l— <d > o

o o

To

JZ O) "c CD > o

o o

3-

CO CO X

"U 0) N >* co Q

"D a> CO JQ 3 o c

CO CO X

a> £

o> c "E "co 4-* c o o CO c o *+z o CO CO D o 'u CO > T3 c CO X Q <

2 E

z o

E CM *0 CD c i2 c o o

10 CM o

CO 0

CD > o J3 CO c 5 o JC CO CO CO c o

c o o CO CD 0C

CO Q o X I— LU

-a 0 n *u o CO 0 T5 >s CO CO CO CO CD o c CO V. co CD Ql a. co CO T5 0 "D c CO >s o

co ~o c CO •*-' CO

CO CD v- <|) 3 X E

c 0? o c o o c 2 o v~ CL

>N JD "D CD u 3 CO CO Q> 2

74

provided but we hypothesize that it may have been due to the

removal of the same dialyzable component.

Requirement of a small dialyzable component for cyanide

oxygenase

The involvement of a dialyzable factor in cyanide oxygenase

activity was further investigated by ultrafiltration. For this

purpose, 1 ml of HSS (12 mg ml"1 protein) was subjected to

ultrafiltration through a 30,000 molecular weight cut-off

Centricon® membrane after centrifugation at 5,000 x g for 90

minutes. The retentate (fraction F1, 0.08 ml [150 mg ml"1 protein])

and filtrate (fraction F2, 0.92 ml [0.55 mg ml"1 protein]) were then

assayed separately and in combination for cyanide-degrading

activity. As shown in Table 14, activity was detected in the

retentate (fraction F1), however, the specific activity was only 51%

of that observed for HSS. When the retentate and filtrate were

combined (F1 [1.2 mg protein] + F2 [50 |xg]) activity comparable to

that of HSS was again observed. Similarly, the addition of F2 to the

combined 0-35% and 35-70% fractions also restored complete

activity. These results further implied the involvement of a small

molecular weight (<30K) component in cyanide oxygenase activity.

Characterization of small dialyzable cyanide oxygenase

component

Further experiments were performed to identify and

characterize the F2 component. Since F2, in comparison with F1,

CO c o .mm

o CO

c o CO

CO U. 4— D

c

<D CO CO c CD O ) > * X o <D

I D C CO > s o

£ CD > o o CD

a :

© JD co

H

o >

e

~ E

> T J CD

" O CO 1— O ) CD

U

O CO

O

2 O a. CO

o E c

•c I % O ) 5 E c CD O c o a

a> o

>* CO CO CO CO

CO T J C CO Q . 2 co

c o o CO

o o

T~ o LO

Y - CO

0 0 <*>

in o

o

c o 00

h -o

CO o i

o o CO CM o

o to

o in

o CO A

CM U L

N ? o CO V

CVJ I L

O h -i m co

CM c c o o u-« . 2 + & E E <?

X < < L L o 8

75 CO c o o CO

c o

CO c

! d E o o

CO u o

' u CO >

" O c CO

x~ Q <

Z

E

z~ o *

E

CM

" U CD c is c o o

in CM

co CD imm

3 X

E c o

o CO a>

0 c

0 ) Q o X H U J

X 5 CD

JQ ' u o CO CD

u

>s CO CO CO CO

0 ) o c CO w. co CD Q . Q . CO CO

" D

0 " U "c CO

a

co T J c co

4 ^ CO

i I I £

Q) c $ o

j c CO

CO co

D CO CO CD

2

76

contained very little protein (0.55 mg ml"1) it was concentrated 6

to 8 fold by lyophilization and rates of cyanide disappearance

measured as a function of F2 concentration. Figure 10 shows that

the rate of F1 -catalyzed cyanide disappearance was F2-

concentration dependent implying a catalytic role.

Table 15 shows a series of experiments performed in an effort

to characterize the F2 component. When F2 was heat treated, over

85% of the initial combined F1 + F2 activity was retained, indicating

that the active component was heat stable. In contrast, the

stimulatory effect of F2 was lost when subjected to acid hydrolysis

or protease treatment. Based on these observations it was

concluded that F2 contained a small molecular weight protein. To

determine its approximate size F2 fractions were subjected to

ultrafiltration and dialysis through 10,000 and 500 MW cut-off

membranes. Various fractions were then assayed in combination

with F1, and based on the results shown in Table 15, it was

concluded that the molecular size of the F2-containing protein was

between 500 and 10,000 Dalton.

