Isolation and characterization of alpha-terpineol dehydratase from Pseudomonas gladioli

MISSING IMAGE

Material Information

Title:
Isolation and characterization of alpha-terpineol dehydratase from Pseudomonas gladioli
Physical Description:
xi, 157 leaves : ill. ; 29 cm.
Language:
English
Creator:
Cadwallader, Keith R., 1963-
Publication Date:

Subjects

Subjects / Keywords:
Limonene   ( lcsh )
Food additives   ( lcsh )
Citrus fruit industry -- By-products   ( lcsh )
Food Science and Human Nutrition thesis Ph. D
Dissertations, Academic -- Food Science and Human Nutrition -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 150-156).
Statement of Responsibility:
by Keith R. Cadwallader.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001709267
oclc - 25526628
notis - AJC1548
sobekcm - AA00004757_00001
System ID:
AA00004757:00001

Full Text











ISOLATION AND CHARACTERIZATION OF ALPHA-TERPINEOL
DEHYDRATASE FROM Pseudomonas gladioli










by


KEITH R. CADWALLADER


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1990














ACKNOWLEDGEMENTS


It has been a privilege and pleasure to have studied

for the past five years under the direction of Professor

Robert J. Braddock. I very much appreciate his patience,

guidance, and support. I wish to acknowledge the other

members of my graduate committee, Dr. Murat Balaban, Dr.

Mickey Parish, Dr. George Sadler, and Dr. Richard Yost, for

their guidance, support, and encouragement throughout this

study.

Special appreciation is extended to Dr. Kenneth Fox and

the FMC Corporation for their financial support of my

graduate studies through a very generous fellowship.

I would like to thank Maria Osorio for her assistance

with experiments during the course of this study.

Finally, I must give credit to my wife Karen, son

Myles, and the rest of my family for their faith, support,

and encouragement.















TABLE OF CONTENTS


Page

ACKNOWLEDGEMENTS ..................................... ii

LIST OF TABLES ..... ................................... v

LIST OF FIGURES ....................................... vi

ABSTRACT ......................... ...... .......... .. ix

INTRODUCTION ............................. .......... 1

REVIEW OF LITERATURE .................................. 6

Limonene ........................................... 6
Physical Properties ............................... 6
Production .................................... 6
a-Terpineol ................. ....... .... .... ..... 7
Physical Properties ....... ..................... 7
Production ................................... 7
Biotransformation of Limonene ..................... 8
Bacteria .. ..... .................... .......... 9
Fungi ...................... ... ................... 14
Plants ........................................... 15
Enzymatic Production of Flavors and Aromas ......... 16
Hydrolases ..................................... 16
Dehyrogenases and Reductases ................... 17
Enzymatic-Catalyzed Eliminations and
Additions of Water ...................... ........ 19
Kinetics ........................................ 19
Nomenclature ......................... ... ...... 22
Solubilization of Particulate-Associated Enzymes ... 23

MATERIALS AND METHODS ................................. 26

Chemicals ............................................. 26
Microbiology ....................................... 27
Liquid Mineral Medium Preparation .............. 27
Enumeration of Bacteria ......................... 27
Growth and Collection of Bacteria .............. 28
Analytical ............................................. 29
Protein Determination ......................... 29
Enzyme Assays .......................... ....... 29
Gas Chromatography ............................. 30

iii








Enzyme Purification ............................... 31
Buffer Preparation ............................. 31
Isolation of Particulate-Associated Enzyme ..... 31
Sucrose Gradient Centrifugation ................ 33
Solubilization of Enzyme ...................... 34
Gel Filtration Chromatography ................ 35
Concentration of Gel Filtration Fractions ...... 35
Removal of Triton X-100 by Affinity
Chromatography ................................ 36
Enzyme Characterization ............................. 36
Molecular Weight Determination ................ 36
Isoelectric Focusing ............... ........ 38
pH Optimum and Stability ......... ....... .... 39
Temperature Optimum and Stability............... 40
Stereospecificity and Stereoselectivity ........ 41

RESULTS AND DISCUSSION ............................... 43

Microbiology ..................................... 43
Isolation of Particulate-Associated Enzyme ........ 46
Enzyme solubilization ............................ 52
Gel Filtration Chromatography ... ................ 54
Partial Purification of Enzyme ..................... 64
Characterization of a-Terpineol Dehydratase ....... 70
Enzyme Nomenclature .......................... 70
Molecular Weight .............................. 71
Isoelectric Point .................. ...... 82
Activity in Organic Solvents ................. 87
pH Optimum and Stability ...................... 97
Temperature Optimum and Stability ............ 100
Initial Velocity ............................. 106
Effect of Triton X-100 on K, and V ........... 109
Stereospecificity and Stereoselectivity ....... 118

CONCLUSIONS ......................................... 130

APPENDIX .......... ........................... ........ 133

REFERENCES ..........................*................. 150

BIOGRAPHICAL SKETCH ................................... 157














LIST OF TABLES


Table Page

1 Cell density and enzyme activity of E.
gladioli as a function of culture age ........... 47

2 Amount of enzyme activity in particulate
fraction as a function of extent of cell
disruption ..................................... 48

3 Fractionation of a-terpineol dehydratase
using Spectra/Gel AcA 44 chromatography ......... 68

4 Fractionation of a-terpineol dehydratase
using Spectra/Gel AcA 22 chromatography ........... 69

5 Results of the molecular weight determination
of native a-terpineol dehydratase in 1.0%
(w/v) Triton X-100 by Sepharose CL-6B gel
filtration chromatography ....................... 77

6 Results of linear regression analyses of
double reciprocal (1/v versus l/[S])
Lineweaver-Burk plots obtained at various
Triton X-100 concentrations .................... 116

7 Relative percent concentration of
a-terpineol enantiomers as a function
of time for the hydration of 0.2 mM
racemic limonene by a-terpineol
dehydratase at 250C ............................ 126














LIST OF FIGURES


Figure Page

1 Reaction sequence of pathway 3 for the
metabolism of limonene by Pseudomonas putida ... 11

2 Solubilization and purification of
particulate-associated enzymes using
detergent solutions .............. .............. 25

3 Growth curve of P. gladioli at 30*C in pH 6.5
liquid mineral medium containing 0.2% (v/v)
limonene ........... ........ ..................... 45

4 Sucrose gradient centrifugation of
particulate-associated enzyme .................. 51

5 Plot of Triton X-100 concentration versus
degree of solubilization of protein
and enzyme .................................. .... 56

6 Plot of sodium trichloroacetate
concentration versus degree of
solubilization of protein and enzyme
in 2.0% (w/v) Triton X-100 ..................... 58

7 Spectra/Gel AcA 44 gel filtration
chromatography of solubilized a-terpineol
dehydratase using two concentrations of
Triton X-100 in the elution buffer:
(a) 0.5% (w/v) Triton X-100 and (b) 1.0%
(w/v) Triton X-100 .............. .............. 60

8 Spectra/Gel AcA 22 gel filtration
chromatography of solubilized a-terpineol
dehydratase .......................... .......... 63

9 Triton X-100 concentration versus catalytic
activity of a-terpineol dehydratase .............. 66

10 Sepharose CL-6B gel filtration chromatography
of a-terpineol dehydratase .................... 74

11 Calibration curve for Sepharose CL-6B column ... 76








12 SDS-PAGE gradient (8-18%) gel of
enzyme fractions ............................. 79

13 Calibration curve for SDS-PAGE gradient
(8-18%) gel .............................. ..... 81

14 Agarose IEF gel of partially purified
a-terpineol dehydratase in 1.0% (w/v)
Triton X-100 and 2.5% (w/v) Servalyte
3-10 ampholytes ............................... 84

15 Calibration curve for agarose IEF gel .......... 86

16 Effect of glycerol, formamide, and N,N-di-
methylformamide concentration on the activity
of a-terpineol dehydratase ..................... 90

17 Effect of 2-propanol and acetonitrile
concentration on the activity of a-terpineol
dehydratase .................................... 93

18 Effect of ethanol concentration on activity
and stability of a-terpineol dehydratase.
Activity curve: activity versus ethanol
concentration. Stability curve: activity after
removal of indicated ethanol concentration ..... 96

19 Effect of pH on activity and stability
of a-terpineol dehydratase. Activity curve:
activity versus pH. Stability curve:
activity at pH 7.0 after 25C pre-incubation
of the enzyme at indicated pH values ........... 99

20 Effect of temperature on the activity of
a-terpineol dehydratase: (a) plot of activity
versus temperature and (b) Arrhenius plot of
10-25C activity-temperature data .............. 103

21 Temperature stability of a-terpineol
dehydratase: activity at 20C after
pre-incubating enzyme at indicated
temperatures for 2 min ......................... 105

22 Formation of a-terpineol versus time for
three concentrations of a-terpineol
dehydratase at 25C and 50 mM limonene ....... 108

23 Initial velocity as a function of
a-terpineol dehydratase concentration
at 25C and 50 mM limonene ...................... 111


vii








24 Effect of varying the concentration of
Triton X-100 on Henri-Michaelis-Menten
(v versus [S]) curves for a-terpineol
dehydratase at 25C ............................ 113

25 Lineweaver-Burk double reciprocal (1/v versus
1/[S]) plots of the data in Figure 24 for
limonene concentrations from 1 mM to 10 mM ..... 115

26 GC profile of the change in concentration
of limonene and a-terpineol enantiomers
during the course of the hydration of
0.2 mM racemic limonene by a-terpineol
dehydratase at 25*C ............................... 123

27 Change in concentration of limonene and
a-terpineol enantiomers during the
course of the hydration of 0.2 mM
racemic limonene by a-terpineol
dehydratase at 25'C ............................ 125

28 Formation of a-terpineol enantiomers
during the course of the hydration of
50 mM racemic limonene by a-terpineol
dehydratase at 25*C ............................ 128

29 Stereo-specificity and -selectivity of
a-terpineol dehydratase ....................... 129


viii














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


ISOLATION AND CHARACTERIZATION OF ALPHA-TERPINEOL
DEHYDRATASE FROM Pseudomonas gladioli


by

Keith R. Cadwallader

December 1990

Chairman: R. J. Braddock
Major Department: Food Science and Human Nutrition

The enzyme-catalyzed hydration of the citrus by-

product, limonene, to the important flavor and aroma

chemical, a-terpineol, was investigated. Particulate-

associated a-terpineol dehydratase was recovered from

Pseudomonas gladioli, solubilized, and partially purified

using detergent extraction and gel filtration

chromatography.

Results of gel filtration chromatography suggested that

a-terpineol dehydratase existed in two soluble forms in 1.0%

(w/v) Triton X-100: a monomer and dimer with apparent

molecular weights of 94,500 and 206,500 daltons,

respectively. SDS-polyacrylamide gel electrophoresis

revealed that a 92,000 dalton polypeptide was enriched

during purification of the enzyme. Activity of a-terpineol








dehydratase was low in non-aqueous solvents. The reaction

was not readily reversible, since enzyme-catalyzed

dehydration of a-terpineol to limonene could not be

demonstrated, even in media with reduced water

concentration.

a-Terpineol dehydratase was characterized in buffers

containing 0.1% (w/v) Triton X-100. In 10 mM MES, 10 mM

BIS-TRIS PROPANE buffer, the pH optimum was 5.5 and the

stability optimum was pH 8.0. The pi of the enzyme was

between 6.5 and 6.8. The temperature optimum at pH 7.0 was

25C in 10 mM HEPES buffer. Using temperature-activity data

for 10-25*C, Ea and Q10 of a-terpineol dehydratase were

determined to be 21.6 2.9 kJ-mol"1 and 1.37 0.07,

respectively. Activity was inhibited by Triton X-100. The

effects were an increase in apparent K and decrease in

apparent Vmx. Average apparent Km of a-terpineol

dehydratase was 2.18 0.19 mM in 10 mM HEPES buffer, pH 7.0

containing 0.1% (w/v) Triton X-100.

a-Terpineol dehydratase stereospecifically catalyzed

the hydration of (4R)-(+)-limonene to (4R)-(+)-a-terpineol

or (4S)-(-)-limonene to (4S)-(-)-a-terpineol. The enzyme

was also stereoselective, since the rate of hydration of

(4R)-(+)-limonene was approximately ten times faster than

the rate of hydration of (4S)-(-)-limonene.

