P

Irr'l-vavgr-

.... ~ - ' .. 2‘ Mafia;
«‘3' :‘f’xu' 7': n ' :{N 1V?“ :fi’.;.' {'4‘ .
7575”?” ' 39‘3'1‘:’3~-.:i'~{ ‘
’ v‘l. “ I'm: - '_ . '
;.;?t3:?’;i&: ’ - ‘ ‘ ~J a,“ ‘ ‘ ‘19:.-

3.: _ ~
' 1- R7L"~ -
«“it‘ f~§§ ‘ LIE-3‘3“?

‘ u ' 's‘g‘"
‘i - . '. ~
6‘ gay: 15‘ ~ A)
”I" 5‘45-
'11 Ht"?

‘ .

..
J.
J 3 {':§‘ ,z‘qfi?
u~~ "i’:z!‘231'-*L"1‘ w . .73, w.
L ‘ 3.9:“ ‘ - ':‘3\.$‘3-.
~ a H. . t ‘0’" 4‘
1‘7“ «13% ~15.” ”7“ :3i1";€2\‘€2..:
if if‘yfi“ qu :\fl: is : .

..(

.
«,3.
$.: “ .:
"“7“: n0
‘3; Jul,

g1:

47L’.‘ f
EtrAi . '
if?» «
an"

@772;

v

‘ 3:
““3 5:21.

' hzu‘
V"

‘5? ‘7: ‘v 1
fix.
o. 2% ’1‘;

“-1.:

,- if";
5'

iii? Fifi K;
.

5‘3}? '

”t

~50
32’! '.
I 5’3

d' N
".‘ I I4
. .91, “

.131?
I‘ '1
53.23::
Wé‘fffl:%
, . .

L II
."

I

. J

("‘37 If”:

I-‘MN’W' :2":
L"

514$ ,-,,_:,.:~
IN)?" N“

- . -_‘_._-.
N¢Z~Lxr

..'(t-’h‘

w
r “‘“I‘U” any”... . ’ .
WW -.' 'x I,
é :3

Mg..." -‘
”I. 44"]

‘1
¢.

‘ -_ ;:‘.’." k
. .A —
,p’:1’r
1.54:9
.\ - -\'.., K
.I ' A
‘x‘ A
'1'
.
..

iii. ’11
:_I:-'£< ‘4‘; I r
u‘r,-[7 of,“
‘1”; 1:4 .
14’4""
" vi" '
f/o.‘

f ‘ ;“'r:;II" "‘1” p
Halal“, {211/ 5.4.1:”,
"'le “'3' :‘M _(IUI' 'f
5? ' . - 1

lb!
1/"
\‘r..- -
" ‘Wuw'll.
Mr

I
I ' f ,(fi‘ FAT: ‘ ; ; 7‘ V 10’")?
.l'.' and}. ’35?” “ K m
“1.. v. ‘. 1’ _/‘.h'
} .s'.

f 3'
1'4"" .1)“ o‘lxle (ill-f"- .‘
A ‘j «I ‘, a. i" - If I
(ii-3.7!. fi’ 2"” “5.),1fi‘?’ 4376” h
‘- Vi}: I’lhé'
;; 4‘ ~- “3" N ,,
. “finfis rtr. ‘79,“: ”t."

”-
‘ I
{Kuhn/’3 . ..

_____1ae‘

A 53:... .

 

 

‘ KW ’ I

'= x- 5‘ ‘ [O i
/*\

96910 menu. AN sum: {mummy LIURAH

l l. ”Inllllnll lll.‘ “ll” Ill ll

_3 1293 00563 5341
LIBRARY
Michigan State
University

 

 

 

 

 

 

 

 

 

This is to certify that the
thesis entitled
The Concentration and Purification of a-Amylase Using A
Methylcellulose-Salt, Two-Phase Partitioning Process

presented by

Steven T. Summerfelt

has been accepted towards fulfillment
of the requirements for

M.S. CHE

degree in

 

 

IflJQW

Major professor

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

‘lV1ESI_J RETURNING MATERIALS:
Place in book drop to
LJBRAfiJES remove this checkout from
-r_‘—. your record. FINES will

 

 

be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

THE CONCENTRATION AND PURIFICATION OF a-AMYLASE USING A

METHYLCELLULOSE-SALT, TWO-PHASE PARTITIONING PROCESS

By

Steven T. Summerfelt

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Chemical Engineering

1988

ABSTRACT

THE CONCENTRATION AND PURIFICATION OF a-AMYLASE USING

A METHYLCELLULOSE-SALT, TWO PHASE PARTITIONING PROCESS

By

Steven T. Summerfelt

The downstream purification of a-amylase from E. 92;; by aqueous
two-phase partitioning with methylcellulose (MC) and ammonium sulfate
was studied. The partitioning of a-amylase into MC was optimized in a
reversible purification scheme.

A portion of a-amylase has been genetically fused to alkaline
phosphatase. Purification of this fusion protein via aqueous two-phase
partitioning was investigated and some enhancement over regular
precipitation was found. Comparing the partitioning, precipitation, and
hydrophobicity of amylase to similar experiments using fi-amylase and
pullanase, indicated that the interaction of a-amylase with MC is not

due to hydrophobicity alone. This interaction seems to be unique to

starch binding enzymes.

ACKNOWLEDGEMENTS

Support for this research was provided by the Research Excellence
Economic Development Fund. I am grateful to the Michigan Biotechnology
Institute for the use of their facilities and equipment. I would like
to thank Professor Kris A. Berglund for his support, guidance and
advice, and to his graduate students Ponnampalam Elankovan and Everson
Miranda for their assistance. Professor Patrick J. Oriel and his
graduate students provided the genetically engineered enzymes and

offered valuable advice and assistance.

iii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
LIST OF FIGURES. . . . . . . . . . . . . . . . . . . . . . . . . . . vii
CHAPTER 1: INTRODUCTION
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . 1
1 2 Literature Cited . . . . . . . . . . . . . . . . . . . . . 3
CHAPTER 2: LITERATURE SURVEY
2.1 Background . 5
2.1.1 Starch Hydrolysis Industry . 5
2.1.2 a- Amylase. . . . . . . 6
2.1.3 Thermostability. . 7
2.1.4 Thermostabile a— Amylase from
539111;; §§earothggmo philus. . . . . . . . . . . . . 8
2.2 Factors Affecting Protein Solubility
and Fractional Precipitation . . . . . . . . . . . . . . . 9
2 2.1 Salting Out. . . . . . . . . . . . . . . . . . . . . 10
2 2 2 Organic Solvents . . . . . . . . . . . . . . . . . 11
2 2 3 Isoelectric Precipitation. . . . . . . . . . . . . . 12
2 2 4 Precipitation with Metal Ions. . . . . . . . . . . . l3
2 2 5 Thermal Precipitation. . . . . . . . . . . . . . . 13
2.2.6 Polyelectrolyte Precipitation. . . . . . . . . . . . 14
2.2.7 Non- ionic Polymer Precipitation . . . . . . . . . . 15
2.3 Biospecific and Affinity Separations . . . . . . . . . 16
2.3.1 Biospecific and Affinity Chromatography. . . . . . . 16
2. 3. 2 Affinity Precipitation . . . . . . . . . . 17
2. 3. 3 Biospecific and Affinity Partitioning. . . . . . . . 17
2.4 Membrane Separation. . . . . . . . . . . . . . . . . . . . 20
2.4.1 Ultrafiltration. . . . . . . . . . . . . . . . . . 20
2. 4. 2 Affinity Ultrafiltration . . . . . . . . . . . . . . 20
2.4 Summary. . . . . . . . . . . . . . . . . . . . . . . 21
2 5 Literature Cited . . . . . . . . . . . . . . . . . . . . . 22
CHAPTER 3: CONCENTRATION AND PURIFICATION OF a-AMYLASE USING A
METHYLCELLULOSE-SALT, TWO PHASE PARTITIONING PROCESS
3.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . 29
3.3 Experimental . . . . . . . . . . . . . . . . . . . . . . 32
3.4 Results and Discussion . . . . . . . . . 36
3.4.1 Optimization of the Partitioning Process . . . . . . 36
3. 4. 2 Purification . . . . . . . . . . . . 42
3. 4. 3 Nature of the Partitioning Process . . . . . . . . . 50
3. 4. 4 Methylcellulose Recovery . . . . . . . . . . . . . . 52
3.5 Conclusions. . . . . . . . . . . . . . . . . . . . . . . 55
3.6 Literature Cited . . . . . . . . . . . . . . . . . . . . . 60

iv

CHAPTER 4:

EFFECT OF A GENETICALLY ATTACHED SECRETION SEQUENCE

ON PARTITIONING ALKALINE PHOSPHATASE FROM E. COLI

J—‘J-‘J-‘L‘
#UDNH

CHAPTER 5:

5
5.
5

“NH

APPENDIX .

Abstract .

Introduction .

Experimental . . .

Results and Discussion .

4.4.1 Comparison of partioning of alkaline
phosphatase, a- amylase, and the
phosphatase/amylase fusion protein . .

4.4.2 Comparison of hydrophobicity of alkaline
phosphatase, a-amylase, and the alkaline
phosphatase fusion protein .

Summary. . .