77

Figure 10. Effect of F2 concentration on stimulation of cyanide

degrading activity of F1. Reaction mixtures contained 2 mM KCN, 4

mM NADH, F1 (1.2 mg protein) and different amounts of F2 (0-200 ^ig

protein) in a final volume of 0.25 ml.

78

4> O c (0 la.

<0 0) a a (0 (0 ^ =5 =

• I

! l >» w

a

<u (Q CC

100

100 150 200

F2 protein ( | ig)

79

Table 15. Effect of F2 treatment on cyanide oxygenase activity

Specific activity3

Fraction nmol KCN-degraded min"1 mg"1

HSS 18.1 (100)

F1 9.3 (51)

F1 + F2 (heated at 100°C/15 min) 15.5 (86)

F1 + F2 (treated with protease)*3 9.3 (51)

F1 + F2 (Acid hydrolysis)0 10.1 (56)

F1 + F2 (Amicon 10K filtrate) 17.2 (95)

F1 + F2 (500 MW cut-off dialyzed) 17.6 (97)

a Reaction mixtures contained 2 mM KCN, 4 mM NADH, HSS (4.8 mg

ml"1) or F1 (4.8 mg ml"1) and variously treated F2 (0.2 mg ml'1) as

indicated. bOne ml of F2 fraction (0.55 mg) was treated with Protease E

(Sigma) at 37°C for 18 h at which time treatment was stopped by

adding EDTA. c The F2 fraction was subjected to acid hydrolysis with 0.5 N HCI at

95°C for 15 minutes.

CHAPTER IV

SUMMARY

The recovery of several cyanide-degrading enzyme activities

in 150,000 x g cell-free extracts of P. fluorescens NCIMB 11764

was found to be consistent with work published previously from this

laboratory indicating that cyanide was metabolized by several

different pathways (Kunz et al., 1992). Results showing that one of

these activities (CDA1), catalyzed the formation of ammonia and

C0 2 and was both oxygen and NADH dependent was consistent with

earlier work suggesting that the enzyme responsible for this

conversion was a cyanide oxygenase (CNO) (Harris and Knowles,

1983b). However, in addition to this enzyme the present work

showed that a second activity (CDA2) composed of two enzymes, a

cyanide dihydratase (CND) (generating formate and ammonia) and a

cyanide hydratase (CNH) producing formamide was also elaborated by

this organism (since it was not possible to distinguish between one

or two different enzymes this activity was referred to as the

combined CND/CNH activity). CND/CNH activity as expected, could be

detected under anaerobic conditions and without added reduced

pyridine nucleotide factors. A comparison of the apparent Km

values toward cyanide revealed a remarkable difference between the

two, with the putative CNO enzyme showing a 10-fold higher

80

81

affinity for cyanide than CND/CNH. This suggested that the putative

cyanide oxygenase plays a significant role in cyanide assimilation as

a growth substrate. Separate studies in our laboratory showing that

a mutant strain having lost the ability to grow on cyanide could no

longer synthesize this enzyme further supported this conclusion

(Kunz et al., 1994). In addition to putative CNO and CND/CNH

enzymes cyanide also induced the synthesis of formate

dehydrogenase. The formation of this enzyme in response to cyanide

has previously not been described and a role for this enzyme in the

recycling of reduced pyridine nucleotide (NADH) has earlier been

discussed (Figure 8).