Isolation and characterization of a-terpineol

dehydratase has not been previously reported. Results of

this study will increase the understanding of








microbiological hydrations of monoterpenes. Considerable

economic potential exists, since the use of a-terpineol

dehydratase to produce pure a-terpineol increases the value

of limonene by 25%. This important natural process has an

advantage over a chemical synthesis of being both stereo-

selective and -specific.














INTRODUCTION


In recent years there has been increasing consumer

preference for food products containing "natural" flavors

over those containing artificial (synthetic) flavors. This

preference has led to an increased demand for natural flavor

and aroma chemicals (Buchel, 1989; Welsh et al., 1989). The

increased demand for natural flavors has in turn placed more

pressure on the production of flavor and aroma compounds by

extraction processes from traditional raw materials such as

plants. Use of these materials presents several problems.

For instance, plants often contain low concentrations of the

desired flavor compound, therefore making extraction

expensive. Furthermore, the supply may be subject to

seasonal, climatic, and geographical variation, as well as

political and socio-economic stability of the producing

regions (Armstrong and Yamazaki, 1986; Welsh et al., 1989).

Such factors have caused an increase in the price of natural

flavors and aromas.

According to the Code of Federal Regulations, compounds

produced or modified by living cells or by their components,

including enzymes, may be designated as natural (Code of

Federal Regulations, 21 CFR 101.22.a.3.). In any case, the

products can be considered natural if they are derived from







2

natural starting materials. Biotechnological or biological

processes (processes involving the use of microorganisms,

plant cell cultures, or enzymes) offer many possibilities

for the production of natural flavor and aroma chemicals.

In addition to their use for production of flavor and

aroma compounds, biological processes provide simple systems

for studying the biosynthetic pathways involved in the

formation of many important flavors and aromas. These

processes have several advantages over alternate physical or

chemical processes, the most important advantage being their

ability to catalyze specific reactions, thereby avoiding

potentially undesirable side reactions which may occur with

less specific processing methods. Another advantage is that

biological processes generally can be accomplished under

mild conditions (i.e., ambient temperature, atmospheric

pressure, and pH values near 7). This automatically results

in lower energy consumption or decreased substrate and

product damage.

The suitability of a process for production of flavor

and aroma chemicals depends on the market demand (total

usage), commercial value (price) of the chemical, and the

technological state of the process. The use of a biological

process for production of a-terpineol from limonene has

economic potential because the annual consumption of

a-terpineol is high, while at the same time its price is

higher than limonene.







3

The monoterpene hydrocarbon limonene [l-methyl-4-

(1-methylethenyl) cyclohexene, chemical formula C1OH16], is

ubiquitous in the plant kingdom. Limonene exists as both

optically active and racemic mixtures (dipentene). Limonene

is the main constituent of the terpene fraction of a number

of essential oils, among which are citrus, caraway, dill,

and American pine (Arctander, 1969a-c; Simonsen, 1947a).

Citrus essential oils are unusual because they contain pure

(4R)-(+)-limonene at concentrations approaching 95% for

orange and grapefruit oils (Shaw, 1979). In addition to its

high chemical and enantiomeric purity, limonene derived from

citrus is in abundant supply, with 8.7 million kilograms

being recovered during the 1988-89 Florida processing season

(Anon, 1989). Limonene from citrus is also relatively

inexpensive, the price being approximately 25% lower than

the price of a-terpineol (Anon, 1990).

a-Terpineol [a,a,4-trimethyl-3-cyclohexene-l-methanol,

chemical formula C1OH180] occurs in nature in both optically

active and racemic forms; the (4R)-(+)-enantiomer has been

identified as a component of petigrain, neroli, and sweet

orange oils, the (4S)-(-)-enantiomer of camphor oils, and

the racemic mixture of American pine and cajuput oils

(Arctander, 1969d; Simonsen, 1947b). a-Terpineol is one of

the most commonly used of all flavor and aroma chemicals.

Its annual consumption for flavor purposes has been

estimated at over 13,000 kg, which places it among the top

30 most commonly used flavors (Welsh et al., 1989).







4

a-Terpineol is commonly used in flavors such as lemon, lime,

nutmeg, orange, ginger, peach, and spices (Arctander,

1969d).

Presently, a-terpineol is commercially available only

as a racemic mixture, which is primarily recovered as a by-

product from the pulp and paper industry. The properties of

flavor and aroma compounds often depend on their optical

purity. For example, (4S)-(+)-carvone has properties

resembling caraway oil, whereas the properties of (4R)-(-)-

carvone resemble spearmint oil. Biological processes are

usually stereospecific; therefore it may be possible to

produce pure (4R)-(+)-a-terpineol from (4R)-(+)-limonene by

using this type of process. Cadwallader et al. (1989)

demonstrated that Pseudomonas gladioli produces pure (4R)-

(+)-a-terpineol from (4R)-(+)-limonene; however, the yield

of a-terpineol was low due to the utilization of limonene by

the bacterium for metabolic purposes. One way of directing

the conversion of limonene toward the production of only

a-terpineol would be by use of an isolated enzyme.

Suitability of an enzyme for the production of flavor

and aroma chemicals depends on its physical and kinetic

properties. It is necessary to isolate and purify enzymes

to relate these properties to important parameters affecting

reactions. For example, stability of enzymes should be

relatively high under the process conditions. These

conditions may include extremes of pH and temperature, as

well as the presence of solvents or other common protein







5
denaturants. Enzymes should also be highly specific for the

reaction of interest so that side products are minimized.

When it is desirable to produce pure enantiomers from

enantiomerically pure substrates, then the stereospecificity

of the enzyme is important. Stereoselectivity is important

when inexpensive racemic compounds are used as substrates

instead of more expensive pure enantiomers.

The above discussion particularly relates to the

importance of this research, which has the major objectives

of (1) to isolate an enzyme from P. gladioli which catalyzes

the hydration of limonene to a-terpineol, (2) to

characterize some of its physical and kinetic properties,

and (3) to define reaction conditions necessary for

production of pure a-terpineol enantiomers.













LITERATURE REVIEW


Limonene

Physical Properties

Limonene (F.W. 136.24) is a colorless liquid with

boiling point 178C (CRC, 1986). It is practically

insoluble in water (13 ppm at 25*C, Massaldi and King,

1973), soluble in alcohol, miscible with oils, but is poorly

soluble in propylene glycol and glycerol (Arctander, 1969c).

Production

Limonene is produced in quantity as both racemic and

(4R)-(+)-limonene. Racemic limonene (commonly referred to

as dipentene) is isolated by fractional distillation of

American Pine oil and rosin oils. It is also recovered as a

by-product from the production of a-terpineol and from

various synthetic products made from a-pinene or turpentine

(Arctander, 1969a; Mattson, 1984).

(4R)-(+)-Limonene (commonly referred to as d-limonene)

is recovered as a by-product from the manufacture of citrus

molasses. Molasses refers to the concentrate produced from

the press liquor, which is expelled from citrus waste

residues after curing with small quantities of lime. Press

liquor, which contains from 0.20 to 0.50% limonene, is

normally passed through preheaters (115-140C) for







7
pasteurization and scale removal and flashed at atmospheric

pressure (100*C) to recover the limonene. The condensate

(60-80% limonene) is put into a closed florentine-type tank

which allows the limonene to float to the top of the tank

where it is continuously decanted into storage tanks or

drums (Kesterson and Braddock, 1976; FMC, 1976).

a-Terpineol

Physical Properties

a-Terpineol (F.W. 154.25) is a colorless, slightly

viscous liquid with melting point 40-41'C and boiling point

220-C (CRC, 1986). It is slightly soluble in water (1980

ppm at 15-20'C, Seidell, 1928), but is soluble in alcohol,

propylene glycol, and mineral oil (Arctander, 1969d).

Production

a-Terpineol is produced primarily in racemic form

either by isolation from American Pine oil or by chemical

synthesis from a-pinene via terpin hydrate (Arctander,

1969d) or directly to a-terpineol by hydration (Arctander,

1969d; Mattson; 1984). Non-commercial synthetic methods for

the production of a-terpineol from limonene include (1)

oxymercuration [aqueous tetrahydrofuran containing Hg(OAc)2

and limonene] followed by reduction with NaBH4 to give 70%

a-terpineol (Brown et al., 1972) and (2) hydration of (4R)-

(+)-limonene using chloroacetic acids to (4R)-(+)-

a-terpineol (Matsubara et al., 1975). Methods involving

acyclic starting materials include (1) cyclization of

pentane tricarboxylic acid followed by esterification, via






8

the hydroxyester to the unsaturated ester, and then reaction

with Grignard reagent to a-terpineol; and (2) by reaction of

isoprene and methyl vinyl ketone with methyl magnesium

iodide (Arctander, 1969d).

Biotransformation of Limonene

Biological approaches for the production of flavor and

aroma compounds can be divided into biosynthetic and

biotransformation methods. Biosynthesis refers to the

production of chemical compounds by cell metabolism

(fermentation or secondary metabolism by plant or microbial

cells), whereas biotransformation is defined as the use of

living cells or enzymes to perform specific modifications or

conversions of chemical compounds (Welsh et al., 1989).

Research concerned with the microbial metabolism of

terpenes was initiated by Bradshaw et al. (1959) and Seubert

(1960). Since that time there has been a great number of

publications in this area. Therefore, this section will be

devoted to discussion of those studies involving the

biotransformation of limonene.

Terpene flavor compounds are synthesized by a variety

of higher plants and microorganisms. Low yield of terpenes

produced by plant cell cultures or microbial fermentations,

in addition to the abundance of natural terpenes from

botanical sources, makes the development of biosynthetic

processes for flavor and aroma terpene production

unfavorable. Microorganisms and enzymes have the greatest

potential for production of natural terpene flavors and






9

aromas via transformations of inexpensive natural terpene

precursors to more valuable terpenes. Furthermore, microbes

can achieve highly selective biotransformations of readily

available substrates leading to products which are rare or

otherwise difficult to synthesize.

Some bacteria and fungi are capable of utilizing

terpenes as their sole source of carbon and energy. Terpene

biotransformations are most often carried out by

pseudomonads because these microbes have the ability to

synthesize a wide variety of oxygenases and related enzymes

necessary for the metabolism of many xenobiotics.

Bacteria and fungi differ in their metabolism of

limonene. Bacteria appear to metabolize limonene primarily

by progressive oxidation of the 7-methyl group. Generation

of small amounts of neutral products occurs to some extent.

The purpose of these "side" reactions is unknown, since

these compounds are not further metabolized. Fungi attack

limonene primarily by hydration of the double bond of the

isopropenyl substituent or by epoxidation-hydrolysis of the

1,2-double bond. a-Terpineol and 1,2-diols are the most

common products of fungal metabolism of limonene. Whether

these compounds are intermediates or end products of minor

pathways is uncertain.

Bacteria

The metabolism of limonene by a strain of Pseudomonas

putida was studied by Dhavlikar and Bhattacharyya (1966) and

Dhavlikar et al. (1966). The bacterium converted limonene






10

into several neutral and acidic products. The neutral

products carveol, carvone, dihydrocarvone, l-p-menthene-6,9-

diol, cis- and trans-8-p-menthene-l,2-diol, and 8-p-menthen-

l-ol-2-one were incapable of supporting the growth or

respiration of the bacterium; however, the acidic products

perillic acid, 2-hydroxy-8-p-menthen-7-oic acid,

B-isopropenyl pimelic acid, and 6,9-dihydroxy-l-p-menthen-7-

oic acid were readily metabolized. These researchers

proposed that P. putida metabolizes limonene by three

distinct pathways, involving (1) allylic oxygenation, (2)

oxygenation of the 1,2-double bond, and (3) progressive

oxidation of the 7-methyl group to perillic acid which

subsequently undergoes hydration, dehydrogenation, and

hydrolysis. The reaction sequence of pathway 3 is

illustrated in Figure 1.