Literature Cited .

SUMMARY
Conclusions.

Proposals For Future Research.
Literature Cited .

63
63
65
69

69

77
80
81

83
84
84

85

LIST OF TABLES

Table 3.1. The a-amylase activity and total protein
of solutions prepared for ammonium sulfate
precipitation and MC/salt partitioning. . . . . . . . . 43

Table 3.2. The downstream refining of a-amylase. . . . . . . . . . 59
Table 4.1. The initial enzyme activity and total protein

of solutions prepared for ammonium sulfate
precipitation and MC/salt partitioning. . . . . . . . . 70

vi

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

.5a.

.5b.

.6a.

.6b.

.9a.

.9b.

LIST OF FIGURES

The action of enzymes involved in starch-
saccharification as they act on the two
components of starch .

The structure of methoxy substituted
cellulose (MC)

The percentage of activity remaining in
the salt phase as a function of the enzyme
loading at 30% of ammonium sulfate saturation.

The percentage of activity remaining in the
salt phase as a funtion of the ammonium
sulfate or sodium chloride saturations .

The percentage of activity remaining in the
salt phase as a function of pH at 30% of
ammonium sulfate saturation.

The percentage of activity or protein
remaining in the salt phase after partitioning
of extracellular amylase .

The percentage of activity or protein
remaining in the salt phase after partitioning
of periplasmic extract amylase .

The percentage of activity or protein
remaining after precipitation of extracellular
amylase.

The percentage of activity or protein
remaining after precipitation of periplasmic
extract amylase.

The percentage of activity remaining in the salt
phase after partitioning with 30% of salt
saturation, as a function of K15M concentration.

The percentage of activity remaining after
partitioning has attained equilibrium as a

function of ammonium sulfate saturation.

The adsorbtion/desorbtion of amylase and
adsorbtion of protein using octyl sepharose.

The adsorbtion/desorbtion of amylase and
adsorbtion of protein using phenyl sepharose

vii

32

38

39

41

44

45

46

47

48

51

53

54

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

. The percentage of activity, total protein

K15M remaining in the salt phase after
partitioning of extracellular amylase.

. The downstream purification of a-amylase

The percentage of phosphatase, amylase or
fusion protein activity remaining after
partitioning upon addition of ammonium sulfate

The percentage of phosphatase, amylase or
fusion protein activity remaining in the salt
phase with respect to percent K15M present .

. The percentage of phosphatase activity remaining

in the salt phase, as a function of salt added,
for either partitioning or precipitation .

. The percentage of amylase activity remaining

in the salt phase, as a function of salt added,
for either partitioning or precipitation .

. The percentage of phosphatase activity, due to

the fusion protein, remaining in the salt phase
with respect to salt added .

The percentage of phosphatase, amylase or fusion
protein activity remaining after precipitation
upon addition of ammonium sulfate.

The adsorbtion of the fusion protein, amylase or

phosphatase to octyl sepharose as a function of
the salt added .

viii

56

58

71

72

74

75

76

78

79

CHAPTER 1
INTRODUCTION

1.1 Introduction

a-Amylase, an enzyme that hydrolyzes starch, is utilized in many
industrial applications. Providing amylase at practical prices for
industrial applications requires that production and refining costs
remain low. The refining cost alone is often half of the total
production costs. To remain economical, the recovery process must
achieve the desired purity of active enzyme at a high final yield
without using costly equipment, or additives, or labor intensive
techniques, and it must not result in waste disposal problems.

Large scale enzyme production processes usually require recovery
techniques significantly different from the standard fractionation and
chromatographic techniques that predominate the research literature1
Aqueous two-phase systems, particularly PEG/salt systems, have been
employed in multistage processes for the purification of proteinsz. The
specific partitioning of an enzyme in aqueous two-phase systems results
mostly from its interaction with the surroundings through hydrophobic,

2,3 .
. These interactions can be controlled to

hydrogen, and ion bonding
enhance partitioning purification by manipulation of the choice of
polymer, its molecular weight, the concentration of the phase system
components, the type and concentration of salts added, the solution pH,
and the temperature during partitioningz.

The specificity of partitioning in aqueous two phase systems can be

increased by tailoring the polymer phase with ligands that are

2
hydrophobic or electrically charged, making the ligand "biospecific"4
The structure and composition of the protein can indicate the choice of

ligand by its effect on proteins of similar compositions.

 

A distinctive characteristic of a-amylase from E; coli, designated
ATCC 29609, is its strong hydrophobic characters. Therefore, a
hydrophobic polymer was chosen to form an aqueous two phase partitioning
system with salt. Instead of synthesizing a hydrophobic polymer as the
literature suggests7, methylcellulose, an existing non-ionic
polysacharide with an unbranched hyrophilic backbone interspersed along
its length with small hydrophobic groupsg, was utilizized.
Methylcellulose was used with dextran in 1960 to perform aqueous two-
phase extraction of several proteinsg. It has many advantages including
well characterized propertiess, low cost at around $2.70 a poundlo, and
FDA approval8 as a polymeric surfactant for use as a food additive. It
also comes in several different molecular weights and as methoxy or
hydroxylpropoxyl and methoxyl substituted cellulose derivativesg. In
aqueous solutions, methylcellulose exists as highly hydrated colloids
which can be gelled or salted out of solution upon addition of certain
limits of solutes or electrolytesa. In addition methylcellulose has an
inverse solubility, i.e., it gels as temperature is increased.

The gene of the a-amylase used in the present study was cloned into

E. 9911 from 5‘ stggrgthgrmgphilusll. a-Amylase from E. coli has the

 

valuable and distinctive characteristics of thermostability and
extracellular excretion, features which enhance its purification11

12,13, and

Extracellular excretion of enzymes by E. coli is rare
therefore the structural regions which confer the a-amylase its

0 14 O O O
excretion properties were investigated to determine if fu51ons

3
containing these regions could produce extracellular release of other
proteins. The enzyme alkaline phosphatase is naturally secreted into
the periplasmic space, but not excreted extracellularlyla. Alkaline
phosphatase has been genetically fused with portions of a-amylasela, and
the fusion product was conferred extracellular release14

The objectives of this research are to:

l. Optimize the partitioning of thermophilic a-amylase in the
methylcellulose/salt aqueous two-phase system with respect to
the type, concentration and molecular weight of cellulose
derivative, the type and concentration of salts added, the
solution pH, and the temperature during extraction.

2. Determine if the partitioning of a-amylase can be reversed upon
decreasing the salt concentration, enabling the methylcellulose
to be recovered from the purified enzyme.

3. Demonstrate a recovery technique for a-amylase from crude broth
using a methylcellulose/salt partitioning process.

4. Determine the nature of the interaction between the methyl-
cellulose and a-amylase.

5. Determine if aqueous two phase partitioning with methylcellulose/
salt can be applied to a phosphatase/amylase fusion protein.

1.2 Literature Cited

1. Crueger. Wulf, and Anneliese Crueger. 1984. Biotechnology: A textbook

9f inggstrial microbiology. Science Tech, Inc., Madison, Wisconsin. 308
PP-

2. Hustedt, H., K. H. Kroner, U. Menge and M.-R. Kula. 1985. Protein
recovery using two-phase systems. Trends in Biotechnology. 3(6):139-144.

3. Mattiasson, B. 1983. Applications of aqueous two-phase systems in
biotechnology. Trends in Biotechnology. 1(1):l6-20.

4. Johansson. G. 1987. Dye-ligand aqueous two-phase systems. Pages 101-
124. In: Y. D. Clonis, T. Atkinson, C. J. Bruton and C.R. Lowe, eds.

4

Reagtive dyes ig proteig and enzymg gechnology. Stockton Press, New
York, NY.

5. Bell, D. J., M. Hoare and P. Dunnill. 1983. The formation of protien
precipitates and their centrifugal recovery. Advances in Biochemical
Engineering. 26:1-72.

6. Oriel, P. Personal communication, 1987.
7. Mattiasson, 3., and R. Kaul. 1986. Use of aqueous two-phase systems

for recovery and purification in biotechnology. American Chemical
Society. 314:78-92.

8. Greminger, G. K. Jr., and K. L. Krumel. 1980. Alkyl and
hydroxyalkylalkylcellulose. Pages 3-1 to 3-25 In: R. L. Davidson, ed.

Handbggk of Wgtg: sgluble gyms and resins. McGraw Hill, New York.
9. Albertsson, P.-A. 1958. Nature. 182:702.

10. Chem1;gl_flg1§g§1ng_figpgytgy. August 1. 1988. Schnell Publishing
Company, New York, New York.

11. Schwartz, J. H., and F. Lipmann. 1961. Phosphate incorporation into
alkaline phosphatase of E. 921;. Biochemistry. 47:1996-2005.

12. Holland, H. B., N. Mackman, and J. M. Nicaud. 1986. Biotechnology.
4:427-431.

13. Oliver, D. 1985. Protein secretion in E. coli. Ann. Rev. Microbiol.
39:615-648.

14. Alexander, P., and P. Oriel. 1988. Excretion of amylase/phosphatase
fusion proteins by E. 9211. Prepublication draft.