Because of the central role of CNO in cyanide assimilation,

particular attention was paid to the identification and unequivocal

verification that this enzyme was a true oxygenase. Since earlier

conclusions (Harris and Knowles, 1983b) that this enzyme was an

oxygenase were based primarily on the reaction stoichiometry,

related experiments were performed. Similar results as those

reported earlier were obtained; that is, one mole each of oxygen and

NADH being consumed per mole cyanide degraded as expected for an

oxygenase enzyme (Hayaishi, 1982). However, to obtain absolute

proof of this it was necessary to show that molecular oxygen was

incorporated during catalytic conversion using isotopically labelled 1 8 0 2 - As summarized in Part II of this dissertation, after

painstaking efforts were taken to limit the extent of isotopic

exchange, convincing evidence that molecular oxygen was

incorporated during catalysis was obtained (Table 12). Parallel

82

experiments showed that oxygen was also incorporated from

labelled water, indicating that one atom each from molecular oxygen

and water are incorporated during cyanide oxidation to CO2 and NH3.

This work is significant for the following reasons. First, it verifies

the actual existence of an oxygenase in cyanide conversion, which

previously could only be inferred from biochemical studies. Second

it shows that this enzyme is a monooxygenase since only one atom

of molecular oxygen is incorporated. Third, since oxygen from water

is incorporated, it infers that an additional step between the

substrate cyanide and the reaction products, CO2 and NH3 must exist.

As discussed in Part II, we hypothesize that this additional step may

involve the hydrolysis of a transient oxygenated intermediate.

Finally, since cyanide disappearance catalyzed by extracts induced

primarily for CNO was observed only when oxygen and NADH were

present it may be concluded that monooxygenation rather than the

addition of water (in which case substrate disappearance would be

expected to occur anaerobically) represents the initial step in

cyanide degradative attack.

Inhibitor studies revealed that cyanide oxidation by high-speed

supernatants was inhibited by classical sulfhydryl-group inhibitors

(see Table 9). This observation is important because the role of an

essential sulfhydryl group in the chemical transformation of related

substrates containing the C=N group have also been proposed (for

example, ricinine nitrilases [E.C. 3.5.5.1] and other nitrilases that

hydrolyze nitriles to carboxylic acids by the addition of 2 moles of

water) (Harper, 1977; Hook and Robinson, 1966; Mahadevan and

83

Thimann, 1964). By analogy, we propose that a related mechanism

involving nucleophilic attack on cyanide by an essential enzyme

sulfhydryl group followed by monooxygenation as shown in equation

may be involved. Oxygenation would be expected to give rise to the

H • C=N • NH

• H " C . + 0 2 + NADH + H*

XE

NH

H O C , 0=C

-H

XE \

( i )

XE

H M * 0=C. \

XE

H.0

H H S k

H - O - C - O H

I? XE

H,0

/O H + A' + BH + 0 = C N ( I I )

XE

H

0 = C &

& ^XE H

0 = C = 0 Carbon Dioxide

+ A" + BH + EXH <NI>

0=C

5 H. U

C»

AS

0=C = N Cyanate

+ A" + BH + EXH (IV)

Scheme 6

corresponding imino ether, which on hydrolysis (equation II) would

give a tetrahedral intermediate. Ammonia would be removed as a

84

consequence of the interaction of adjacent acidic and basic groups

to leave the carboxylated enzyme, which in turn could give rise to

C 0 2 and the regenerated enzyme (equation III). Alternatively, if the

enzyme serves as the initial leaving group then cyanate (HOCN)

would be generated (equation IV). However, no evidence for cyanate

as intermediate was obtained during this work or other

investigations conducted in this laboratory. Therefore, the latter

mechanism seems unlikely. Further experiments with purified

enzyme are, of course, required to confirm the proposed enzyme

mechanism. Part III of this dissertation was directed at resolving

the enzyme(s) responsible for cyanide oxidation. We now think this

system consists of a cyanide monooxygenase (CNO) and a possible

additional enzyme necessary for conversion of an as yet unidentified

oxygenated intermediate. Efforts to resolve this system resulted in

the identification of two protein fractions, one including the bulk

protein (>30K) fraction (F1) and the other a small molecular weight

protein fraction (0.5-1 OK) (F2) of as yet unknown nature. Thus, at

this point we think that at least two protein components are

required. Findings are analogous to other oxygenases which usually

are multicomponent in nature (Gibson, 1988; Whited and Gibson,

1991).