The enzymes in pathway 3 were demonstrated in cell-free

sonicates of the bacterium. The NADPH- and oxygen-dependent

C-7 hydroxylation of limonene was shown to be associated

with the 100,000 x g sediment of the sonicated cells. All

other enzymes were present in the supernatant fraction. A

cytochrome P-450 mixed function oxidase system was isolated

from the same organism which catalyzes the C-7 hydroxylation

of p-cymene (Madhyastha et al., 1968). Similarities between

the structures of limonene and p-cymene suggest this

cytochrome P-450 system, or one which is similar,

catalyzes the C-7 hydroxylation of limonene to perillyl

alcohol.









0

U
+0
I >





A
0) O
O






o
o o 0
CL 0
^400 \ E rl
4r







t
0.0


0
44


ao










0 40
C 0


o ( / -- 5




i .



'I
-1 4 )






I '-I






12

An NAD-linked alcohol dehydrogenase, perillyl alcohol

dehydrogenase, was isolated from P. putida (Ballal et al.,

1966). Partially purified (6-7 fold) enzyme was examined

for its substrate specificity and cofactor requirement. The

enzyme was absolutely specific towards NAD, with NADP being

inactive. The enzyme demonstrated a broad substrate

specificity. The following structural features of the

substrate were found to favor dehydrogenation: (1) a

primary alcohol group, preferably allylic to an endocyclic

double bond, (2) a six-membered ring, and (3) an alkyl

substituent in the para position.

An aldehyde dehydrogenase, perillyl aldehyde

dehydrogenase, was also isolated from this organism (Ballal

et al., 1967). The partially purified (10 fold) enzyme was

highly specific towards NAD with relatively less activity

with NADP. Besides perillyl aldehyde, this enzyme showed

activity on several other aldehydes. No specific structural

requirements were observed with respect to the substrate

specificity of the aldehyde dehydrogenase as compared to

that of the alcohol dehydrogenase.

The metabolism of limonene by Pseudomonas incognita was

studied by Rama Devi and Bhattacharyya (1977a). Major

conversion products were perillic acid and B-isopropenyl

pimelic acid. Small amounts of neutral products were

recovered, but were not identified. Rama Devi and

Bhattacharyya (1977b) demonstrated that the pathway of

degradation of limonene by this bacterium occurred in the






13

following sequence: limonene perillyl alcohol perillic

acid B-isopropenyl pimelic acid -* -. CO2 + H20, i.e. same

as pathway 3 for the metabolism of limonene by P. putida

(see Figure 1).

Fermentation of limonene by an enterobacterium was

shown to result in the formation of dihydroperillic acid,

perillic acid, and traces of unidentified neutral products

(Dhere and Dhavlikar, 1970). The formation of perillic acid

suggests that this organism metabolizes limonene using a

similar pathway as P. putida.

Corynebacterium hydrocarboclastus has been reported to

oxidize limonene to carvone (Takagi et al., 1969). The lack

of proper controls (i.e., experiments to test for the

autoxidation of limonene) and the very low yield of carvone

(32 mg from 5.9 g limonene, in 30 hr) suggests that the

appearance of carvone in the medium may have been due to

chemical oxidation of limonene. Carvone has been shown to

be an autoxidation product of limonene (Buckholz and Daun,

1978).

Cadwallader et al. (1989) demonstrated that Pseudomonas

gladioli converts (4R)-(+)-limonene into (4R)-(+)-perillic

acid, (4R)-(+)-a-terpineol, and at least one other

unidentified acidic product. a-Terpineol was found to be

resistant to further degradation, whereas perillic acid was

readily degraded. This suggests that P. gladioli may

metabolize limonene using a similar pathway as that

described for P. putida.






14

P. gladioli is not the only bacterium capable of

converting limonene into a-terpineol. Murdock et al. (1967,

1969) and Murdock and Hunter (1970) demonstrated that in

some citrus oil emulsions limonene concentration diminished

in direct proportion to an increase in a-terpineol

concentration as a result of microbial growth. Several of

these microbes were isolated but were not identified.

Fungi

Cladosporium spp. (T12) has been shown to convert (4R)-

(+)-limonene into pure (4R)-(+)-a-terpineol (Kraidman et

al., 1969). Similarly, Penicillium digitatum (DSM 62840)

was shown to produce (4R)-(+)-a-terpineol from either (4R)-

(+)-limonene or racemic limonene (Abraham et al., 1985).

This could be explained by the exclusive hydration of (4R)-

(+)-limonene.

Epoxidation-hydrolysis of the 1,2-double bond of

limonene leads to 1,2-diols (p-menth-8-ene-l,2-diols).

Mukherjee et al. (1973) demonstrated that Cladosporium spp.

(T7) converts limonene to trans-l,2-diol plus a small amount

of the corresponding cis-1,2-diol.

Corynespora cassiicola (DSM 62475) and Diplodia

gossypina (ATCC 10936) were found to be stereospecific for

the conversion of limonene to 1,2-diols. (4R)-(+)-limonene

was converted to the (1S,2S)-diol, while (4S)-(-)-limonene

yielded the (lR,2R)-diol (Abraham et al., 1984; Abraham et

al., 1985).






15

The conversion of limonene by Penicillium italicum and

P. digitatum resulted in low yields of 1,2-diols with

respect to the other products isolated (Bowen, 1975). Cis-

and trans-carveol were in highest yield followed by cis- and

trans-p-mentha-2,8-diene-l-ol, carvone, p-menthena-2,8-

diene-l-ol, carvone, p-mentha-l,8-diene-4-ol, and perillyl

alcohol, respectively.

Plants

In addition to microorganisms, isolated plant enzymes

and plant cell cultures have been shown to convert limonene

to more important flavor compounds. Knorr et al. (1990)

described the biotransformation of limonene to carvone by

dill (Anethum graveolens) cultures. Maximum carvone

production occurred with 5 mM limonene after 4 days of

treatment. Prolonged treatment for 8 to 14 days resulted in

degradation of carvone. It was proposed that this technique

would be feasible for a continuous process involving

continuous product recovery.

Microsomal preparations containing limonene hydroxylase

activity were isolated from peppermint (Mentha piperita),

spearmint (Mentha spicata), and perilla (Perilla frutescens)

leaves (Karp et al. 1990). Each microsomal preparation

generated only one product from limonene: trans-

isopiperitenol from M. piperita, trans-carveol from M.

spicata, and perillyl alcohol from P. frutescens. These

compounds originated from hydroxylation of limonene at C-3,

C-6, and C-7, respectively. The three hydroxylases had an






16

absolute requirement for molecular oxygen and NADPH. Both

(4R)-(+)- and (4S)-(-)-limonene were. hydroxylated by the

three enzymes.

Enzymatic Production of Flavors and Aromas

Enzymes have traditionally been used as processing aids

in the food industry. Most widely used enzymes are

amylases, glucose oxidases, proteases, pectic enzymes, and

lipases. Reviews on production and utilization of food

enzymes are available (Wasserman, 1990; Whitaker, 1972).

Largest use of enzymes for flavor production is for the

production of syrups and sweeteners. Manufacture of glucose

syrup and high fructose corn syrup is done solely by

enzymatic means. This accounts for about 25% of the total

usage of enzymes worldwide (West, 1987). Hydrolases,

dehydrogenases and reductases have demonstrated the greatest

potential for the production of flavor and aroma compounds.

Excellent reviews of these enzymes are available (West 1987;

Welsh et al., 1989).

Hydrolases

Hydrolases perform reactions such as esterifications,

transesterifications, hydrolyses, and lactonizations

(internal ester formation). Lipases, esterases, and

proteases hydrolyze fat, carbohydrate, and protein

molecules, respectively. Some of the most potent flavor

compounds are esters. Natural esters retail at much higher

prices than their synthetic counterparts. Lipases have been

used to perform stereospecific hydrolyses and






17

esterifications to yield pure, optically active aliphatic

and aromatic esters, alcohols, acids, and lactones (Welsh,

1989).

Lipases have been shown to mediate esterifications and

transesterifications in organic media (Cambou and Klibanov

1984; Engel et al., 1989; Gerlach et al. 1988; Yokozeki et

al., 1982). Enzymatic resolution of enantiomers can be

achieved with some esterases and may represent an

alternative to traditional chemical purification techniques

(Engel et al., 1989; Omata et al., 1981)

Esterifications and transesterifications can be

mediated by proteases. Proteases have been used to create

unique hydrolyzed proteins which enhance or add flavors,

particularly savory flavors, to foods. Use of proteases to

produce modified food proteins has been recently reviewed

(Alder-Nissen, 1985). Catalytic behavior of proteases can

be modified by their use in organic solvents. Rate

enhancements afforded by subtilisin and a-chymotrypsin for

transesterification in octane were of the order of 100

billion-fold (Zaks and Klibanov, 1988). Similarly, peptide

synthesis was favored instead of hydrolysis when

a-chymotrypsin was reacted in ethanol or acetonitrile (Kisee

et al., 1988).

Dehydroqenases and Reductases

Only two oxido-reductases have been seriously

considered for flavor and aroma production (Welsh et al.,

1989). These enzymes, alcohol dehydrogenase and alcohol






18

oxidase, perform the oxidation of aliphatic alcohols to

their respective aldehydes. Alcohol dehydrogenase (ADH) can

potentially be produced in commercial quantities from yeasts

or plant cell cultures (West, 1987). The conversion of

geraniol to geranial stands out as a typical conversion

which can be mediated by ADH (Legoy et al., 1985).

The main drawback to using oxido-reductases is that

these enzyme require cofactors for their catalytic function.

Cofactors may act as either cosubstrates or active site

participants. For many cofactors, especially those which

are oxidized or reduced during the reaction, the initial

cost of the cofactor prevents one-time usage on a large

scale. High cost requires cofactor recycling if the process

is to be economical. Recycling of NADH can be achieved by

electrochemical means or substrate-driven reactions (Legoy

et al., 1985; Nakamura et al., 1988); however, cofactor

regeneration rates are not high enough for commercial

purposes.

To expand the range of possible processes and to

improve the economics of current enzyme processes, increased

knowledge is needed concerning enzyme isolation and

characterization, mechanisms of enzyme action, and

incorporation of enzyme processes for natural flavor and

aroma production. Specific needs are to understand the

mechanisms of enzyme activation/inactivation, and to utilize

enzymes in processes and redox reactions relevant to flavor

production, including low-cost production and recycling of

cofactors.







19

Enzyme-Catalyzed Eliminations and Additions of Water

Kinetics

Aconitase, fumarase, enolase, and crotonase are

examples of enzymes that catalyze reversible dehydration/

hydration reactions. This type of reaction can be

illustrated by the hydration of limonene to a-terpineol:






+ H20


OH


Limonene a-Terpineol

The biochemistry of enzymes is a subject of great depth

and complexity. This section will attempt to describe only

the aspects of enzyme kinetics applicable to the present

study. For additional information, the reader may find

greater detail elsewhere (Hammes, 1978; Segel, 1976; Walsh,

1979).

Kinetic equations can be readily derived assuming the

simplest mechanism, i.e. that involving a single substrate

and product. In fact, this mechanism is valid for the

hydration of limonene to a-terpineol if it is assumed that

water is always present in saturating amounts. The general

mechanism is as follows:

k, k2
E + S -- ES E + P
k.i k.2








d[ES]
------ = (k2 + k._) [ES] k,[E][S] k.2[E][P] (1-1),
dt

where S is the substrate, P is the product, E is the

free enzyme, and ES is the enzyme-substrate complex. Since

ES is rapidly destroyed, and therefore is present at very

low concentration, the steady state condition d[ES]/dt = 0

can be assumed. Also, conservation of mass requires that

[Et] = [E] + [ES] (1-2)

and

[St] = [S] + [P] + [ES] (1-3),

where Et and St represent total enzyme and total substrate,

respectively. Since [S] >> [ES] z [E], [ES] in equation

(1-3) can be neglected. The rate equation for the

disappearance of S or the appearance of P can be written as

follows:

d d
---[S] ----[P] = kI[E][S] k.,[ES] (1-4).
dt dt

Setting equation (1-1) equal to zero and substituting in

equation (1-4) gives

d d (klk2[S] k.lk_2[P])[Et]
--[S] = ----[P] =---------------- (1-5).
dt dt k1[S] + k-2[P] + k_1 + k

Measurements of the reaction velocity, v, are typically

carried out at initial times, when [S] >> [P]. Therefore,

the terms containing the concentration of product can be

neglected in equation (1-5) and

d k2[Et]
v ---S] = -----------------(1-6).
dt 1 + (k_1 + k2)/(k,[S])






21

Defining Vmx = k2[Et] and Km = (k_. + k2)/kl, then equation

(1-6) takes the form

Vx Vax [S]
V = ------- = ---(1-7),
1 + KO[S] K + [S]

which is known as the Henri-Michaelis-Menten equation.