CHAPTER 2
LITERATURE SURVEY

2.1 Background

Enzymes used for large scale commercial applications such as food
processing, household cleaning aids, and effluent treatment have to be
produced and refined at low costs. Even so, industrial use of enzymes
has been of increasing importance in recent years due to developments in
microbial genetics, fermentation, and enzyme recovery and purification
techniques. These developments were instrumental in making the
production of these enzymes economically feasible.

In particular, the ability to recover and purify an enzyme
economically often determines the overall feasibility of an industrial
process. As well as a achieving a high final yield, the purification
process must not reduce the biological activityof the enzyme. High
yield and biological activity are difficult to obtain because enzymes
are fragile molecules produced in very dilute solutions that can contain
soluble portions of residual substrates, metabolic pathway inter-
mediates, and cellular debris. Therefore, to recover the enzyme
economically at the desired concentration and purity, the separation
process must be efficient and reliable.

2.1.1 Starch-Hydrolysis Industry

Starch is a naturally produced and widely available glucose
polymer. Amylases are enzymes used industrially to hydrolyze starch for
such applications as brewing, baking, alcohol, milling, paper, textiles,
feed, detergent, and sweetenersl. Bacterial amylases are commercially

5

6
produced in large quantities and generally used for industrial
applications after only minimal refinement. In 1979, 300 tons of
amylases were produced2 and marketed for about 8 million dollars3
Studies on purification and crystallization of amylase from bacteria are
numerous4 due to its commercial importance.

The commercial production of sweeteners, such as glucose, fructose,
and maltose from starch is one of the main uses of amylases. Amylases
are used to hydrolyze starch into smaller saccharides including glucose.
Because glucose is not as sweet as its isomer fructose, it is converted
to fructose with the enzyme glucose isomerase. The enzymatic hydrolysis
of starch, called starch-saccharification, includes a—amylases, fi-
amylases, glucoamylases, glucose isomerases, pullanases, and
isoamylasesl. The different actions of each of these enzymes is shown
in Figure 2.1. The focus of the present study is on a-amylase.

2.1.2 a-Amylase

a-Amylase specifically hydrolyzes a-1,4-g1ucosidic bonds in
polyglucans such as starch, glycogen, and dextrins while leaving the a-
1,6 glucosidic bonds (i.e. branch points) unhydrolyzeds. The hydrolysis
products are reducing groups such as maltose, short limit dextrins
usually containing a-1,6-glucosidic branch points, and small amounts of
glucoseé.

a-Amylases differ from fl-amylases by attacking the interior of
polyglucan chains (i.e. endoases) as opposed to attacking from a chain
end (i.e. exoases)5. The products of both a- and 6- amylases are

maltose.

 

«1 lose

“1 lover: in

0—0 at 1-4l glucosidic linkage

0—0 at I-6l glucosidic linkage

o-oo-amyloso action

“allmmyluo action

G—qlucoomylaso action

O-adobnncho'nq enzyme
(puuulanuo. isoamyuul action

 

 

 

 

Figure 2.1. The action of enzymes involyed in starch-saccharification as
they act on the two components of starch .

Bacteria and fungi produce several different a-amylases, sometimes
even within the same organism. These different a-amylases are
classified according to their pH optimum, temperature range, stability,
and starch liquefying and/or saccharogenic effectl. Free sugars are
produced only by saccharogenic a-amylases. Starch liquefying a-amylase
breaks down the starch into maltose and short limit dextrins.

2.1.3 Thermostability

Many a-amylases from bacteria of the genus Bagillgs are
thermostable, greatly facilitating the enzymes' industrial utilization.
For example, operating at higher temperatures reduces contamination by
foreign organisms during production. In the starch industry a
thermostable a-amylase allows liquefaction of starch at temperatures
above the gelatinising temperature of starch granulesS

In addition, the number of purification steps required to isolate a
thermostabile enzyme downstream from a fermentor are reduced by its

thermostability. Just heating the broth for a short period of time to a

l;—

 

8

temperature slightly below the limit of the enzyme's stability
precipitates by denaturation many contaminants in solution while leaving
virtually all the thermostable enzyme.
2.1.4 Nature of the Thermostabile a-Amylase from E. stearothermophilus

As thermostability in a-amylase is a desired characteristic for
commercial applications, discovery of such an enzyme in Bacillus
Egggzgghgzmgphilgg was of considerable importance. Subsequently, the a-
amylase gene from Eggillgg figggxgghgxmgphilgg was cloned into E. 92;;
(designated as EC 147). The genetically modified E. 99;; releases
thermostable a-amylase extracellularly and will grow on starch as the
sole carbon source7. It was advantagous to clone the thermostable a-
amylase gene into E. 9911 because the molecular biology of Escherichia
£911 is well known.

The a-amylase from both the parent E. ggggzgghggmgphilgg and EC 147
were found to have optimum temperature and pH of 70°C and 5.1,
respectively. The amylase from both organisms were found to be stable

with little loss of activity at 90°C over a 1 h period if in the
7

2 .

The EC 147 did not grow efficiently on M9 minimal agar with

presence of 5 mM CaCl

glycerol7. However, a mutation of it was found that did grow well on
M9—glycerol, -starch, or -solub1e starch with significant release of a-
amylase7. This mutation, designated EC 148, did not differ from the EC
147 in release of a-amylase or growth in L brotha. The EC 148 organism
was used to produce the a-amylase in the present study.

Information on structure and biocomposition of thermostable a-
amylase from E. gggarothegmophilus is highly variable, with significant

differences reported for thermostability, molecular weights (15 to 90

kilodaltons), and other propertiesg-11. SDS-polyacrylamide
electrophoresis indicated a molecular weight of 57 kilodaltons7 for the
a-amylase from ATCC 29609 which was used in this work. In addition,
this enzyme appeared to be very hydrophobic.
2.2 Factors Affecting Protein Solubility, Fractional Precipitation,

and Partitioning

The term precipitation will be used to describe an operation in
which a reagent is added to a protein solution causing the formation of
insoluble particles of protein. The intent of precipitation is to
recover a protein in either an unchanged molecular form or one which is
readily returned to that form.

The present research is concerned with proteins whose solubility
properties are determined largely by their polypeptide structure. For
example, proteins with relatively small non-peptide groups such as
lipo-, nucleo-, and glycoproteins have solubility properties often
distinctive to these groups. Thus, glycoproteins are very soluble in
aqueous solution due to hydration of their carbohydrate moiety.
However, lipoproteins are relatively insoluble in aqueous solution due
the hydrophobic nature of their lipid components.

The choice of precipitating reagents can be suggested by the known
effect of the reagent on the solubility of proteins of similar
structure. The polypeptide chain of water soluble proteins are folded
in such a way that the majority of polar hydrophilic amino acid side-
chain groups will be on the exteriorlz. However, this division is not
so well defined that changes in the exterior environment brought about
by the precipitating reagent will not affect both hydrophobic and

hydrophilic groups. The overall effect of the precipitating reagent on

10

the protein results from the sum of the individual effects which will
often be opposed to one another. Thus, the resulting protein
conformation may either increase, decrease or not change the solubility
of the protein depending upon the moieties contained with the protein.

Because the solubility of a protein is largly determined by its
primary structure, proteins may be genetically engineered to facilitate
their purification. Recombinant DNA technology has made it possible to
produce foreign gene products in bacteria7. This technique also makes
it possible to direct the proteins to be secreted extracellularly using
secretion vectorslB-15 and to simplify the protein's purification by
another genetic modificationlé. One method for enhancing purification
is to genetically fuse an additional peptide to the protein which would
enable a simple, cheap, and efficient purification. The only
requirements for the fusion are retention of biological activity and
ready removal after purification16. Finally, if the same peptide is
fused to other recombinant proteins, the same purification method may be
applicable. 7
2.2.1 Salting out

Salting out involves precipitation of proteins from solution using
high concentrations of neutral saltslz. As the salt is added and the
solution's ionic strength increases the protein's solubility decreases
until it is driven out of solution. This change in solubility is
described by the interaction between two opposing effects. Increasing
the ionic strenth of solution increases the electrostatic effects which
increases the protein's solubility17. Opposing this is the hydrophobic
effect, which decreases the proteins solubility as the ionic strength

17
increases

11

Changing a protein's solubility by salting out is therefore a
function of choice of salt and concentration. The Hofmeister
series17 rates the relative effectiveness of anions on salting out,
where:

citrate > phosphate > sulfate >
acetate or chloride > nitrate > thiocyanate.

Salting out, however, may denature the protein and affect its function.
The amount of protein denaturation caused by the salt is inversely
related to its position in the Hofmeister seriesl7, e.g. nitrate would
be more damaging than phosphate.

Cations have less effect than anions on the precipitation.
Monovalent cations are preferred, with order of effectiveness being12

N114+ > K+ > Na+

The salt of choice must also have a pH range within the limits of
the enzymes stability. Finally, it must be inexpensive and either
recoverable or have managable disposal problems.