The primary goal of this dissertation was to begin biochemical

investigations of cyanide-degrading enzymes at the cell-free level

in P. fluorescens NCIMB 11764. This has been accomplished with

the recovery and preliminary characterization of enzymes by

identification of reaction products and using isotope-labelling

85

methods. The results show that cyanide metabolism in this

organism is more sophisticated than previously recognized. At least

four different enzymes involved in cyanide metabolism were

discovered in P. fluorescens NCIMB 11764. This includes, CNO,

CND, CNH, formate dehydrogenase, and an enzyme catalyzing the

possible hydrolysis of an oxygenated intermediate generated by

cyanide monooxygenase. Of these enzymes CNO has been shown to be

most important for cyanide growth. Additional studies on this and

the other enzymes synthesized by P. fluorescens NCIMB 11764

should reveal further insight into the unique biochemistry and

physiology of cyanide metabolism in this as well as other potential

cyanide-degrading microorganisms.

REFERENCES

Alexander, M., and K. Volini. 1987. Properties of an E. coli rhodanase. J. Biol. Chem. 262:6595-6604.

Brysk, M. M., W. A. Corpe, and L. V. Hankes. 1969. [3-Cyanoalanine formation by Chromobacterium violacem. J. Bacteriol. 97:322-327.

Castric, P. A., K. F. Castric, and R. Meganathan. 1981. The metabolism of hydrogen cyanide by bacteria, p. 263-274. In B. Vennesland et al. (ed.), Cyanide in biology. Academic Press, Inc., New York.

Cornish-Bowden, A., and R. Eisenthal. 1974. Statistical considerations in the estimation of enzyme kinetic parameters by the direct linear plot and other methods. Biochem. J. 139:721-730.

Dagley, S. 1988. Microbial catabolism and the carbon cycle: a perspective, p. 3-13. In S. R. Hagedorn et a/.(ed), Microbial metabolism and the carbon cycle. Harwood Science Publishers, New York.

Dorr, P. K., and C. J. Knowles. 1989. Cyanide oxygenase and cyanase activities of Pseudomonas fluorescens NCIMB 11764. FEMS Microbiol. Lett. 60:289-294.

Dowd, J. E., and D. S. Riggs. 1965. A comparison of estimates of Michaelis-Menten kinetic constants from various linear transformations. J. Biol. Chem. 240:863-869.

Dunhill, P. M., and L. Fowden. 1965. Enzymic formation of p-cyanoalanine from cyanide by Escherichia coli extracts. Nature 208:1206-1207.

86

87

Eisenthal, R., and A. Cornish-Bowden. 1974. The direct linear plot: A new graphical procedure for estimating enzyme kinetic parameters. Biochem. J. 139:715-720.

Fallon, R. D., D. A. Cooper, R. Speece, and M. Henson. 1991. Anaerobic biodegradation of cyanide under methanogenic conditions. Appl. Environ. Microbiol. 57:1656-1662.

Fawcett, J. K., and J. E. Scott. 1960. A rapid and precise method for the determination of urea. J. Clin. Pathol. 13:156-159.

Ferry, J. G. 1990. Formate dehydrogenase. FEMS Microbiol. Rev. 87:377-382.

Finneagan, J., S. Toerlen, L. Abbot, F. Smit, and H. G. Raubenheimer. 1991. Identification and characterization of an Acinetobacter species capable of assimilation of a range of cyano-metal complexes, free cyanide and simple organic nitriles. Appl. Environ. Microbiol. 36:142-144.

Fry, W. E., and R. L. Millar. 1972. Cyanide degradation by an enzyme from Stemphylium loti. Arch. Biochem. Biophys. 151:468-474.

Fry, W. A., and D. C. munch. 1975. Hydrogen cyanide detoxification by Gloeocercospora sorghi. Physiol. Plant. Patho. 7:23-33.

Gibson, D. T. 1988. Microbial metabolism of aromatic hydrocarbons and the carbon cycle, p. 33-58. In S. R. Hagedorn et al. (3 ed.), Microbial metabolism and the carbon cycle. A symposium in honor of Stanley Dagley. Harwood Academic Publishers., Switzerland.