Equation (1-7) shows dependence of v on [S], since as

[S] co, v Vmx, and also when [S] = Km, v = Vmx/2.

Michaelis constant, Km, is not an equilibrium constant but a

"steady-state" constant and measures the ratio of steady

state concentrations [E][S]/[ES]. Maximum initial velocity,

Vx, is directly proportional to [Et].
The numerical value of KY is important because it

represents an approximate value for the intracellular

concentration of the substrate. There would be little

advantage in maintaining a substrate concentration higher

than Km since activity cannot exceed Vax. The difference

between the activity at [S] equal to Km and at [S] equal to

1000Km is only two-fold. This property of YK can be

utilized when designing an enzymatic process. The K is

constant for a given enzyme and can be used to make

comparisons between enzymes from different sources. The K,

can also be used to indicate the relative suitability of

alternate substrates for a particular enzyme. The substrate

with the lowest K, would have the highest apparent affinity

for the enzyme. The "best" substrate is that which has the

highest Vax/K ratio (Segel, 1976).








Nomenclature

Nomenclature and classification of enzymes is

accomplished using rules and guidelines established by the

Nomenclature Committee of the International Union of

Biochemistry (1984). Enzymes which cleave C-C, C-O,

C-N and other bonds by elimination leaving double bonds or

rings, or conversely adding groups to double bonds, are

classified as lyases. The systematic name is formed

according to 'substrate group-lyase'.

Carbon oxygen lyases catalyze the breakage of C-O bonds

leading to unsaturated products. In the case of hydro-

lyases, this is by elimination of water or by elimination of

an alcohol from a polysaccharide. The name 'dehydratase' is

recommended for those enzymes which eliminate or add water

to double bonds. The name 'synthase' may be used instead

when it is desired to emphasize the synthetic aspect of the

reaction or when the reverse reaction is much more

important.

Where equilibrium warrants it, or where the enzyme has

traditionally been named after a particular substrate, the

reverse reaction may be used as the basis of the name using

'hydratase'. In the case of reversible reactions, the

direction chosen for naming should be the same for all the

enzymes in a given class, even when the direction has not

been demonstrated.






23

Solubilization of Particulate-Associated Enzymes

This section is devoted to methods of approaching the

problem of solubilizing enzymes or proteins which are

associated with insoluble parts of the cell, such as

membranes.

Particulate-associated proteins fall into two general

categories. The first of these includes those proteins

which, once solubilized, behave much like any other water-

soluble enzyme. This category consists of two types of

proteins: those which are present in a space surrounded by

a membrane, but are not true components of the membrane

itself; and those enzymes which are intrinsic components of

a membrane, but upon being released are water-soluble.

Procedures for the solubilization of these types of proteins

have been summarized by Penefsky and Tzagoloff (1971).

The second category includes those proteins which are

components of lipoprotein complexes. General procedures for

the solubilization and purification of these proteins have

been summarized by Tzagoloff and Penefsky (1971).

Solubilization can be accomplished using detergents, in

which the detergent binds to the protein and takes the place

of the lipid-containing membrane, thus solubilizing the

protein in the medium. During purification it is often

necessary to include detergent in solutions to maintain

protein solubility. Solubilization and purification of

proteins using detergents is illustrated in Figure 2.






24

The use of chaotropic agents for the isolation of

membrane-bound enzymes has been discussed by Hatefi and

Hanstein (1974). Chaotropic agents decrease water structure

thereby destabilizing membranes and enzyme complexes. Some

commonly used chaotropic agents include sodium

trichloroacetate, sodium and potassium perchlorate, and

guanidine hydrochloride.
















"I>?~
\I/ ,r "


'1"


"7 1

I 7'nz H


"7

N,


I-
z
U
0
U N
UJ

0


*%1-\7

'iN


I-

uJ
0
CC

UJ
L Ut
in


10

N
4)


4-)

0

It





*4
0




0

.p
(4

I

0



t0
0


.4o
U







0 4
-l



0




HO
0,
a)


Or





t0 *














MATERIALS AND METHODS


Chemicals

(4R)-(+)-Limonene was a gift from Hercules Inc.,
24
Lakeland, FL; [a]D +122.5 (neat), 99.0% pure by gas

chromatography (GC). (4S)-(-)-Limonene was obtained from
19
Aldrich Chemical Company Inc., Milwaukee, WI; [a]D -86.6

(neat), 97.1% pure by GC. a-Terpineol racemicc) was a gift
24
from International Flavors, Union Beach, NJ; [a]D -10.2

(c=5.1, chloroform), 98.7% pure by GC.

(4R)-(+)-a-Terpineol was prepared by fermentation of

(4R)-(+)-limonene by P. gladioli. Bacteria were grown in a

group of 9 flasks as described under Growth and Collection

of Bacteria, except that after an initial 24 hr incubation

period 1 mL of (4R)-(+)-limonene was added to each flask and

the flasks incubated for an additional 6 days. The culture

broth was centrifuged for 20 min at 15,000 x g to remove

bacteria. The bacterial pellet was resuspended in 200 mL

distilled water and centrifuged as described above. The

clarified culture broth and pellet wash supernatant were

pooled and then extracted with diethyl ether using the

method of Cadwallader et al. (1989). Crude a-terpineol

extract (65% pure by GC) was purified by silica gel column

chromatography. a-Terpineol in n-hexane (2.5 mL) was







27
applied to a slurry packed silica gel column (2.0 x 20 cm)

which was equilibrated in n-hexane. The column was

developed using the following sequence of solvents: 100 mL

n-hexane, 100 mL of 50% n-hexane/50% methylene chloride, and

200 mL methylene chloride. Fractions (20 mL) were collected

and analyzed by GC. Fractions containing the highest

amounts of a-terpineol were pooled and the solvent

evaporated under a stream of nitrogen to yield 0.2150 g of
24
purified (4R)-(+)-a-terpineol; [a]D +102.0 (c=l.l,

chloroform), 94.0% pure by GC.

Microbiology
Liquid Mineral Medium Preparation

Liquid mineral medium (Monod and Wollman, 1947) was

prepared by adding the following chemicals consecutively to

3 L of distilled water until completely dissolved: KH2PO4

(12 g), Na2HPO4 (108 g), MgSO4 (10.8 g), NH4C1 (16 g), CaCI2

(8 mL of a 1% (w/v) solution), and FeSO4*7H20 (4 mL of a

fresh 0.1% (w/v) solution). The solution pH was adjusted to

6.5 with 85% phosphoric acid and then diluted to 4 L to make

4X strength liquid mineral medium. To make single strength

liquid mineral medium 1 L of 4X liquid mineral medium was

diluted to 4 L with distilled water. The pH was measured

and adjusted if necessary to pH 6.5 with 85% phosphoric

acid.

Enumeration of Bacteria

Microbial populations (viable cells) were estimated

according to procedures described by the American Public







28

Health Association (1989) using Standard Methods agar (BBL

Microbiology Systems, Cockeysville, MD). Serial dilutions

were made in 0.1% peptone. Aliquots of each dilution (0.1

mL or 1 mL) were transferred to duplicate petri dishes and

pour plated with 15-20 mL of agar tempered at ca. 45*C.

Plates were incubated at 30C for 48 hr.

Bacterial growth was also monitored by measuring the

absorbance (660 nm) of the culture broth.

Growth and Collection of Bacteria

P. gladioli was grown in 1 L Erlenmeyer flasks

containing 250 mL liquid mineral medium, 0.5 mL sterile

(4R)-(+)-limonene, and 25 mL inoculum. Flasks containing

liquid mineral medium were sterilized at 121*C for 20 min.

Limonene was filter-sterilized using a 0.22 Am MSI membrane

filter (Micron Separations Inc., Westboro, MA). Flasks were

incubated for 24 hr at 30 10C on an Environ orbital shaker

(Lab-Line Instruments, Inc., Melrose Park, IL) at 200 rpm

with 2 cm orbit. Inoculum was grown under the same

conditions as described above. A total of 9 flasks were

incubated per batch with one flask from each batch being

used as the inoculum of the following batch.

Bacteria were collected by centrifuging the culture

broth for 20 min at 15,000 x g in a Sorvall RC-5B

refrigerated centrifuge (Dupont Co., Newtown, CT). The

yield of bacterial cells from 8 flasks (2.2 L of broth) was

about 6-7 g. Cells were stored at 4'C.







29

Analytical

Protein Determination

Protein concentrations were measured by the Coomassie

blue protein assay (Bradford, 1976) using bovine serum

albumin as the standard.

Enzyme Assays

In a typical enzyme assay the sample was transferred

into a test tube (13 x 100 mm) and the final volume of the

sample adjusted to 0.990 mL with 10 mM N-2-hydroxyethyl-

piperazine-N'-2-ethane sulfonic acid (HEPES) buffer, pH 7.0.

The reaction was started by addition of 10 AL of substrate

solution which contained 5 M limonene and 50 ppm decanol

(internal standard) in absolute ethanol. The tube was then

quickly sealed with a PTFE lined screw cap, mixed by

vortexing, and placed in a temperature-controlled water

bath. Unless otherwise stated, assays were done for 2 min

at 25*C. The reaction was stopped by addition of 1 mL of

n-hexane, which denatured the enzyme. Anhydrous sodium

sulfate (3 g) was added to break the hexane emulsion. The

hexane layer was transferred to a 1 mL vial and concentrated

to about 0.1 mL under a stream of nitrogen. A small amount

of anhydrous sodium sulfate was then added to the vial to

completely dry the concentrated hexane phase before GC

analysis.

Sample blanks were prepared by mixing 1 mL of n-hexane

with the sample to denature the enzyme before adding 10 gL

of substrate solution. Limonene blanks consisted of 0.990







30

mL buffer (typically 10 mM HEPES buffer, pH 7.0) and 10 pL

of substrate stock solution. Limonene blanks were incubated

under the same conditions as enzyme assays.

a-Terpineol concentration was determined by comparing

the a-terpineol/decanol GC area ratio of the unknown with a

calibration curve of a-terpineol/decanol GC area ratio

versus a-terpineol concentration (Amol.L'1 or nmol-mL'1).

A typical standard curve for a-terpineol is shown in Figure

A-l. The curve is linear for concentrations of a-terpineol

between 1.04 and 522 Amol*L'.

The enzyme activity (nmol.min'1.mL"1) was calculated by

dividing the concentration of a-terpineol (nmol.mL-')) by the

assay time (min). Specific enzyme activity (nmol-min'1.mg'1

protein) was calculated by dividing the enzyme activity

(nmol-min'l.mL'1) by the protein concentration (mg.mL'1).

Gas Chromatographv

The GC system consisted of an HP 5890 gas chromatograph

equipped with a flame ionization detector (FID) and

connected to an HP 3396A integrator (Hewlett-Packard Co.,

Avondale, PA). Chromatograms were analyzed using a computer

integration program (Chrom-Perfect, Justice Innovations,

Palo Alto, CA). Unless otherwise stated, a 0.53 mm i.d. x

15 m SE 54 fused silica capillary column (Alltech

Associates, Inc., Deerfield, IL) was used for the

separations. The film thickness of the liquid phase was

1.25 Am. Conditions were as follows: 5 AL injection with

1:3 split; helium carrier gas at 5.5 mL-min'1; injector port







31

at 175C; detector at 250*C; column temperature programmed

from 100"C to 200' at 10O'Cmin'1 with a final hold time of 5

min. A gas chromatogram of a typical enzyme assay extract

is shown in Figure A-2.