The preferred salts used for isolating a-amylase by salting out are
sodium chloridels.20 and ammonium sulfate4’6’18.
2.2.2 Organic solvent precipitation

Organic solvent precipitation involves the precipitation of protein
from solution upon the addition of a weakly polar solvent. Solubility
reduction by a polar solvent is a result of a shift in the solvent's
dielectric constant away from the protein’s isoelectric point, producing
an increase in the effectiveness of intra- and intermolecular
attraction17. The solubility of metal-protein complexes such as the

21,22

calcium-a-amylase is very dependent on the dielectric constant of

12
the medium17. Protein damage is reduced because of the lower solvent
concentrations.

Precipitation by organic solvents can be combined with variation in
pH, temperature, and ionic stength to provide very careful control of
the protein fractionation. Additional advantagous properties of organic
solvents include their relatively high volatility, which facilitates
their recovery, and their bactericidal effect, which keeps the media
sterile17.

Some disadvantages of solvent extraction exist. There is a
tendency for irreversibly denaturation of the protein, but this problem
can be minimized by operating at low temperatures. Other disadvantages
include flammability and government control of ethanol.

Solvent extraction of aoamylase has been frequently reported in
conjunction with other extraction steps. The solvents of choice in

4,6,20,23-27 20,23-25

previous methods include acetone ethanol

20,23-25 and NHhOH20’26’27.

2.2.3 Isoelectric precipitation

ether

A protein's solubility is substantially reduced at its isoelectric
pointlz, i.e., the pH where the protein has no net charge17. In low
ionic strength solutions, the isoelectric point can be reached by
addition of either acids or bases to the solution.

Isoelectric precipitation is enhanced for proteins with high
surface hydrophobicitiesl7. A well defined isoelectric point enhances
separation specificity. Additional advantages are low cost and
acceptance for use in food products. Phosphoric, hydrochloric, and

sulfuric acids are all available at low cost and are allowed food

additives.

13
The overriding disadvantage of isoelectric precipitation is the
potential for irreversible denaturation at pH extremes.

The preferred reagents for isoelectrically precipitating a-amylase

are ammoniaza, and acetic acidza’zs.

2.2.4 Precipitation by metal ions
Polyvalent metal ions are even more effective at precipitating
proteins than the salt ions mentioned above. Polyvalent ions have been

classified into three groups according to the site they selectively

bind17. Carboxylic acids and nitrogenous compounds such as amines and

heterocyclics are bound strongly by ions such as Mn2+, Fe2+, 002+, N12+,

Cu2+, Zn2+ and Cd

Mg2+ and Pb2+ bind carboxylic acids specifically but not nitrogenous

2+. A different set of ions, including Ca2+, Ba2+,

groups. The third group, including Ag+, Hg2+ and Pb2+ strongly binds
sulfhydryl groups.

Metal ions have the advantage of precipitating proteins at low
concentrations; they may be removed from solution by ion exchange.
a-Amylase has been precipitated4 using Zn2+.
2.2.5 Thermal precipitation

By increasing the temperature of a crude solution of proteins,
fractionation can occur by selectively denaturing and hence
precipitating proteinslz. The selective denaturation of protein is not
applicable to all protein systems, but it is of particular usefulness
for the isolation of thermostable enzymes (see section 2.1.3). Thermal
precipitation is inexpensive because it does not require the addition or
removal of precipitating reagents.

The use of thermal precipitation to isolate a—amylase has been

reported26-28.

14
2.2.6 Polyelectrolyte precipitation

The recovery of proteins through reversible precipitation with
polyelectrolytes, has been adapted for the purification and recovery of
several industrial enzymeszg. Polyelectrolytes such as polyacrylic
acidszg, carboxymethylcellulose30, anionic hydrocolloidsBl, and
heteropolyacids32 have been used for precipitation of impurities,
precipitation of the desired enzyme, and fractional precipitation of
more than one enzyme.

Upon addition of a polyelectrolyte to a crude enzyme solution at pH
3.0-5.8, a precipitate forms between some of the protein and the
polyelectrolyte33. The selectivity of the precipitation is such that at
pH < 6.0 mono-, oligo- and polysaccharides, amino acids, oligopeptides,
lipids, uncleotides, nucleic acids and inorganic salts do not form
precipitates with polyacrylic acids29. The desired enzyme may end up in
the precipitate or stay in solution depending upon the conditions of
reaction and its chemical reactivity. In either instance a
fractionation of the enzyme has begun. The polyelectrolye-enzyme
complex can be isolated by centrifugation or ultrafiltration, and the
enzyme unbound and solubilized at pH > 6.0 upon addition of a divalent
metallic cation such as Ca2+ or Mg2+. Thus the enzyme is isolated and
the polyelectrolyte recovered as an insoluble salt of the polyacrylic
acid is formed33.

Polyelectrolytes have the advantage of low cost, recoverability and
an overall effectiveness at low concentrationszg. Their limitations
for protein purification are due to denaturation within the pH range of

precipitation33

The isolation of a-amylase using polyelectrolytes has been reported

15

using polyacrylic acid29’33.

2.2.7 Non-ionic polymer partitioning
Aqueous two-phase partitioning results from the incompatability
between aqueous solutions of two polymers or between one polymer and an

34,35

appropriate salt if above certain reagent concentrations The

phases formed have characteristics ranging from precipitate to liquid
droplet, or to something in between. The most common two-phase system
employed are with polyethylene glycol (PEG) and dextran. However, other

36,37

polymers including methylcellulose , polyvinyl alcohol35, and

ficol35 have been used to form aqueous two-phase systems.
Aqueous two phase partitioning is a well described purification
method particularly suited to the separation of cell debris and

34,38,39

enrichment of a desired protein found in low concentrations.

These systems provide very mild conditions for partitioning due to their

40-42. A

high water content, approximately 75-95 percent for both phases
gentle partitioning is also due to the very low surface tension between
the two phases, about 0.1 dyne/cm43, allowing the creation of an
emulsion even upon only light mixing. However, water-organic solvent
two-phase extraction systems are harsh on biomolecules, partly due to
surface tension around 400 times greater than for aqueous systems42

The partitioning of biomolecules between two-phases is governed

mainly by their surface properties”’42

system39’42. The phase system is characterized by the partition

and the composition of the phase

coefficient, K a which is defined as the ratio of the biomolecule's

p rt'

concentration in the top and bottom phases. The partitioning results

from the biomolecule's interactions with the surrounding phases, which

34,42

are mainly through hydrophobic, hydrogen and ionic bonding The

16

many different interactions provides several methods of controlling the
partitioning by manipulation of:

1. kind of polymers,

2. molecular weight of the polymer,

3. concentration of the phase systems components,

4. type and concentration of salts added,

5. solution pH, and
6. temperature during extraction3a.
Over a wide range, partitioning does not depend on the

39,41

concentration of the desired product In addition, it has been

reported34 that it is the Kpa value that has the largest influence on

rt
the economics of an extraction.

The advantages for using PEG, and dextran or salt to form aqueous
two phase systems are the density and viscosity differences between the
two phasesao. In addition, dextran as well as PEG are nontoxic
substances and have been thoroughly tested for pharmaceutical and food
purposes.

However, most of the aqueous two phase systems have the
disadvantage of high costs because of highly purified phase components,
particularly dextranaa. Even so, two-phase systems are economical for

large scale isolation of certain proteins; e g., utilizing PEG/salt34

and PEG/crude dextran34’44.
The purification of a-amylase by non-ionic partitioning has been

reported previously using PEG/crude dextran44 and PEG/dextran44

2.3 Biospecific and Affinity Separations

2.3.1 Biospecific and Affinity Chromatography

Biospecific and affinity chromatography techniques utilize

17

immobilized ligand adsorbents to form reversible complexes with the

complementary biomolecules to be isolatedas-47. Affinity chromatography

techniques have absolute enzyme specificity by covalently attaching

ligands that function with the enzyme as either substrates, inhibitors,

cofactors, products or antibodie345-48. Biospecific chomatography

techniques have selectivity based upon ligands that are group

specificao, such as electrically charged groupsa or groups

hydrophobic in natureSI-SB; these ligands are group specific and are

capable of binding a number of similar biomolecules.

Techniques have been reported that use chromatography of a-amylase

26,27,59 60,61 l8,19,23,26-

by ion exchange , and affinity

28,62-64

, hydrophobic
adsorbents.
2.3.2 Affinity precipitation
Affinity precipitation utilizes specific interactions between the
enzyme and its substrates, inhibitors, cofactors, products or antibodies
to selectively form complexes that can be isolated by precipitation.
Two methods of precipitation are available, one wherein the complex is
insoluble upon formation and the other wherein the solution buffer must
be changed to cause the complex to precipitate65
a-Amylase has been purified by affinity precipitation by utilizing

20’23’24 and dextrinzs.

its interactions with glycogen
2.3.3 Biospecific and affinity partitioning

Aqueous two-phase partitioning has been developed by using polymers
that have been tailored to increase their specificity. This procedure
is sometimes referred to as biospecific and affinity partitioning39’42.