Harayama, S., M. Kok, and E. L. Neidle. 1992. Functional and evolutionary relationships among diverse oxygenases. Annu. Rev. Microbiol. 46:565-601.

88

Harper, D. B. 1977. Microbial metabolism of aromatic nitriles. Enzymology of C-N cleavage by Nocardia sp (rhodochrous group) N.C.I.B. 11216. Biochem. J. 165:309-319.

Harris, R., and C. J. Knowles. 1983a. Isolation and growth of a Pseudomonas species that utilizes cyanide as a source of nitrogen. J. Gen. Microbiol. 129:1005-1011.

Harris, R. E., and C. J. Knowles. 1983b. The conversion of cyanide to ammonia by extracts of a strain of Pseudomonas fluorescens that utilizes cyanide as a source of nitrogen. FEMS Microbiol. Lett. 20:337-341.

Harris, R. E., A. W. Bunch, and C. J. Knowles. 1987. Microbial cyanide and nirtile metabolism. Aci. Prog. Oxf. 71:293-304.

Hayaishi, O. 1982. My life and oxygen, retrospective and perspective. In Oxygenase and oxygen metabolism a symposium in honor for O. Hayaishi. P. 1-13. Academic Press, New York, N.Y.

Hendrickson, H. R., and E. E. Conn. 1968. Cyanide metabolism in higher plants. IV. Purification and properties of the (3-cyanoalanine synthase of blue lupine. J. Biol. Chem. 244:2632-2640.

Hook, R. H., and W. G. Robinson. 1966. Ricinine nitrilase. II. Purification and properties. J. Biol. Chem. 239:4263-4267.

Hope, K. M., and C. J. Knowles. 1991. The anaerobic utilization of cyanide in the presence of sugars by microbial cultures can involve an abiotic process. FEMS Microbiol. Lett. 80:217-220.

Hopner, T., and J. Knappe. 1974. Formate: determination with formate dehydrogenase, P. 1551-1555. Methods of enzymatic analysis, vol. 3. Academic Press Inc., New York.

89

Ingvorsen, K., B. Hojer-Pedersen, and S. E. Godtfredsen. 1991. Novel cyanide-hydrolyzing enzyme from Alcaligenes xylosoxidans subsp. denitrificans. Appl. Environ. Microbiol. 57:1783-1789.

Ivanoff, N. N., and E. S. Zwetkoff. 1936. The biochemistry of fungi. Annu. Rev. Biochem. 1:585-612.

Knowles, C. J. 1976. Microorganisms and cyanide. Bacteriol. Rev. 40:652-680.

Knowles, C. J. 1988. Cyanide utilization and degradation by microorganisms. CIBA Found. Symp. 140:3-15.

Knowles, C. J., and A. W. Bunch. 1986. Microbial cyanide metabolism. Adv. Microbial. Physiol. 27:73-111.

Kunz, D. A., and O. Nagappan. 1989. Cyanase-mediated utilization of cyanate in Pseudomonas fluorescens NCIMB 11764. Appl. Environ. Microbiol. 55:256-258.

Kunz, D. A., O. Nagappan, J. Silva-Avalos, and G. T. Delong. 1992. Utilization of cyanide as a nitrogen substrate by Pseudomonas fluorescens NCIMB 11764: evidence for multiple pathways of metabolic conversion. Appl. Environ. Microbiol. 58:2022-2029.

Kunz, D. A., D. W. Ribbons, and P. J. Chapman. 1981. Metabolism of allglycine and c/'s-crotylglycine by Pseudomonas putida (arvilla) mt-2 harboring a TOL plasmid. J. Bacteriol. 148:72-82.

Kunz, D. A., C. -S. Wang, and J. -L. Chen. 1994. Alternative routes of enzymic cyanide metabolism in Pseudomonas fluorescens NCIMB 11764. Microbiology. 140:1705-1712.

Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London)

90

227:680-685. Lambert, J. L., J. Ramasamy, and J. V. Paukstells. 1975.

Stable reagents for the colorimetric determination of cyanide by modified Konig reactions. Anal. Chem. 47:916-918.

Lennox, E. S. 1955. Transduction of linked genetic characters of the host bacteriophage P1. Virology 1:190-206.