Enzyme Purification

Buffer Preparation

All buffers were prepared at room temperature (20-25C)

using double distilled water. The pH of each buffer was

adjusted with NaOH or HC1 before its final dilution. The pH

measurements were made at room temperature using an Accumet

915 pH meter (Fisher Scientific, Pittsburgh, PA) with

automatic temperature compensation.

Isolation of Particulate-Associated Enzyme

The first step in isolation of a bacterial intra-

cellular enzyme is to disrupt the organism. There are

several techniques available for the disruption of bacteria:

sonication, bead-milling, French press, and digestive

enzymes such as lysozyme. Initial attempts to disrupt the

bacterium involved a combined nonionic detergent-osmotic

shock-lysozyme treatment as described by Schwinghamer (1980)

and Scopes (1987). This technique was only partially

successful in disrupting the bacterium and required a long

incubation period. Ultrasonication was also attempted, but

the conditions required to obtain adequate disruption of the

bacterium were very vigorous and resulted in loss of enzyme

activity, possibly due to heat denaturation. Small scale

bead-milling resulted in adequate bacterial disruption and

was used for initial studies.







32

The general procedure used for small scale bead-milling

is described by Scopes (1987) and involves the use of small

glass beads (0.1 mm diameter), which when vigorously stirred

along with a suspension of bacteria disintegrate the

bacteria by grinding action. Glass beads (8 g) were added

to centrifuge tubes (28.5 x 104 mm) along with cell

suspensions (1 g wet cells in 5 mL of 20 mM potassium

phosphate buffer containing 5 mM MgSO4 and 100 jig-mL'1

DNaseI) and the mixtures vortex-mixed for 10-20 min.

Suspensions were centrifuged for 30 min at 20,000 x g in a

Sorvall RC-5B refrigerated centrifuge in order to sediment

the enzyme. All operations were carried out as close to 0C

as possible, except for centrifugation which was carried out

at 4-60C.

Large scale bacterial disruption was achieved using a

commercially available bead-mill. Bacterial cells (20-30 g)

were re-suspended in 200 mL of 50 mM HEPES buffer, pH 7.0

containing 0.2 mM dithioerythritol (DTE). The suspension

was transferred, along with 500 g of dry glass beads (0.1 mm

diameter), to a Bead-Beater (Biospec Products, Bartlesville,

OK) equipped with a stainless steel chamber. The bacteria

were disrupted for a total of 20 min (30 sec on, 60 sec off)

with the chamber immersed in an ice bath. The cell

homogenate was separated from the glass beads by vacuum

filtration through nylon mesh filter cloth and the beads

washed with an additional 50-100 mL of buffer. After adding

2.5 mL of 1 M MgSO4 and 1 mL of a DNaseI solution (2500







33

units*mL"' in distilled water) the cell homogenate was

centrifuged for 20 min at 15,000 x g in a Sorvall RC-5B

refrigerated centrifuge in order to sediment unbroken cells

and large debris. The supernatant was further clarified by

re-centrifuging under the same conditions. The purpose of

DNaseI was to reduce the viscosity of the cell homogenate.

Magnesium sulfate was added because Mg2 is a required

cofactor of DNaseI.

In order to sediment the membranous material containing

the enzyme, the clarified supernatant was re-centrifuged for

2 hr at 25,000 rpm in a Beckman Model L5-65 ultracentrifuge

using a Beckman SW28 rotor (Beckman Instruments, Inc., Palo

Alto, CA). The pellet from this centrifugation was re-

suspended and washed twice with 70-75 ml of 10 mM HEPES

buffer, pH 7.0 containing 0.2 mM DTE, with centrifugation

the same as previously described. All operations were

carried out as close to 0C as possible, except for

centrifugation which was carried out at 4-60C.

Sucrose Gradient Centrifugation

The method of Osborn and Munson (1971) was used for

sucrose gradient centrifugation with some modifications.

Sucrose step gradients were prepared in ultracentrifuge

tubes (14 x 89 mm) by layering 2.1 mL each of 50, 45, 40,

35, and 30% (w/v) sucrose over a cushion (0.5 mL) of 55%

(w/v) sucrose. Sucrose solutions were prepared in 5 mM

HEPES buffer, pH 7.0 containing 5 mM EDTA. Samples were

prepared by re-suspending washed membrane pellets in 25%







34

(w/v) sucrose solution to a protein concentration of 2.2

mg/mL. One mL of sample was layered onto each of three

tubes. One mL of 25% (w/v) sucrose was layered onto a forth

tube for buoyant density determinations. Tubes were

centrifuged for 18 hr at 35,000 rpm in a Beckman Model L5-65

ultracentrifuge using a Beckman SW 41 rotor (Beckman

Instruments, Inc.).

Gradients were fractionated using an ISCO Model 640

density gradient fractionator (ISCO Inc., Lincoln, NB).

Fractions (0.6 mL) were assayed for both enzyme activity and

protein concentration. Buoyant densities were determined by

measurements of refractive index using an Abbe Mark II

refractometer (Rehert Scientific Instruments, Buffalo, NY).

All operations were carried out as close to 0OC as possible,

except for centrifugation which was carried out at 4-60C.

Solubilization of Enzyme

Washed membrane pellets were re-suspended in 25-50 mL

of 10 mM HEPES buffer, pH 7.0 containing 2.0% (w/v) Triton

X-100, 0.5 M sodium trichloroacetate and 0.2 mM DTE.

Suspensions were kept on ice and stirred for 30-60 min, then

diluted with additional buffer until the protein

concentration was equal to 5-10 mg*mL-1. Suspensions were

centrifuged for 1 hr at 55,000 rpm (100,000 x g) in a

Beckman TL-100 ultracentrifuge using a Beckman TLA100.3

rotor (Beckman Instruments, Inc.). All operations were

carried out as close to *0C as possible, except for

centrifugation which was done at 4'C. Supernatants







35

containing the solubilized enzyme were decanted and kept at

4C.

Gel Filtration Chromatoaraphv

Solubilized enzyme (20-30 mL) was applied to a 5.0 x

50 cm Spectra/Gel AcA 44 or AcA 22 column (Spectrum Medical

Industries, Inc., Los Angeles, CA) which was equilibrated in

10 mM HEPES buffer, pH 7.0 containing 1.0% (w/v) Triton

X-100, 0.2 mM DTE, and 0.02% (w/v) sodium azide. The enzyme

was eluted from the column at a flow rate of 70-75 mL-hr"'

and 10 mL fractions were collected. Fractions were assayed

for enzyme activity (30 min at 25C) and protein

concentration. All operations were carried out in a 4C

cold room.

Concentration of Gel Filtration Fractions

Gel filtration fractions containing enzyme activity

were pooled and concentrated using a 150 mL capacity

Omegacell ultrafiltration (UF) unit (Filtron Technology

Corp., Clinton, MA) with a nominal molecular weight limit

(NMWL) of 10,000 daltons. Nitrogen at 30-40 psi was used to

pressurize the unit. Solutions containing the pooled

fractions were concentrated to 10 mL and then 100 mL of 10

mM HEPES buffer, pH 7.0 was added and the solution

reconcentrated to 10 mL. This procedure was repeated until

the UF permeate had an absorbance (275 nm) of less than 1.0.

All operations were carried out in a 4C cold room.







36
Removal of Triton X-100 by Affinity Chromatography

Enzyme concentrate (10-15 mL) was applied to a 1.6 x

6.5 cm Spectra/Gel D column (Spectrum Medical Industries,

Inc.) which was equilibrated in 10 mM HEPES buffer, pH 7.0.

The enzyme was eluted from the column at a flow rate of 5

mL*hr"1. An additional 10 mL of 10 mM HEPES buffer, pH 7.0

was applied to the column in order to elute any remaining

enzyme. The column was regenerated using the procedure

recommended by the manufacturer, as follows: Column was

washed with 1 bed volume distilled water, 1 bed volume of

ethanol, 2 bed volumes of butanol, 1 bed volume of ethanol,

1 bed volume of distilled water followed by equilibration in

buffer. All operations were carried out in a 40C cold room.

Triton X-100 concentration in the Spectra/Gel D treated

enzyme solution was estimated by measuring the absorbance

(275 nm) of the UF permeate. Standard curves of absorbance

versus Triton X-100 concentration for standard solutions and

their respective UF permeates are shown in Figure A-3. The

UF permeate curve was estimated by a straight line for

Triton X-100 concentrations between 0 and 0.05% (w/v) and

was used to estimate the Triton X-100 concentrations of

sample enzyme solutions.

Enzyme Characterization

Molecular Weight Determination

The molecular weight (MW) of the native enzyme was

determined by gel filtration chromatography. Solubilized

enzyme (0.75 mL) was applied to a 1.0 x 50 cm Sepharose







37

CL-6B column (Aldrich Chemical Company, Inc.) which was

equilibrated in 10 mM HEPES buffer, pH 7.0 containing 1.0%

(w/v) Triton X-100, 0.2 mM DTE, and 0.02 % (w/v) sodium

azide. The enzyme was eluted from the column at a flow rate

of 5 mL.hr"' and 0.4 mL fractions were collected.

Fractions were assayed for both enzyme activity (30 min at

250C) and protein concentration. The calibration curve was

prepared with bovine thyroglobulin (MW = 669,000 daltons),

horse spleen apoferritin (MW = 443,000), sweet potato

B-amylase (MW = 200,000), yeast alcohol dehydrogenase

(MW = 150,000), and bovine erythrocyte carbonic anhydrase

(MW = 29,000). Protein standards were obtained from Sigma

Chemical Company, St. Louis, MO. The void volume was

determined with Blue Dextran 2000. Protein standards

(5 mg.mL'') and Blue Dextran 2000 (10 mg.mL'1) were dissolved

in gel filtration buffer. Blue Dextran 2000 fractions were

monitored by measuring the absorbance (600 nm) after

dilution of each fraction with 2.5 mL of distilled water.

Gel filtration was carried out in a 4*C cold room.

The molecular weight of the enzyme in SDS denaturing

conditions was determined by horizontal SDS polyacrylamide

gel electrophoresis (SDS-PAGE) using ExcelGel SDS gradient

(8-18% acrylamide) precast gels and buffer strips (Pharmacia

LKB Biotechnology, Uppsala, Sweden). The electrophoresis

system consisted of an LKB 2217 Multiphor II horizontal

electrophoresis unit and an LKB 2297 Macrodrive S constant

power supply (Pharmacia LKB Biotechnology). The running







38

conditions were as follows: 15C; 50 mA constant current;

80 min. Gels were silver-stained using a PhastGel Silver

Kit (Pharmacia LKB Biotechnology). The calibration curve

was prepared using Bio-Rad SDS-PAGE high and low molecular

weight standards (Bio-Rad Laboratories, Richmond, CA).

Isoelectric Focusing

The isoelectric point of the enzyme was determined by

horizontal isoelectric focusing (IEF) using FisherBiotech

agarose IEF gels and electrode solutions for pH range 3 to

10 (Fisher Scientific). The electrophoresis system was the

same as that used for SDS-PAGE. Gels were equilibrated for

1 hr in a solution containing 2.5% (w/v) Servalyte 3-10

ampholytes (Fisher Scientific) and 1.0% (w/v) Triton X-100.

In order to remove any salt or buffer ions which would

disturb the pH gradient during focusing, enzyme solutions

were dialyzed against either distilled water or 1.0% (w/v)

glycine, pH 7.0 (Vesterberg, 1971). Enzyme concentrate (1

mL) was dialyzed against 2 x 250 mL of 1% (w/v) glycine

buffer, pH 7.0 containing 1% (w/v) Triton X-100 (using

dialysis tubing with MWCO 12,000-14,000 daltons) or against

2 x 250 mL distilled water containing 1.0% (w/v) Triton

X-100 (Vesterberg, 1971). Dialyzed samples were filtered

thru 0.2 Am Acrodisc filters (Gelman Sciences, Ann Arbor,

MI) to remove any debris before focusing. Samples were

applied at various regions of the gel using a large template

(1 x 1 cm wells), which was supplied with the IEF gels.