Biospecific partitioning of proteins has been most effective by

covalently combining polymers with ligands that are electrically charged

18
. 4O . . . .
or hydrOphobic in nature . These polymer-l1gands are biospec1f1c
adsorbents, i.e., capable of binding a number of similar biomolecules.
Biospecific partitioning differs from affinity partitioning only in

the degree of polymer-enzyme binding specificity42’47.

Affinity
partioning like affinity chromatography has absolute enzyme specificity
by covalently attaching substrates, products, inhibitors, cofactors or
antibodies specific to the enzyme to one of the polymer

38’39’42'47. Both biospecific and affinity partitioning take

phases
advantage of principles of affinity chromatography to chemically modify
a ligand so that it partitions exclusively, or nearly exclusively, to
the phase opposite of the impurities. A favorable effect on the
distribution of the protein will be obtained after it has been
specifically bound by the ligand-polymerBa’38-40’42’47.

A multistage process for the isolation of proteins by affinity
partitioning and the subsequent recycle of modified liquid polymer is a
somewhat more complex process than conventional affinity chromatography
using solid matrices. However, there are many advantages of multistage
affinty partitioning processes in large scale enzyme purification.
Extraction techniques scale up much better than chromatograpic
techniques41. Aqueous phase systems have considerably higher binding
capacities per unit volume than chromatography columns due to their
higher ligand density and availability66. In addition, approach to
equilibrium binding is faster in solution38. Finally, chromatography is
intrinsically a batch operation, while partitioning can be performed in
multistage operationsaa

The literature on hydrophobic and affinity chromatography includes

techniques using the covalent attachment of ligands to carbohydrate

19

supports45'46’48’67-80. Because the carbohydrate support is composed of
crosslinked carbohydrate polymers, the techniques used to attach ligands
for affinity chromatography can be used to design specific affinity and
biospecific partitioning reagents. However, there are reasonably priced
and FDA approved modified polymers available, though as yet not
developed for this use, for biospecific partioning of enzymes. The
polymer of choice for the present research is one of these existing
polymers.

Methylcellulose is a non-ionic polymer that was used with dextran
over 28 years ago to perform aqueous two-phase partitioning of
proteins37 and virus particles36. Methylcellulose has many advantages in

that its properties have been well characterizedgl’82

3

, it is inexpensive
at about 2.70/1b8 , and it has been approved many times by the Food and
Drug Administration as a polymeric surfactant for use as a food
additivesl. Methylcellulose is available as a methoxyl substituted or
as a methoxyl/-hydroxypropoxyl substituted cellulose derivative; it is
available in several molecular weights ranging from 15 to 250
kilodaltons. It also has unique solubility characteristics81’82,
allowing it to be gelled or salted out of solution. Also,
methylcellulose exhibits an inverse solubility, gelling and
precipitating at high temperatures more readilly than at low ones. This
is a good property to exploit with a-amylase thermostability.

The diversity of physical and chemical properties and FDA approval

provide impetus to use methylcellulose in aqueous two phase partitioning

of a-amylase.

20

2.4 Membrane Separations
2.4.1 Ultrafiltration

Ultrafiltration and microfiltration are membrane separation
processes that separate dissolved substances according to molecular
weight and size based upon their ability to pass through membrane pores
while under an applied pressure. Pore sizes of the membranes ranges
from 10 to 200 angstroms. Dissolved substances and solvents whose size
and molecular weight are below the membrane cutoff will pass through the
membrane while large molecules are retained. Molecules that pass
through the membrane are called the permeate; while those which are
retained by the membrane are called the concentrate.

Ultrafiltration is a relatively new but already popular
concentration and separation process. It still has drawbacks that
include low operating pressures84 and fouling problemsss. However,
improvements in its performance and new applications such as affinity
ultrafiltrationgé'89 are occuring rapidly.

2.4.2 Affinity ultrafiltration

Affinity ultrafiltrafiltration is a new purification method that
combines affinity binding with ultrafiltration separation586-89.
Affinity ultrafiltration is initiated by contacting a crude enzyme
extract with a solution of macromolecular ligand specific for the
desired enzyme (see section on biospecific partitioning). Upon contact,
the enzyme binds reversibly to the high-molecular-weight ligand, this
complex is retained on one side of a membrane while other material is

washed out through the membrane pores. Following isolation of the

ligand-enzyme macromolecule complex, the enzyme is liberated from the

21
ligand by addition of a dissociative media. The enzyme is isolated and
the ligand-macromolecule recovered by making a second pass through the
ultrafiltration membrane.
2.5 Summary
There are many publications describing protein separation, including
major reviews covering developments in separation and purification of

12’17’89. Techniques such as aqueous two-phase extraction

34,38-44 29,30,32,33

biomolecules

with non-ionic polymers and

40.42.47

, polyelectrolytes

34,38,39,42,47

biospecific and affinity polymers have been

detailed. There are also numerous papers on the chromatography of

enzymes by hydrophobic51'58, ion exchangeag'SI, and affinityl‘s'48

84,85 86-89

adsorbents. Ultrafiltration , and affinity ultrafiltration

provide alternatives for protein isolation.

There are many reports dealing with the purification of a-amylases

6,18,20,24,29,63 23,26-28,62 29
nts ,

isolated from animals fungi , and

4,19,25,33,59-61,64,91,92

pla

bacteria a-Amylase has been isolated using
fractionation techniques such as salting out4’6’18’19’20, solvent
precipitation4’6'23-27’61, isoelectric precipitationza, metal ion

precipitation“, thermal precipitation26-28, polyelectrolyte

precipitation29’33, and non-ionic polymer precipitationlg’hh. Affinity
20,23-25 .

precipitation has been used as well. Techniques involving

chromatography of a-amylase by ion exchange26’27’59, hydrophobic6o’6l

and affinityla’19'23v25'23.62-64

adorbents have been developed to a
similar extent.
With the number of purification techniques available, selection of

the "best" technique for the commercial production of protein becomes a

function of economics, and requirements for final purity and yield.

22
Techniques such as aqueous two-phase biospecific or affinity
partitioning can be economical, with high yields at high puritities39
A technique such as this was used in the present study of a-amylase
(chapter 3). Recombinant-DNA technology has been tried to enhance
purification techniques by modifying the protein to facilitate its
isolation. The effectiveness of some protein modifications are
evaluated in chapters 3 and 4.
2.5 Literature Cited

1. Crueger. W., and A. Crueger. 1984. o e o o ' te tboo

ingg§§11§l_gigyghiglggy. Science Tech, Inc., Madison, Wisconsin. 308 pp.

2. Aunstrup, K., 0. Anderson, E. A. Falch, and T. K. Nielsen. 1979.
Production of microbial enzymes. Pages 281-309. In: H. J. Peppler, and
D. Perlman, eds. Eigrobigl teghgology, vol.l. Academic Press, New York,
New York.

3. Wolnak, B., and Associates. December, 1972. Present and future
technological and commercial status. Report prepared for the National
Science Foundation. U.S. Dept. Commerce Natl. Tech. Inf. Serv. Doc. PB-
219 636.

4. Welker, N. E., and L. Leon Campbell. 1967. Crystallization and

properties of a-amylase from five strains of bacillus amyloliguefaciens.
Biochemistry. 6(12):3681-3689.

5. French, D. 1975. Chemistry and biochemistry of starch. Pages 267-335.
In: W. J. Whelan, o e st a be drates. Biochem series one, vol.
5. Butterworths, London.

6. Bernfeld, P. 1955. Amylases, alpha and beta. Pages 149-154. In: S. O.
Colowick and N. 0. Kaplan, eds. Me ods in enz olo . vol I. Academic
Press, New York, New York.

7. Oriel, P. and A. Schwacha. 1988. Growth on starch and extracellular
production of thermostable amylase by E. coli. Enzyme Microb. Technol.
10:42-46.

8. Davis, R. W., D. Botstein, and J. R. Roth. 1980. Advanced Bacterial
Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York,
New York. 201 pp.

9. Manning, G. B., and L. L. Campbell. 1961. Thermostable a-amylase from
Eagillus steggotgggmgphilus. J. Biol. Chem. 236:2952-2957.

10. Tsukagoshi, 8., et a1. 1984. Cloning and expression of a

23

thermophilic a-amylase gene from Egcillus stearothermophilus in E. coli.
Mol. Gen. Genet. 193:58-63.

ll. Tamuri, M., et a1. 1981. US Pat. Appl. 4 284 722.

12. Belter P. A., E. L. Cussler, and Wei-Shou Hu. 1988. Bioseparations:
W- John Wiley & Sons. New York.
New Yerk. 368 pp.

13. Oliver, D. 1985. Protein secretion in E. coli. Ann. Rev. Microbiol.
39:615-648.

 

 

14. Ghrayeb, J., et a1. 1984. Secretion cloning vectors in E. coli. EMBO
J. 3(10):2437-2442.

15. Lunn, C. A., M. Takahara, and M. Inouye. 1986. Secretion cloning
vectors for guiding the localization of proteins in vivo. Current Topics
In Microbiology and Immunology. 125:59-74.