Li, J., B. K. Burgess, and J. L. Corbin. 1982. Nitrogenase activity: Cyanide as substrate and inhibitor. Biochem. 21:4393-4402.

Lowry, O. H., N. J. Rosenbrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.

Macadam, A. M., and C. J. Knowles. 1984. Purification and properties of p-cyano-alanine synthase from the cyanide-producing bacterium, Chromobacterium violaceum. Biochim. Biophys. Acta 786:123-132.

Mahadevan, S., and K. Thimann. 1964. Nitrilase. II. Substrate specificity and possible mode of action. Arch. Biochem. Biophys. 107:62-68.

Meyers, P. R., P. Gokool, D. E. rawlings, and D. R. Woods. 1991. An efficient cyanide-degrading Bacillus pumilus strain. J. Gen. Microbiol. 137:1397-1400.

Meyers, P. R., D. E. Rawlings, D. R. Woods, and G. G. Lindsey. 1993. Isolation and characterization of a cyanide dihydratase from Bacillus pumilus CI. 1993:6105-6112.

Mills, G. A. and H. C. Urey. 1940. The kinetics of isotopic exchange between carbon dioxide, bicarbonate ion, carbonate ion and water. J. Am. Chem. Soc. 62:1019-1026.

Mudder, T. I., and J. L. Whitlock. 1984. Biological treatment of cyanidation wastewaters. Min. and Metal Proc. 161-165.

91

Nazly, N. and C. J. Knowies. 1981. Cyanide degradation by immobilized fungi. Biotechnol. Lett. 3:363-368.

Nazly, N., C. J. Knowies, A. J. Beardsmore, W. T. Taylor, and E. G. Corcoran. 1983. Detoxification of cyanide by immobilized fungi. J. Chem. Tech. Biotech. 33:119-126.

Neidhardt, F. C., J. L. Ingraham, and M. Schaechter. 1990. Physiology of the bacterial cell: a molecular approach. Sinauer Associates Inc., Publishers, Sunderland. Mass.

Pettet, A. E. J., and J. B. Ware. 1955. Disposal of cyanide wastes. Chem. Ind. 1955:1232-1238.

Pocker, Y., and D. W. Bjorkquist. 1977. Stopped-flow studies of carbon dioxide hydration and bicarbonate dehydration in H 2 0 and D20. Acid-base and metal ion catalysis. J. Am. Chem. Soc. 99:6537-6543.

Quayle, J. R. 1966. Formate dehydrogenase. Methods. Enzymol. 9:360-364.

Rangaswami, G., and A. Balasubramanian. 1963. Studies on the rhizosphere microflora of sorghum in relation to hydrocyanic acid content of roots. Can. J. Microbiol. 9:719-725.

Ressler, C., U. H. Giza, and S. N. Nigam. 1969. p-Cyanoalanine, product of cyanide fixation and intermediate in asparagine biosynthesis in certain species of Lathyrus and Vicia. J. Amer. Chem. Soc. 91:2766-2775.

Rodgers, P. B. 1981. Cyanide degradation by Chromobacterium violaceum. p. 301-310. In B. Vennesland et al. (ed.), Cyanide in biology. Academic Press, Inc., New York.

Rollinson, G., R. Jones, M. P. Meadows, R. E. Harris, and C. J. Knowies. 1987. The growth of a cyanide-utilizing strain of Pseudomonas fluorescens in liquid culture on nickel cyanide as a source of nitrogen. FEMS Microbiol. Lett. 40:199-205.

92

Ryan, R. W., and R. C. Tilton. 1977. Isolation of rhodanese from Pseudomonas aeruginosa by affinity chromatography. J. Gen. Microbiol. 103:197-199.

Sakai, T., H. Yanese, M. Sawada, and K. Tonomura. 1981. Formation of p-cyanoalanine by cyanide-resistant strain Enterobacter sp. 10-1. Agric. Biol. Chem. 45:2053-2062.

Sharp, A. G. 1976. The chemistry of cyano complexes of the transition metals. Academic Press, Inc. (London), Ltd., London.