Calibration curves were prepared using Bio-Rad broad range







39

IEF standards (Bio-Rad Laboratories). Gels were focused at

10C using the following running conditions: 3 W/5 min, 5

W/30 min, 10 W/30 min, 15 W/30 min, and 25 W/5 min.

The gels were halved immediately after focusing. One

gel half, which contained one lane of focused enzyme sample

and two lanes of focused IEF standards, was soaked in a

solution of 15% (w/v) trichloroacetic acid to fix the

protein bands and then stained with Coomassie G-250. The

other gel half, which contained 4 lanes of focused enzyme

sample, was cut horizontally into 7.5 mm strips and each

strip immediately suspended in 2 mL of 50 mM HEPES buffer,

pH 7.0. The pH of each suspension was adjusted to 7.0 and

the activity assayed (2 hr at 250C).

pH Optimum and Stability

Enzyme concentrate (30 mL) was dialyzed against

distilled water (1 L) containing 0.1% (w/v) Triton X-100

for 12 hr at 4*C using dialysis tubing with molecular weight

cutoff (MWCO) of 12,000-14,000 daltons. The pH optimum of

the enzyme was determined as follows: Aliquots of dialyzed

enzyme (0.25 mL; 1.9 mg.mL'' protein) were diluted to 2.5 mL

in 10 mM 2-(4-morpholino)-ethane sulfonic acid (MES), 10 mM

1,2-bis[tris(hydroxymethyl)methylamino]-propane (BIS-TRIS

PROPANE) buffer, with pH values ranging from 4.0 to 9.0 by

increments of 0.5; The pH of each solution was measured and

adjusted if necessary with 0.5 M HCl or 0.5 M NaOH to within

0.02 units of the indicated pH; Each solution was incubated

for 30 min at 25*C and then assayed for activity (30 min at







40

25C). The combined MES/BIS-TRIS PROPANE buffer was

suitable for pH range 5.5 to 9.5.

The pH stability of the enzyme was determined using a

similar procedure as above except for the following:

Dialyzed enzyme aliquots (0.2 mL) were diluted to 2.0 mL in

each buffer; solutions were incubated for 30 min at 250C and

then assayed for activity after adjusting the pH to 7.0 by

addition of 0.5 mL of 1 M HEPES buffer, pH 7.0.

Temperature Optimum and Stability

Temperature optimum of the enzyme was determined by

assaying the enzyme (1.7 mg*mL"1 protein in 10 mM HEPES

buffer, pH 7.0 containing 0.1% (w/v) Triton X-100) for

activity at various temperatures (10-50C) after

equilibration for 2 min. The temperature-activity data

obtained from 10-25C was used to estimate the activation

energy (E,) using an Arrhenius plot (log enzyme activity

versus l/T, where T is the absolute temperature in degree

Kelvin). The slope of this plot is equal to -Eg(2.3R),

where R is the gas constant (8.31441 JK-'1-mol-1). The

activation energy (J.mol"') was calculated from the slope.

The Q10 (the factor by which the rate increases by

raising the temperature 10*C) of the enzyme was calculated

using the following equation:

Q10 = [activity at (X + 10) C]/[activity at XC].

The temperature stability of the enzyme was determined

as follows: Enzyme (2.0 mg-mL-1 protein in 10 mM HEPES

buffer, pH 7.0) was incubated for 2 min at various







41

temperatures (20-60C), immediately cooled on ice, and then

assayed for activity at 20'C.

Stereospecificity and Stereoselectivity

Enzyme assays were the same as described earlier except

that the substrate stock solution was composed of 0.02 M
25
racemic limonene ([Qa]D +0.7; neat) and 1000 ppm (1S)-(-)-

cis-pinane (internal standard) in absolute ethanol. Assays

were done at 25"C for 5, 10, 20, and 40 hr. The GC system

was the same as previously described except that a 0.254 mm

i.d. x 30 m Cyclodex B fused silica capillary column (J & W

Scientific, Inc., Folsom, CA) was used for the separation of

limonene and a-terpineol enantiomers. The film thickness of

the liquid phase was 0.25 Am. Conditions were as follows: 1

gL injection with 1:100 split; helium carrier gas at 1.26

mL'min''; injector port at 200"C; detector at 250'C; column

temperature programmed from 70C to 200'C at 5'C*min"1 with

an initial hold time of 10 min and a final hold time of 5

min.

For determination of initial kinetic rates, enzyme

assays were done using a substrate stock solution composed

of 5 M racemic limonene and 500 ppm n-decanol (internal

standard) in absolute ethanol. Assays were done at 250C for

2, 5, and 10 min. GC conditions were the same as above

except for the following: 1 AL injection with 1:10 split;

helium carrier gas at 0.92 mL*min'1; column temperature

isothermal at 115"C. A chiral gas chromatogram of a typical

enzyme assay extract is shown in Figure A-4.







42

A calibration curve of a-terpineol/decanol GC area

ratios versus a-terpineol concentration (pmol-L'') is shown

in Figure A-5. The actual concentration of each

a-tetpineol enantiomer was calculated from its relative

concentration (i.e. the GC peak area of each enantiomer

divided by the sum of the GC peak areas for both

enantiomers). Enzyme activity was calculated using the same

method previously described under Enzyme Assays.














RESULTS AND DISCUSSION


Microbiology

Pseudomonas gladioli, which was used throughout this

study, was the same bacterium described by Cadwallader et

al. (1989). The bacterium exhibited optimum growth at pH

6.5 and 300C in both tryptic soy broth and in liquid mineral

medium containing 0.2% (v/v) limonene (Figure A-4). The

concentration of limonene in the medium also affected the

growth of the bacterium, with optimum growth occurring at a

concentration less than 1.0% (v/v) (Cadwallader et al.,

1989).

In this study the bacterium was grown exclusively in pH

6.5 liquid mineral medium containing 0.2% (v/v) limonene

with incubation at 30C. A typical growth curve of the

bacterium under these conditions is shown in Figure 3. The

exponential growth phase of the bacterium occurred between 6

and 18 hr with maximum number of viable cells occurring at

about 24 hr. The ideal time for collecting cells is toward

the end of the log phase before the growth rate slows, since

this gives the highest yield of cells. The enzyme may not

be at maximum concentration at this time. Preliminary

studies determined which physiological state of the

bacterium contained the most enzyme. Cultures of various
































Figure 3. Growth curve of E. gladioli at 30'C in pH 6.5
liquid mineral medium containing 0.2% (v/v)
limonene.








10.4


10.0




9.6


9.2


8.8




8.4


8.0


0 12 24 36


...J
E

IL
0

0
0


48


60


Time (hr)







46
ages (24, 48, and 72 hr) were examined for cell density and

enzyme activity. Enzyme assays (12 hr at 30C) were

conducted on cells suspended in 20 mM potassium phosphate

buffer, pH 7.0 (1 g wet cells in 5 mL buffer). Results

(Table 1) showed no significant difference (a = 0.05) in

enzyme activity between the 24 hr and 72 hr cultures;

however, there was a significant decline (a = 0.05) in cell

density after 72 hr. It was concluded from these results

that 24 hr was the most suitable time for collecting the

bacteria.

Isolation of Particulate-Associated Enzyme

After centrifugation of the bacterial extract the

majority of the enzyme activity was found in the particulate

fraction. Initial small scale bead-mill experiments

demonstrated that the enzyme activity of the particulate

fraction was due to bound enzyme and not due to surviving

bacteria. Enzyme activity and viable cells were determined

before and after bead-milling for various times (Table 2).

After 10 min of bead-milling greater than 99.999% of the

bacteria were disrupted, while the amount of activity in the

particulate fraction did not decrease. The supernatant

fraction, which contained over 20 mg*mL'1 protein, showed

only 7.9% of the total activity. The high protein

concentration of the supernatant fraction was another

indication of efficient cell disruption. Bead-milling the

suspension for a total of 20 min resulted in an additional

99.98% reduction of viable cells over that observed at







47

TABLE 1

Cell density and enzyme activity of P. gladioli
as a function of culture age


Culture Cell Density Enzyme Activit
Age (hr) (mg.mL-1)a (nmol.min- mg'
wet cells)


24 3.44 0.16 0.68 0.06

48 3.53 0.09 0.52 0.06

72 2.79 0.29 0.65 0.07


average standard deviation (n = 2)

average standard deviation (n = 4)









TABLE 2

Amount of enzyme activity in particulate fraction
as a function of extent of cell disruption


Extraction Enzyme Activity Viable Cells
Time (min) (nmol.min' .mL ')a (Log CFU.mL"1)b


0 4.54 0.17 13.2

10 4.7 0.6 8.0

20 2.66 0.08 4.2


average standard deviation (n = 2)


average (n = 4)







49
10 min. The decrease in enzyme activity after 20 min was

possibly due to denaturation of the enzyme, since only 2.6%

of the total activity was found in the supernatant fraction.

Small scale bead-milling was very efficient, but the

sample processing capacity was low. The technique was

scaled up by utilizing a laboratory scale bead-mill. This

unit was capable of disrupting up to 100 g of cells (wet

wt.) in 200 mL medium. Disruption of 100 g of cells was not

feasible because of the low yield of cells (6-7 g) from each

batch of culture. The technique effectively disrupted

20-30 g of cells, but was apparently much more vigorous than

the small scale method, since a higher centrifugal force was

required to sediment the enzyme. This was not necessarily a

disadvantage since a lower centrifugal force could be used

to remove unbroken cells and large debris from the cell

homogenate prior to sedimentation of the enzyme.

Isolated particulate-associated enzyme was partially

purified by washing the pellet with a buffer which did not

solubilize the enzyme. The particulate material was

characterized by differential centrifugation. This

technique was also evaluated for its potential as an enzyme

purification step. Results of sucrose gradient

centrifugation are shown in Figure 4. Protein was

effectively fractionated into four bands. The intensity of

these bands was highest at the sucrose layer interfaces.

Enzyme activity was found in each band, with the majority of

the activity being found in the two bands of lowest density.



























N



4)
4,


0
u

02

C1





4-)
0












,4
O
m












-4



44
0















C)



0






*a)
An
4-4
104
Ow















(0

ty





oc
Lr












-,-- (_w.o-6) A,!sueQ


N 0 c D 0







0***..* (L-'LtU!Wit-oWuu) AI!IA~I*O


-- (L_.Wu--6u) u!ieOJd


I-




qL.

u-
.2




L0
U-


0 0 0 0 0 0 0 0
I O D I O O M 1 O- O
0 d d d d d0d






52

The results of this experiment suggest that the enzyme

was associated with particulate material of varying density

and size. One explanation that might account for these

results is that the enzyme was associated with a specific

membrane of the bacterium and during bead-milling the

membrane was sheared into smaller heterogeneous fragments.

On the other hand, the enzyme may not have been membrane-

bound, but instead became associated with its nearest

hydrophobic neighbor after being released into the

hydrophilic medium (buffer). The utility of sucrose

gradient centrifugation as a purification step was limited

by the low sample processing capacity and the time required

to achieve equilibrium (18 hr). In addition, the degree of

purification obtained was low (less than two-fold);

therefore it was decided to solubilize the enzyme without

any prior fractionation.

Enzyme Solubilization

Initial attempts to solubilize the enzyme involved

those techniques described by Penefsky and Tzagoloff (1971)

for water-soluble membrane proteins, but it was evident

after several experiments that the enzyme was not

characteristic of this type of protein. Extraction of the

particulate-associated enzyme with concentrated salt (2 M

NaCl and 6 M urea), alkali/EDTA, enzymes (phospholipase A2,

lysozyme, and pectolyase) and organic solvents (DMSO,

ethanol, acetone, n-butanol, and dioxane) had no effect on

the solubility of the enzyme. Extraction with organic

solvents actually denatured the enzyme.