16. Brewer, S. J., and H. M. Sassenfeld. 1985. The purification of
recombinant proteins using C-terminal polyarginine fusions. Trends in
Biotechnology. 3(5):119-121.

17. Bell, D. J., M. Hoare and P. Dunnill. 1983. The formation of protien
precipitates and their centrifugal recovery. Advances in Biochemical
Engineering. 26:1-72.

18. Buonocore, V., and E. Poerio. 1975. Affinity column purification of
amylases on protein inhibitors from wheat kernel. Journal of
Chromatography. 114:109-114.

19. Monma M., K. Mikuni, H. Ishigami, and K. Kainuma. 1987. Purification
of the glucoamylase components of ghglgyg pagggoxa by affinity
chromatography and chromatofucusing. Carbohydrate Research. 159:255-261.

20. Schramm, M., and A. Loyter. 1966. Purification of a-amylase by
precipitation of amylase-glycogen complexes. Methods in Enzymology.
8:533-537.

21. Toda, H., and K. Narita. 1968. Correlations of the sulfhydryl group
with the essential calcium in E. subtilis saccharifying a-amylase. J.
Biochem. 63(3):302-307.

22. Vallee, B. L., E. A. Stein, W. M. Summerwell, and E. H. Fisher.
1959. Metal content of a-amylase of various origins. J. Biol. Chem.
234:2901.

23. Silvanovich, M. P., and R. D. Hill. 1976. Affinity chromatography of
cereal a-amylase. Analytical Biochemistry. 73:430-433.

24. Loyter, A., and M. Schramm. 1963. The glycogen-amylase complex as a
means of obtaining highly purified a-amylase. Biochim. Biophys. Acta.
80:200-206.

24

25. Levitzki, A., J. Heller and M. Schramm. 1964. Specific precipitation
of enzyme by its substrate: The a-amylase-macrodextrin complex.
Biochim. Biophys. Acta. 81:101-107. ‘

26. Kruger, J. E., and R. Tkachuk. 1969. Wheat a-amylase. I. isolation.
Cereal Chem. 46:219-226.

27. Tkachuk, R., and J. E. Kruger. 1974. Wheat a-amylase. II. physical
characterization. Cereal Chem. 51:508-529.

28. Tkachuk, R. 1975. Competitive affinity chromatography of wheat a-
amylase. FEBS Letters. 52(1):66-68.

29. Sternberg, M., and D. Hershberger. 1974. Separation of proteins with
polyacrylic acid. Biochim. Biophys. Acta. 342:195-206.

30. Clark, K. M., and C. E. Glatz, 1987. Polymer dosage considerations
in polyelectrolyte precipitation of protein. Biotechnology Progress.
3(4):241-247.

31. Smith, A. K., A. M. Nash, A. C. Eldridge, and W. Wolf. 1962.
Recovery of soybean whey protein with edible gums and detergents. J.
Agricul. Food Chem. 10:302-304.

32. Sternberg, M. 1970. The separation of proteins with heteropolyacids.
Biotechnology and Bioengineering. 12:1-17.

33. Sternberg, M. 1976. Purification of industrial enzymes with
polyacrylic acids. Process Biochemistry. 11(7):11-12.

34. Hustedt, H., K. H. Kroner, U. Menge, and M.-R. Kula. 1985. Protein
recovery using two-phase systems. Trends in Biotechnology. 3(6):139-144.

35. Albertsson, P. A. 1986. Eargitigg of cell particles and
mggggmglggglgg. 3rd ed., Wiley-Interscience, New York, New York.

36. Albertsson, P. A., and G. Frick. 1960. Partition of virus particles
in a liquid two-phase system. Biochim. Biophys. Acta. 37:230-237.

37. Albertsson, P. A. 1958. Nature. 182:702.

38. Kula, M.-R., K. H. Kroner, and H. Hustedt. 1982. Purification of
enzymes by liquid-liquid extraction. Pages 73-117. In: A. Fiechter, eds.

Eggggign_Enginggzing. Springer-Verlag, New York, New York.

39. Mattiasson, B., and R. Kaul. 1986. Use of aqueous two-phase systems
for recovery and purification in biotechnology. American Chemical
Society. 314:78-92.

40. Johansson, G. 1987. Dye-ligand aqueous two-phase systems. Pages 101-
124. In: Y. D. Clonis, T. Atkinson, C. J. Bruton, and C. R. Lowe, eds.

Egagtive dyes in pgotein and enzyme technology. Stockton Press, New
York, New York.

25

41. Kroner, K. H., H. Hustedt, S. Granda, and M.-R. Kula. 1978.
Technical aspects of separation using aqueous two-phase systems in

enzyme isolation processes. Biotechnology and Bioengineering. 20:1967-
1988.

42. Mattiasson B. 1983. Applications of aqueous two-phase systems in
biotechnology. Trends in Biotechnology. l(1):16-20.

43. Kula, M.-R., K. H. Kroner, H. Hustedt, and H. Schutte. 1981.
Technical aspects of extractive enzyme purification. Ann. N. Y. Acad.
Sci. 369:341-354.

44. Kroner, K. H., H. Hustedt, and M.-R. Kula. 1982. Evaluation of crude
dextran as phase-forming polymer for the extraction of enzymes in

aqueous two-phase systems in large scale. Biotechnology and
Bioengineering. 24:1015-1045.

45. Vretblad, P. 1974. Immobilization of ligands for biospecific
affinity chromatography via their hydroxyl groups: The
cyclohexylamylose-A-amylase system. FEBS Letters. 47(1):86-89.

46. Cuatrecassas, P. 1972. Affinity chromatography of macromolecules.
Advan. Enzymol. 36:29-89.

47. Clonis, Y. D. 1987. Dye-ligand chromatography. Pages 33-49 13 Y. D.
Clonis, T. Atkinson, C. J. Bruton, and C. R. Lowe. Eeactive dyes in

pIg5g1n_§ng_gngyn§_;gghnglggy. Stockton Press, New York, New York.

48. Schwacha. A. 1983. Affinity chromatography: Principles and methods.
P8868 12-110. Wiggle. Piscataway. New

Jersey.

49. Simmonds, R. J., and R. J. Yon. 1976. Protein chromatography on
adsorbents with hydrophobic and ionic groups. Biochem. J. 157:153-159.

50. Yon, R. J., and R. J. Simmonds. 1979. The adsorption of proteins and
protein-dodecyl sulphate complexes on N-(3-carboxypropionyl) aminodecyl-
sepharose. Biochem. J. 177:417-424.

51. Halperin, G., M. Breitenback, M. Tauber-Finkelstein, and S.
Shaltiel. 1981. Hydrophobic chromatography on homologous series of

alkylagaroses: A comparison of charged and electrically neutral column
material. J. Chroma. 215:211-228.

52. Yon, R. J. 1978. Recent developments in protein chromatography
involving hydrophobic interactions. Internat. J. Biochem. 9(6):373-379.

53. Melander, V., and C. Horvath. 1977. Salt effects on hydrophobic
interactions in precipitation and chromatography of proteins: an
interpretation of the lyotropic series. Arch. Biochem. Biophys. 183:200-215.

54. Jennison, H. P., and L.M. Heilmeyer, Jr. 1975. General aspects of

26

hydrophobic chromatography. adsorption and elution characteristics of
some skeletal muscle enzymes. Biochemistry. 14(4):754-760.

55. Srinivasan, R., and E. Ruckenstein. 1980. Role of physical forces in
hydrophobic interaction chromatography. Separation and Purification
Methods. 9(2):267-370.

56. Shaltiel. S. 1975. Hydrophobic chromatography: Use in the
resolution, purification and probing of proteins. Proceedings of the
Tenth FEBS Meeting. 117-127.

57. Rosengren, J., J. Pahlmany, S. Glad, M. et a1. 1975. Hydrophobic
interaction chromatography on non-charged sepharose derivatives.
Biochim. Biophys. Acta. 412:51-61.

58. Nemat-Gorgani, M., and K. Karimian. 1982. Non-ionic adsorptive
immobilization of proteins to palmityl-substituted sepharose 4B. Eur. J.
Biochem. 123:601-610.

59. walker, N. E., and L. Leon Campbell. 1967. Comparison of the a-

amylase of Eggillgg fighgilig and Egcillus amyloliguefaciens. J.
Bacteriology. 94(4):1l31-1135.

60. Sada, Eizo, Shigeo Katoh, Tsuneo Inoue, and Masami Shiozawa. 1985.
Performance of hydrophobic chromatography in purification of a-amylase.
Biotechnology and Bioengineering. 27:514-518.

61. Nemat-Gorgani, M., K. Karimian and A. R. Massih. 1984. Effect of
salt concentration on binding of proteins to a non-ionic adsorbent.
Experintia. 40:81-83.

62. Vretblad, P. 1974. Biospecific affinity chromatography of sweet-
potato B-amylase. Biochemical Society Transactions. 2:1327-28.

63. Burrill, P. H., P. N. Brannon, and N. Kretchmer. 1981. A single-step
purification of rat pancreatic and salivary amylase by affinity
chromatography. Anal. Biochem. 117:402-405.