Silva-Avalos J., M. G. Richmond, O. Nagappan, and D. A. Kunz. 1990. Degradation of the metal-cyano complex tetracyanonickelate (II) by cyanide-utilizing bacterial isolates. Appl. Environ. Microbiol. 56:3664-3670.

Silver, M., and D. P. Kelly. 1976. Rhodanese from Thiobacillus A2: Catalysis of reactions of thiosulphate with dihydrolipoate and dihydrolipoamide. J. Gen. Microbiol. 97:277-284.

Silverman, D. N., C. K. Tu, and N. Roessler. 1981. Diffusion-

limited exchange of 1 8 0 between CO2 and water in red cell suspensions. Resp. Physiol. 44:285-298.

Solomonson, L. P. 1981. Cyanide as a metabolic inhibitor, p. 11 28. In B. Vennesland et at. (ed.), Cyanide in biology. Academic Press, Inc., New York.

Sorbo, B. H. 1953. Crystalline rhodanese I. Purification and physicochemical examination. Acta. Chem. Scand. 7:1129-1136.

Strobel, G. A. 1967. 4-Amino-4-cyanobutyric acid as an intermediate in glutamate biosynthesis. J. Biol. Chem. 242:3265-3269.

93

Suh, Y-J., J. M. Park, and J-W. Yang. 1994. Biodegradation of cyanide compounds by Pseudomonas fluorescens immobilized on zeolite. Enzyme Microb. Technol. 16:529-533.

Ting, I. P., and W. C. Zschoche. 1970. Asparagine biosynthesis in cotton roots. Carbon dioxide fixation and cyanide incorporation. Plant Physiol. 45:439-444.

Towill, L. E., J. S. Drury, B. L. Whitfield, E. B. Lewis, E. L. Galyan, and A. S. Hammons. 1978. Reviews of the environmental effects of pollutants. V. Cyanide. U.S. Environmental Protection Agency publication no. 600/1-78-027. U.S. Environmental Protection Agency, Washington, D.C.

Tyler, B. 1978. Regulation of the assimilation of nitrogen compounds. Annu. Rev. Biochem. 47:1127-1162.

Umbreit, W. W., R. Burris, and J. F. Stauffer. 1964. Manometric techniques. Burgess Publishing company. Minneapolis, MN.

Wang, P., D. E. Mathews, and H. D. Van Etten. 1992. Purification and characterization of cyanide hydratase from the phytopathogenic fungus Gloeocercospora sorghi. Arch. Biochem. Biophys. 298:569-575.

Wang, P., and H. D. VanEtten. 1992. Cloning and properties of a cyanide hydratase gene from the phytopathogenic fungus Gloeocercospora sorghi. Biochem. Biophys. Res. Comm. 187:1048-1054.

Ware, G. S., and H. A. Painter. 1955. Bacterial utilization of cyanide. Nature 175:900.

Way, J. L. 1981. Pharmacologic aspects of cyanide and its antagonism, p. 29-49. In B. Vennesland et al. (ed.), Cyanide in biology. Academic Press, Inc., New York.

94

Westley, J. 1981. Cyanide and sulfane sulfur, p.61-76. In B. Vennesland et al. (ed.), Cyanide in biology. Academic Press, Inc., New York.

White, J. M., D. D. Jones, D. Huang, and J. J. gauthier. 1988. Conversion of cyanide to formate and ammonia by a pseudomonad obtained from industrial wastewater. J. Ind. Microbiol. 3:263-272.

Whited, G. M. and D. T. Gibson. 1991. Toluene-4-monooxygenase, a three-component enzyme system that catalyzes the oxidation of toluene to p-cresol in Pseudomonas mendocina KR1. J. Bacteriol. 173:3010-3016.

Yanese, H., T. Sakai, and K. Tonomura. 1982. Microbial utilization of p-cyano-L-alanine by Pseudomonas sp. 13. Agric. Biol. Chem. 46:2925-2932.

Yanese, H., T. Sakai, and K. Tonomura. 1983. Purification, crystallization and some properties of p-cyano-L-alanine-degrading enzyme in Pseudomonas sp. 13. Agric. Biol. Chem. 47:473-482.