53
Partial solubilization of the enzyme could be achieved

using detergents. Nonionic detergents, such as Triton

X-100, Nonidet P-40, and n-octyl a-D-glucopyranoside were

more effective then the ionic detergents; sodium cholate,

SDS, and 3-[(3-choramidopropyl)dimethyammonio]-l-

propanesulfonate (CHAPS). Triton X-100 was chosen for

further enzyme solubilization studies. The degree of

solubilization of the enzyme was found to be dependent on

the concentration of Triton X-100 with 2.0% (w/v) giving

more consistent results than 1.0% (w/v) (Figure 5).

Nonionic detergents have been reported to be more effective

when used in combination with salts (Tzagoloff and Penefsky,

1971); however, inclusion of various concentrations of NaCl,

KC1 or urea with 2% (w/v) Triton X-100 did not increase the

solubility of the enzyme. Inclusion of the chaotropic

agent, sodium trichloroacetate, in extraction buffers

greatly increased the effectiveness of Triton X-100 for

solubilizing the enzyme. Effect of including various

concentrations of sodium trichloroacetate in solutions of

2.0% (w/v) Triton X-100 on the solubility of the enzyme is

shown in Figure 6. Optimum concentration of sodium

trichloroacetate was found to be 0.5 M when the protein

concentration of the initial enzyme suspension was less than

10 mg.mL". Sodium trichloroacetate (0.5 M) alone had no

effect on the solubility of the enzyme.







54

Gel Filtration Chromatography

Initial attempts to purify the enzyme using anion

exchange (DEAE Spectra/Gel) and hydroxylapatite (Bio-Gel

HTP) chromatography were unsuccessful due to irreversible

adsorption of the enzyme, which occurred even when 1.0%

(w/v) Triton X-100 was included in the elution buffers. The

degree of success of gel filtration chromatography for

enzyme purification depended on the concentration of Triton

X-100 in the elution buffer. Gel filtration profiles of

solubilized enzyme using Spectra/Gel AcA 44 with 0.5 and

1.0% (w/v) Triton X-100 in the elution buffer are shown in

Figures 7a and 7b. With 0.5% detergent only 12-46% of the

enzyme activity was recovered, whereas with 1.0% detergent

yields close to 100% were obtained.

Low recovery of activity with 0.5% detergent was

apparently due to adsorption of the enzyme to the gel. This

is evident from the greater volume required to elute the

enzyme with 0.5% detergent compared to with 1.0% detergent.

With 0.5% detergent the enzyme eluted at the leading edge of

the last and largest peak. This was the inclusion volume of

the column and was composed mainly of excess detergent

originating from the solubilized enzyme solution. After

being adsorbed to the gel, the enzyme was apparently

partially re-solubilized by the concentrated detergent

volume. With 1.0% detergent the inclusion volume appeared

as a negative peak. This was probably composed of excess

sodium trichloroacetate originating from the solubilized

enzyme solution.

































Figure 5. Plot of Triton X-100 concentration versus degree
of solubilization of protein and enzyme. (Buffer
consisted of 50 mM HEPES, pH 7.0 and 0.2 mM DTE.)









60.0


55.0


50.0-..
5 0 .0 .................... ...... .. .


45.0


C 40.0
O

N o35.0


2 30.0
O
O)
S 25.0 ---- Protein


S 20.0 ......... Enzyme Activity
0)

15.0


10.0


5.0


0.0o ------I
0.0 0.5 1.0 1.5 2.0 2.5 3.0

Triton X100 (% w/v)












100.0 ............................................
9 0.0 ..... ..... o. o **.

90.0



80.0--



- 70.0

O
0
-,I
4 60.0



S50.0 -


0
0 40.0-

L.

0 30.0
O Protein

0.0 ........ Enzyme Activity
20.0



10.0--

0.0


0.00 0.25 0.50 0.75 1.00

Sodium Trichloroacetate (M)






























Figure 7.


Spectra/Gel AcA 44 gel filtration chromatography
of solubilized a-terpineol dehydratase using two
concentrations of Triton X-100 in the elution
buffer: (a) 0.5% (w/v) Triton X-100 and (b) 1.0%
(w/v) Triton X-100. (Elution buffer consisted of
10 mM HEPES, pH 7.0, 0.2 mM DTE, and 0.02% (w/v)
sodium azide.)










1.00 1.25


0.80 -1.00

(a) 0
::-

S0.60-- 0.75 E


E 0.40- 0.50
S*CE
00

2 0.20 0.25 Z


0.00 I 0.00
0 100 200 300 400 500 600 700

Elution Volume (mL)



1.25 1.25

1.00- 1.00

0.75 (b) -0.75

0.50 -- --0.50
E C
1 0.25-- -- 0.25I
E 0.25
S0.00 0.00





0 100 200 300 400 500 600 700
Elution Volume (mL)







61
The purpose of gel filtration chromatography was not

only for purification of the enzyme, but also as a criterion

to establish the solubility of the enzyme in 1.0% (w/v)

Triton X-100. It is generally excepted that an enzyme

preparation is "soluble" if it remains in the supernatant

solution after centrifugation at 100,000 x g for one or more

hours (Penefsky and Tzagoloff, 1971); however, small

membrane fragments which do not sediment readily may appear

to be soluble. Additional evidence of solubility can be

obtained by passage of the enzyme preparation through a gel

filtration column (Eiberger and Wasserman, 1987). The

enzyme should elute within the linear fractionation range of

the column. This is because large insoluble material would

elute in the exclusion volume or void volume, unless it

precipitates in the column or adsorbs to the gel.

When passed through Spectra/Gel AcA 44 in 1.0% (w/v)

Triton X-100 the enzyme eluted in the void volume. The

linear fractionation range of AcA 44 is 10,000-130,000

daltons and the exclusion limit is 200,000 daltons.

Therefore, the enzyme either eluted as an insoluble enzyme-

particulate complex or its molecular weight was larger than

the exclusion limit of the gel. When the enzyme was

fractionated using Spectra/Gel AcA 22 (with 1.0% (w/v)

Triton X-100 in the elution buffer) it eluted as two peaks

(Figure 8). The linear fractionation range of AcA 22 is

100,000-1,200,000 daltons and the exclusion limit is

3,000,000 daltons. Both peaks eluted within the linear

































Figure 8.


Spectra/Gel AcA 22 gel filtration chromatography
of solubilized a-terpineol dehydratase. (Elution
buffer consisted of 10 mM HEPES, pH 7.0, 1.0%
(w/v) Triton X-100, 0.2 mM DTE, and 0.02% sodium
azide.)
























0.80 0.60

0.70 -
0-6-0.50
0.60 --

0.50 0.40
* I-

rE oooE
E 0.30 0.30

S: 0.20 E
S0.20 -
0
0.10 0.10
.0.100>
0.00 0.00
0 100 200 300 400 500 600 700 800 900
Elution Volume (mL)







64
fractionation range of the gel, thereby confirming that the

enzyme was soluble in 1.0% (w/v) Triton X-100. One

explanation for the occurrence of two activity peaks might

be that the enzyme existed as both a native monomer and

dimer in the presence of 1.0% (w/v) Triton X-100.

Alternatively, the two activity peaks may have been two

different enzymes (isozymes).

Partial Purification of Enzyme

Samples from various stages of enzyme fractionation

were analyzed for enzyme activity and protein concentration.

It was necessary for each sample to contain the same amount

of Triton X-100 because concentrations above 0.1% (w/v) had

an inhibitory effect on the enzyme (Figure 9). By adjusting

the detergent concentration to 0.5% it was possible to

compare enzyme activities of different samples.

Protein determinations were of high precision; however,

there was a high uncertainty in their accuracy. This was

because results obtained using the Coomassie blue protein

assay depend on the composition of the protein (Bradford,

1976). Much of the uncertainty could be eliminated by using

appropriate blanks to account for the sample matrix. For

example, Coomassie blue reacts with Triton X-100 as well as

protein. In order to use this method to monitor gel

filtration fractions, blanks consisting of the gel

filtration buffer were required. The background in the

presence of 1.0% detergent was relatively high; however,

this method was still preferable to monitoring the

































Figure 9. Triton X-100 concentration versus catalytic
activity of a-terpineol dehydratase. (Buffer
consisted of 10 mM HEPES, pH 7.0. Error bars
represent standard deviations, n = 4.)








110


100





- 90
o g






(I)
80





70 .-





60





50 -
0.0 1.0 2.0 3.0 4.0 5.0

Triton X-100 (% w/v)







67
absorbance (280 nm) of the gel filtration eluent. Triton

X-100 has a very high absorptivity at this wavelength.

Uncertainty associated with using bovine serum albumin as

the standard could not be eliminated, since this would

require using purified enzyme as the standard.

Results of the fractionation of the enzyme using

Spectra/Gel AcA 44 chromatography as the final purification

step are given in Table 3. The highest degree of

purification was obtained after solubilization of the enzyme

(5.0 fold). The final degree of purification depended on

the amount of particulate-associated enzyme recovered from

the 15,000 x g supernatant. Nearly 100% of the enzyme

activity was recovered after gel filtration; however, the

degree of purification obtained was only about two-fold over

the solubilized enzyme. Despite the low purification fold,

gel filtration removed much of the pigmented material and

appeared to lower the viscosity of the enzyme solution.

Results of the fractionation of the enzyme using

Spectra/Gel AcA 22 chromatography as the final purification

step are given in Table 4. Again, the highest degree of

purification was obtained after solubilization of the enzyme

(6.5 fold). The purification factor after gel filtration on

AcA 44 could not be accurately determined because the pooled

fractions contained an excessively high amount of Triton

X-100. This caused a high estimation of the protein

concentration. Inflated protein concentration values in

turn caused the calculated specific activities to appear


















.4
01
9-i

0
O



P4

0








N
a)



er






44






0

r,







0
0)
-)





a




0(


0



-sI


>o










.4 4. 00
S0





4Q)






0 >i.
44-17 C


-. 4 .
()341 0 0










410

.r4











C fi
00











O 0



0
.0



0.r


















S.'
0z


0
0
+1






0
0
H







0
0
+1
0
o
*






0\





0
+1
H

















0)
+1-
0,



H
o
+1







0\
H


.4'
H
0 '-0
0
0
+1 0
Nj +1
.4' Lo
(1) Lo

0 0


tU a)


U 0
I X
a) X M



M N
SH N


S+I
0
%D +1
N- N
* LA
cM in


co m oC







mr o
.n H 0
( 1 H 'O
~ CN (


a)



Ln
*





.U


0
-r 4



<-s













M .
00

w44'



4 -














,QM
0



."








'1








IV
40








0
0

H0

a)0



M04














>1


0
O

0

4.


N
4)


4-
0)



0

S-
0
4)












-L
I








el
0


0




Oi
o
0)


0
0)

5-I
-)

0)

-*
.4l
0,
cd
-H
&r


0>

40



H 7 "


rl -P 0
V 04o 4







4J

u>
0









**-1 *
o 04
--
VI

4-)










U


*cJ.
0







.>
O 0
S









0.
uO


0 1o
0 0
H H
*-4 .-


C*
0










*44








'r-4
0



140





















00
eI!e




































0
H0



















0

0
S'
O














Oa
t o
89

















k^0
<0OQ
O^2r
Q)'7

2^~


0









4)
04


0


























$.4
0






























41




O O

4J 4-)
-t -,fu
So
c



0)



0








































ON
4L
0)




0)




0(















0 0

0a 0


c c











0) 0

0 0









lower than the actual values. The main disadvantage of AcA

22 gel filtration chromatography was that the total volume

of fractions containing enzyme activity was about double

that from AcA 44 gel filtration chromatography. This meant

additional time was required to concentrate the fractions

and reduce the detergent concentration.