64. Schell, H. D., M. A. Mateescu, T. Bentia and A. Jifcu. 1981. a-
Amylase purification and separation from glugoamylase by affinity
chromatography on cross linked amylose. Analytical Letters. 14:1501-
1514.

65. Flygare, S., T. Griffin, P. Larsson, and P. Mosbach. 1983. Affinity
precipitation of dehydrogenases. Anal. Biochem. 133:409-416.

66. Kroner, K. H. et al. 1982. Page 491. In: T. C. Gribnau, J. Visser, R.
J. F. Nivard, eds. it ch to ra h and related techni ues.
Elsevier, Amsterdam.

67. Kennedy, J. F. 1978. Chemical synthesis and modification of
oligosaccharides, polysaccharides, glycoproteins, enzymes and
glycolipids. Carbo. Chem. 10:427-497.

27

68. Butler, L. G. 1975. Enzyme immobilization by adsorption on
hydrophobic derivatives of cellulose and other hydrophobic materials.
Arch. Biochem Biophys. 171:645-650.

69. Hjerten, S., J. Rosengren, S. Pahlman. 1974. Hydrophobic interaction
chromatography: The synthesis and the use of some alkyl and aryl
derivatives of agarose. J. Chrom. 101:281-288.

70. Horejsi, V., and J. Kocourek. 1973. Studies on Phytohemagglutinins.
XII. o-glycosyl polyacroamide gels for affinity chromatography of
phytohemagglutinins. Biochim. Biophys. Acta. 297:346-351.

71. Matsumoto, T., and T. Osawa. 1972. The specific purification of
various carbohydrates binding hemaglutinins. Biochemical and Biophysical
Research Communications. 46(5):1810-1815.

72. Kristiansen, T., L. Sundberg, and J. Porath. 1969. Studies on blood
group substances II. coupling of blood group substance A to hydroxyl-
containing matrices, including animoethyl cellulose and agarose.
Biochim. Biophys. Acta. 184:93-98.

73. Gordon, J., et a1. 1972. Purification of soybean agglutinin by
affinity chomatography on sepharose derivatives. FEBS Letters.
24(2):193-196.

74. Barker, R., et a1. 1972. Agarose derivatives of uridine diphosphate
and N-acetylglucosamine for the purification of a galactosyltransferase.
J. Biol. Chem. 247(22):7135-7147.

75. Junowicz, E., and J. E. Paris. 1973. Affinity chomatography by
enzyme-subsrate interaction. purification of some rat liver
glycosidases. Biochim. Biophys. Acta. 231:234-245.

76. Rafestin, M. E., et a1. 1974. Purification of N-acetyl D-
glucosamine-binding proteins by affinity chromatography. FEBS Letters.
40(1):62-66.

77. Hayes, C. E., and I. J. Goldstein. 1974. An a-D-galactoseyl-binding

lectin from Egnggigggg simpligifolia seeds: Isolation by affinity
chromatography and characterization. J. Biol. Chem. 249(6) 1904-1914.

78. Ellingboie, J., et a1. 1970. Liquid-gel chromatography on .
lipophilic-hydrophobic sephadex derivatives. J. Lipid Res. 11:266-273.

79. Rosengren, I., et a1. 1975. Hydrophobic interaction chromatography
on non-charged sepharose derivatives: Binding of a model protein,
related to ionic strength, hydrophobicity of the substituent, and degree
of substitution (determined by NMR). Biochim. Biophys. Acta. 412:51-61.

80. Keillich, G., et a1. 1972. Optical rotary dispersion and circular
dichroism of benzoyl polysaccharides. Biopolymers. 11:1997-2013.

28

81. Greminger, G. K. Jr., and K. L. Krumel. 1980. Alkyl and
hydroxyalkylalkylcellulose. Pages 3-1 to 3-25 In: R. L. Davidson, ed.

flandbook of Wate; goluplg gpgg and resins. McGraw Hill, New York.

82. Neely, W. B. 1963. Solution properties of polysaccharides. IV.
Molecular weight and aggregate formation in methylcelluse solutions, J.
Polymer. Sci. 1(1):311-320.

83. ghgmippl_E§;k§§1pg_ngp;§g;. August 1. 1988. Schnell Publishing
Company, New York, New York.

84. Lonsdale, H. K. 1979. Theory and practice of reverse osmosis and
ultrafiltration. Pages 123-178. In: R. E. Lacey, and S. Loeb. ngpstrial

pLpggggipg_yigh_mgmh1gng§. Robert E. Krieger Publishing Co., Huntington,
New York.

85. Devereux, N., and M. Hoare. 1986. Membrane separation of protein
precipitates: Studies with cross flow in hollow fibers. Biotechnology
and Bioengineering. 28:422-431.

86. Mattiasson, B., and M. Ramstorp. 1983. Ultrafiltration affinity
purification. Ann. N. Y. Acad. Sci. 413:307-309.

87. Mattiasson, B, and T. G. I. Ling. 1986. Ultrafiltration affinity
purification: A process for large-scale biospecific separations. Pages

99-114. In: W. C. McGregor, edt. Egppggpe separations in Eiotechnolpgy.
Marcel Dekker, Inc., New York, NY.

88. Luong, J. H. T., A.-L. Nguyen, and K. B. Male. 1987. Affinity cross
flow filtration for purifying biomolecules. Biotechnology. 5:564-566.

89. Pungor, E. Jr., N. B. Afeyan, N. F. Gordon and C. L. Cooney. 1987. A
continuous affinity-recycle extraction: A novel protein separation
technology. Biotechnology. 5:604-608.

90. Null, Harold R.1987. Selection of a separation Process. Pages 982-

995. In: R. W. Rousseau, edt. nggbopg of separatiop process
ggghnplpgy. John Wiley & Sons, New York, New York.

91. Kochhar S., Batra R., Dua R.D., 1984. Purification and
characterization of thermostable a-amylase from Bacillus
amyloliquefaciens. Enzyme Microbiol Tech. Communicated.

92. Nagata, Y., S. Suga, O. Kado and B. Maruo. 1980. N-Terminal amino
acid sequence of a-amylase from Epgillpg subtilis var

ticu : comparison with that of liquefying type a-amylase.
Agric. Biol. Chem. 44:215-216.

CHAPTER 3

CONCENTRATION AND PURIFICATION OF a-AMYLASE USING A

METHYLCELLULOSE-SALT, TWO-PHASE PARTITIONING PROCESS

3.1 Abstract

The downstream purification of thermostable a-amylase was studied
utilizing a technique involving aqueous two-phase partitioning with
methylcellulose and ammonium sulfate. The partitioning was optimized
with respect to the type of cellulose derivative, its molecular weight
and concentration, the type and concentration of salts added, the
solution pH, and the temperature during extraction. The partitioning
was shown to be reversible upon lowering the salt concentration,
allowing the recovery of methylcellulose from the enzyme. A process was
demonstrated for recovering a-amylase from crude broth using
methylcellulose/salt partitioning.
3.2 Introduction

Molecular biology and recombinant-DNA technology have combined to
provide control over the kinds and composition of proteins
biosynthesized by microorganisms such as E. p913. Commercial
applications require development of efficient methodologies to
concentrate and purify extracellular secreted enzymes from culture
broths. Production of low priced industrial enzymes on a large scale,
however, often requires significantly different recovery processes than
the standard fractionation and chromatographic techniques that have
predominated in the research literaturel. Aqueous two-phase systems,e.g.
PEG/salt systems, have seen some use in multistage processes for the

29

3O
purification of proteins, but require PEG recycling to remain economical2
This technique has been fairly successful due to the gentleness of

partitioning1'3'a, the large density and viscosity differences between

the two phasess, the ability to control the partitioning specificityz’s,
and the acceptance of PEG by the Food and Drug Administration.

The specific partitioning results from the various interactions of
the biomolecules with the surrounding phases mainly through hydrophobic,

hydrogen, and ionic bonding2 These interactions provide options for
controlling the partitioning by manipulation of2

1. kind of polymers,

2. molecular weight of the polymer,

3. concentration of the phase system components,

4. type and concentration of salts added,

5. solution pH, and

6. temperature during extraction.

To increase the specificity of partitioning in aqueous two phase
systems, the polymer phase can be tailored with ligands that are
hydrophobic or electrically charged3. The ligand-tailored polymer phase
is then group specific, or in this case biospecific, because it is
capable of binding a number of similar biomolecules. Selecting a
biospecific polymer to partition the desired enzyme therefore depends
upon the chemistry of the enzyme. The choice of ligand may be
indicated by the ligands effect on the solubility of proteins with
similar structure6. For example, proteins with relatively small non-
peptide groups such as lipo-, nucleo-, and glycoproteins have solubility

properties often distinctive to these groups.

When the purification of the enzyme a-amylase from E. coli

31
(designated ATCC 29609) was investigated, it was found that the enzyme
was strongly hydrophobic7. Thus, a hydrophobic polymer/salt, two-phase
aqueous partitioning system was selected.