Characterization of a-Terpineol Dehydratase

Enzyme-catalyzed dehydration of a-terpineol to limonene

could not demonstrated. In fact, no change in limonene

concentration could be detected in the assay medium even

after 20 hr incubation of the enzyme with 50 mM racemic

a-terpineol or 5 mM (4R)-(+)-a-terpineol. The kinetic

properties of the enzyme were therefore determined only for

the hydration of limonene to a-terpineol. Furthermore,

(4R)-(+)-limonene was used as the substrate for all enzyme

assays, except for stereospecificity/stereoselectivity

studies.

Enzyme Nomenclature

It was decided that the enzyme would be named

a-terpineol dehydratase even though the dehydration of

a-terpineol to limonene could not be demonstrated. This

name was chosen because it was most descriptive of the

reaction. It is obvious that dehydration of a-terpineol

leads only to limonene. The name limonene hydratase was not

chosen because it is ambiguous in regards to which double

bond of limonene is hydrated. The name a-terpineol synthase

was not chosen because it describes only the product.








Molecular Weight

Initial attempts to determine the native molecular

weight of a-terpineol dehydratase involved nondenaturing

polyacrylamide gel electrophoresis (ND-PAGE). The method of

Doerner and White (1990) was used except that 0.1 or 1.0%

(w/v) Triton X-100 was included in all gels and buffers. In

0.1% detergent the sample proteins were unable to migrate

through the separating gels. In order to increase the

mobility of the proteins, 1.0% detergent was used. Several

components of the enzyme solution were partially resolved

with ND-PAGE in 1.0% detergent. Gels were composed of 5% or

6% acrylamide (with stacking gels composed of 4%

acrylamide). Gels were prepared from an acrylamide stock

solution containing 30%T and 2.7%C. It was difficult to

distinguish most of the protein bands because the gel

backgrounds were very dark due to the high concentration of

detergent. No enzyme activity was detected on either gel;

therefore it was not possible to identify the enzyme band.

It was not surprising that a-terpineol dehydratase could not

be separated by ND-PAGE, since the method is generally not

applicable to hydrophobic membrane proteins (Tzagoloff and

Penefsky, 1971).

The molecular weight of native a-terpineol dehydratase

in the presence of 1.0% (w/v) Triton X-100 was determined by

gel filtration. A profile for Sepharose CL-6B gel

filtration chromatography of a-terpineol dehydratase with

1.0% (w/v) Triton X-100 in the elution buffer is shown in







72

Figure 10. The linear fractionation range of Sepharose

CL-6B is 10,000-4,000,000 daltons. Enzyme activity eluted

as two peaks, as was also observed on Spectra/Gel AcA 22.

Molecular weights of the two forms of a-terpineol

dehydratase were estimated by comparing their VJ/V (where Ve

is elution volume of sample and Vo is void volume of column)

values with a calibration curve of molecular weight versus

Ve/V for protein standards (Figure 11). Results (Table 5)

suggest that a-terpineol dehydratase existed as two forms in

1.0% (w/v) Triton X-100; a monomer with an apparent

molecular weight of 94,500 daltons, and a dimer with an

apparent molecular weight of 206,500 daltons. Solubilized

membrane enzymes often exist as detergent-enzyme complexes

(Nalecz et al., 1986). It is possible that molecular

weights determined by this technique are those of detergent-

enzyme complexes and are not necessarily the actual

molecular weights of the enzyme monomer and dimer.

Several enzyme fractions were analyzed by SDS-PAGE

(Figure 12) to determine their protein subunit composition.

Molecular weights of protein subunits were determined by

comparing their migration distances with a calibration curve

of molecular weight versus migration distance for protein

standards (Figure 13). At least three bands were enriched

during the purification of the enzyme, corresponding to

polypeptides with molecular weights of 92,000, 38,000, and

28,000 daltons. The 92,000 dalton polypeptide had the

highest molecular weight of any other polypeptide in the

































Figure 10. Sepharose CL-6B gel filtration chromatography of
a-terpineol dehydratase. (Elution buffer
consisted of 10 mM HEPES, pH 7.0, 1.0% (w/v)
Triton X-100, 0.2 mM DTE, and 0.02% (w/v) sodium
azide.)































9



.,
*

0?
0i




I I I I~~J ~ I I I I


i I J I I1 1 1
5.0 10.0 15.0 20.0

Elution Volume (mL)


I 2
25.0


1.00

0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20

0.10

0.00 --
0.0


0.55

0.50

0.45

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0.00
).0


30


' '


































Figure 11. Calibration curve for Sepharose CL-6B column.
(Values represent means, n = 2.)








106



5x105





>


0) 10

5 105


o 5x104



Log Y -2.01(X) + 8.27
XSE= 0.08, YSE = 0.04
R Squared = 0.995



104 I I I I I I I I I
1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90

V/Vo, x








TABLE 5

Results of the molecular weight determination of native
a-terpineol dehydratase in 1.0% (w/v) Triton X-100 by
Sepharose CL-6B gel filtration chromatography


Peak Number Ve/Vo, Molecular Weight
(daltons)

1 1.4690.013b 206,5001800b

2 1.6380.003 94,500200


'V. = elution volume of sample, Vo = void volume of column
average standard deviation (n = 2)

























Figure 12. SDS-PAGE gradient (8-18%) gel of enzyme
fractions. [Lane 1, 10 AL of 1:100 diluted low
molecular weight standards. Lane 2, 10 gL of
recovered particulate-associated enzyme (0.25
mg-mL'' protein). Lane 3, 10 gL of solubilized
enzyme (0.25 mg-mL'" protein). Lane 4, 10 ML of
concentrated AcA 44 gel filtration fractions
(0.25 mg-mL-1 protein). Lane 5, 10 AL of 1:100
diluted high molecular weight standards.
Molecular weight standards were as follows:

a. Myosin (MW = 200,000)
b. E. coli B-galactosidase (MW = 116,250)
c. Rabbit muscle phosphorylase b (MW = 97,400)
d. Bovine serum albumin (MW = 66,200)
e. Hen egg white ovalbumin (MW = 45,000)
f. Bovine carbonic anhydrase (MW = 31,000)
g. Soybean trypsin inhibitor (MW = 21,500)
h. Hen egg-white lysozyme (MW = 14,400).]






79

















123 4 5





S ~-~
A b
c

S-d

e e

f-
h -
*"i'z ^"' '- W*



































Figure 13. Calibration curve for SDS-PAGE gradient (8-18%)
gel.









3X 105


2X 10-


5-



105



7X 104-



5X 104-


3X 10-


3 X 10


2X 104







104


0




















o





Log Y = -0.0184(X) + 5.793

XSE= 0.0005, YSE= 0.028
R Squared = 0.996


r I I I


20 30 40


50 60


70 80


Distance, X(mm)


Si
c*


(U
0)



0
0
j)
O







82

fractions and was essentially the same size as the native

a-terpineol dehydratase monomer. This implies that at least

some portion of the native monomer did not fractionate into

smaller subunits in SDS denaturing conditions. In order to

provide conclusive evidence this band represents a-terpineol

dehydratase, SDS-PAGE analysis of the purified enzyme would

be required.

Isoelectric Point

The isoelectric point (pI) of a protein is the pH at

which it has no net charge. One of the easiest and fastest

methods for determining the pi of a protein is by

isoelectric focusing (IEF), which involves setting up a pH

gradient (using ampholytes) and allowing the protein to

migrate in an electric field to a point in the system where

the pH equals its pI. The pi of a-terpineol dehydratase was

determined by agarose IEF in the presence of 1.0% (w/v)

Triton X-100 and 2.5% (w/v) ampholytes (Figure 14). Agarose

was used as the stabilizing medium instead of polyacrylamide

because of its greater porosity. Greater porosity was

required so that the high molecular weight proteins could

migrate unhindered. The IEF conditions used were suitable

for analyzing any protein with a pi between 3 and 10. The

pi of a-terpineol dehydratase was determined by comparing

its migration distance with a calibration curve of pi versus

migration distance for protein standards (Figure 15).

Enzyme activity was found throughout most of the gel, with

the majority being located in a zone where there were




























Figure 14.


Agarose IEF gel of partially purified
a-terpineol dehydratase in 1.0% (w/v) Triton
X-100 and 2.5% (w/v) Servalyte 3-10 ampholytes.
[Middle lane, 20 pL (50 Ag protein) enzyme
solution dialyzed in 1.0% glycine, pH 7.0,
containing 1.0% (w/v) Triton X-100. Outside
lanes, 10 PL of 1:4 diluted IEF standards.
IEF standards were as follows:

a. Phycocyanin (pI = 4.65)
b. B-Lactoglobulin B (pI = 5.10)
c. Bovine carbonic anhydrase (pI = 6.00)
d. Human carbonic anhydrase (pI = 6.50)
e. Human hemoglobin A (pI = 7.10)
f. Lentil lectin, 3 bands (pI = 7.80, 8.10,
8.20)]






84













4% f
i -a. b. c d. .
C- 0.2 g j "




% I I II 11l
0.0




1 2 3 4 5 6 7 8 9 10 11 12

Fraction Number

































Figure 15. Calibration curve for agarose IEF gel. (Values
represent means, n = 2.)















Y = 20.5(X) 94.4

XSE= 1.2, YE= 4.2

R Squared = 0.980


5I I I
5.0 5.5 6.0 6.5


Si I I I


7.0 7.5 8.0 8.5
7.0 7.5 8.0 8.5


pl, X


80.0


70.0




60.0


50.0




40.0




30.0


G)



(0


20.0


10.0


0.0 1
4.5







87

several protein bands. This zone actually represented the

0.75 mm gel fraction which had the highest activity. The

diffusion of enzyme activity throughout most of the gel was

probably due to the association of the enzyme with several

ampholyte components, those species having combined pi

values different from that of the native enzyme (Scopes,

1987). For the enzyme solution which was dialyzed against

1.0% glycine, the zone of highest activity was composed of

proteins with pi values between 6.5 and 6.8. The band of

highest intensity in this zone had a pi of 6.6. For the

enzyme solution which was dialyzed against distilled water,

the zone of highest activity was composed of proteins with

pi values between 6.0 and 6.8. The band of highest

intensity in this zone also had a pi of 6.6. It was

concluded, by comparing the overlap of the zones of highest

activity from these two experiments, that the pi of

a-terpineol dehydratase was between 6.5 and 6.8. There was

not enough evidence to determine which band actually

represented a-terpineol dehydratase. Such a determination

would require IEF of the purified enzyme.

Activity in Organic Solvents

Determination of the kinetic properties of an enzyme

becomes complicated when the substrate is insoluble in the

medium containing the enzyme. Indeed, classical enzyme

kinetic theory assumes that both enzyme and substrate(s) are

soluble in the same medium. Therefore, the potential of

using water-miscible organic solvents to dissolve limonene

in the enzyme reaction medium was examined.







88
Solvents used in the experiments were carefully

selected using the criterion of Khmelnitsky et al. (1988).

These researchers stated that hydrophobic interactions play

the dominant role in maintaining the catalytically active

conformation of enzymes. Thus, the most favored

conformation would be produced by solvents that can replace

water in the hydration shell of an enzyme molecule without

significantly distorting hydrophobic interactions (i.e.

solvents would maintain solvophobic interactions).

Ray (1971) classified organic solvents according to

their capacity for solvophobic interactions. Solvents were

classified into three groups, for example: (i) water,

glycerol, ethylene glycol, aminoethanol, and formamide, (ii)

methyl formamide and dimethyl formamide; and (iii) methanol,

ethanol, and toluene. According to Ray (1971), solvophobic

interactions are best realized in solvents of group (i),

much less so in solvents of group (ii), and are practically

absent in solvents of group (iii). This is because the

solvents of group (i) contain at least two groups capable of

forming hydrogen bonds. Among the solvents in class (i),

water and glycerol have the highest capacity for solvophobic

interactions.

Effect of several group (i) and (ii) solvents on

a-terpineol dehydratase activity is shown in Figure 16. The

effect of glycerol was examined for comparison purposes

only, since limonene is scarcely more soluble in glycerol

than in water. a-Terpineol dehydratase activity decreased

































Figure 16. Effect of glycerol, formamide, and N,N-dimethyl-
formamide concentration on the activity of
a-terpineol dehydratase. (Error bars represent
standard deviations, n = 4.)