An existing hydrophobic polymer could be utilized or one could be
synthesizeds, utilizing a PEG or carbohydrate backbone. Hydrophobic
ligands can be attached to a carbohydrate polymer backbone, as descibed
by techniques used for hydrophobic and affinity chromatography8'24,
because the carbohydrate support matrix is composed of crosslinked
carbohydrate polymers, and similarly for PEG backbones. However, there
are reasonably priced and FDA approved hydrophobic polymers available,
though not as yet developed for this use, that fill the requirements for
a two-phase partitioning process that could be biospecific.

Methylcellulose is one such non-ionic polymer that was used with
dextran over 28 years ago to perform aqueous two-phase extractions of
proteinszs. In addition, it has many advantages including well-

26’27, low cost at $2.70 per lb28, and FDA

characterized properties
approval26 as a polymeric surfactant for use as a food additive. Also,
because methylcellulose comes in several molecular weights and as
methoxy (MC) or hydroxylpropoxyl/ methoxy (HPMC) substituted cellulose
derivatives, the partitioning can be optimized for molecular weight and
somewhat for ligand substitution.

The nature of interaction that methylcellulose brings to the
partitioning is indicated by its structure (Figure 3.1). MC and HPMC
are neutral, non-ionic polysaccharides whose unbranched hydrophilic
backbones, are interspersed with small hydrophobic groups along their

length826. They have an inverse solubility, gelling at increasing

temperatures, which could be used advantageously with thermostable a-

H0

32
amylase. They are not susceptible to chemical gelation or precipitation
with di- or trivalent metals, with borates, or by interaction with other
polymers to form complexes or coacervate526. However, in aqueous
solution, where they exist as highly hydrated colloids, MC and HPMC can
be gelled or salted out of solution when the concentration of added

. . ~ 6
solutes or electrolytes exceeds certain l1m1ts’2

 

 

 

H OH H OCH,
H
OCH, H OCH, H
H H
C) H r_____43 (DH
CH20CH, _. H OCH, -n CH10H

Figure 3.1. The structure of methoxy substituted cellulose (MC)
a-Amylase excreted extracellularly by E. p21; is thermostable (with
little loss of activity at 90°C for 1 hr)29. Taking advantage of the
characteristics of a-amylase and methylcellulose described above, a
process for concentration and purification of the enzyme downstream from
the fermentor has been studied. The process involves removal of
insolubles by centrifugation, concentration by ultrafiltration, soluble
contaminant removal by thermal precipitation, and aqueous two-phase
partitioning for further amylase purification.
3.3 Experimental
Materials

The methylcellulose and hydroxypropyl methylcellulose were donated

33

by Dow Chemical and have brand name METHOCEL type A4M, K4M, KISM,
and KlOOM designating the cellulose derivative (A for 26-33 %
methoxy substitution and K for 19-24 % methoxy/7-12 % hydroxypropoxyl
substitution) and the approximate molecular weight (4, 15 and 100
corresponding to 85, 15, and 250 kilodaltons, respectively). The a-
amylase was obtained as described below. All other chemicals were
reagent grade.
Amylase Preparation

a-Amylase purified to homogeniety was used in the six partition
optimization experiments. The a-amylase was prepared and purified
from transformant EC 147 periplasmic extractsBo, grown overnight in
cultures shaken at 37°C in L broth31, as described by P. Orielzg.
Purification of the amylase extract involved fractionation by heat

treatment at 80°C for 45 min in the presence of 15 mM CaCl and salting

2.
out with ammonium sulfate, retaining the O to 55% saturation fraction.
After redissolving the precipitate in buffer (0.05 M Tris, pH 8.0,
containing 0.025 M CaClZ), further impurities are removed by DEAE
Sepharose chromatography, which passes the a-amylase. Purification to
homogeneity is obtained by gel permeation chromatography on Sephadex G-
200, and validated by SDS-PAGE electrophoresis. The pure a-amylase was
stored frozen in distilled water, and prior to experiments diluted in 50
mM acetate buffer, pH 6.0, containing 5 mM CaC12.

After the a-amylase partitioning was optimized, further experiments
were carried out to evaluate the a-amylase purification by
methylcellulose/salt partitioning from a broth. The a-amylase

was produced as just described. However, prior to separation of the

extracellular fraction from the cells by centrifugation, the volume was

34

reduced with a 10,000 moleculer weight cut off, cross flow
ultrafiltration unit. Following a-amylase extraction from the periplasm
by osmotic shock with sucrose30, both the periplasmic extract and the
supernatant were further fractionated by heat treatment as described
above. Following the heat fractionation step, the further purification
of a-amylase from the periplasmic extract and the extracellular
supernatant was studied for comparison purposes (sections 3.4.2 &
3.4.3).
Methylcellulose Preparation

The methylcellulose powder was dispersed and wetted in 50 mM

acetate buffer, pH 6.0, containing 5 mM CaCl solution by slow addition

2
of the powder to the hot buffer, around 80-90°C. The wetted
methylcellulose was then poured into an equal volume of cold buffer,
around 4°C, and mixed. Methylcellulose solutions thus formed26 were
stored at 4°C between experiments for several months with no apparent
changes.
Partitioning

The term partitioning is used here to indicate the localization of
protein upon formation of two phases. The exact physical nature of the
two phases is somewhat unclear. With ammonium sulfate-methylcellulose
partitioning, the polymer phase can appear as liquid droplets under a
microscope while resembling a precipitate in other characteristics.

Experiments involving homogeneously pure a-amylase were performed
on a small scale using 1.5 m1 Eppendorf tubes. However, experiments
evaluating purification and the nature of partitioning were performed on

a scale of several ml.

Equal volumes of methylcellulose and a-amylase solutions, prepared

35

as described above, were added in various concentrations and mixed
vertically at room temperature. After approximately 30 minutes with
occasional mixing, saturated solutions of ammonium sulfate were added to
the samples, then immediately mixed to produce the desired salt
concentration. The addition of ammonium sulfate to the a-amylase and
dilute (0.5-0.0001%) methylcellulose solution creates a precipitate-like
polymer phase and a salt phase. After waiting an additional 30 minutes,
while mixing ocasionally, the precipitate was pelleted by centrifuging
the Eppendorf tubes at 13,000 rpm for 20 minutes. The resulting
supernatant and/or pellet were then tested, after appropriate dilutions,
for total reducing activity, total carbohydyrate, and/or total protein.
A measure of total protein was not always possible because effort to
obtain highly pure a-amylase and the dilutions used in most of the
experiments reduced the protein levels to less than that assayable by
the Lowry et a1. procedure33
Assays

a-Amylase activity was estimated by a modification of the
dinitrosalicylate assay by Bernfe1d32. The assay requires photometric
measurement at 640 nm of the optical density arising from the reducing
power of a solution of soluble starch in the presence of a-amylase32
One unit of activity is defined as the amount of enzyme that liberates
starch hydrolysis products at a rate equivalent to the reducing capacity
of one umol of maltose per minute at 60°C.

The method of Lowry et al32 was used to estimate total protein;
bovine serum albumin was the protein standard. The contribution of the
methylcellulose to the total protein measurement was negligible.

The phenol sulfuric method by Dubois et al34 was used to estimate

36
total carbohydrate; the appropriate methylcellulose calibration curve
was the carbohydrate standard. The contribution of the pure a-amylase
to the total carbohydrate measurement was insignificant.
3.4 Results and Discussion
3.4.1 Optimization of the Partitioning Process

Partitioning can be optimized by the choice of polymer, its
molecular weight, its concentration, the type and concentration of salt
added, the solution pH, and the temperature during extractionz. The
effects of all of these variables on the partitioning of homogeneously
pure a-amylase were evaluated in the present study.

The partitioning is presented in terms of the amount of starch
reducing activity remaining in the aqueous salt phase. An experiment
was carried out to verify that the activity missing from the salt phase
is actually in the methylcellulose phase. The results showed 93% of the
total activity remained in the polymer phase, 1% remained in the salt
phase, and the 6% unaccounted was possibly due to experimental error.
The two phase system included 0.02% w/v MC, 30% saturation ammonium
sulfate and 90 units/ml a-amylase.

Choice of polymer and the effect of its molecular weight

The choice of polymer and the effect of its molecular weight on
partitioning a-amylase was evaluated for solutions of 0.0005% w/v
polymer A4M, K4M, K15M, and K100M containing 30% saturated ammonium
sulfate, and 90 units/m1 a-amylase. The percent activity remaining in
the salt phase after partitioning was 13.8, 12.9, 6.0, and 5.4% for A4M,
K4M, KlSM and K100M polymer types, respectively. The HPMC, of molecular
weights 15 and 250 kilodaltons, enables better partitioning than MC or

HPMC of molecular weight 85 kilodaltons, with MC and HPMC of similar

37

molecular weights partitioning similarly. In addition, when choosing
between HPMC of molecular weights 15 or 250 kilodaltons, the lower
molecular weight was used for lower solution viscosity.
Effect of methylcellulose concentration

The effect of methylcellulose concentration on partitioning a-
amylase was evaluated for solutions of varied methylcellulose
concentrations, each containing 30% saturated ammonium sulfate, and 90
units/ml a-amylase. The results were evaluated as a percentage of
activity remaining in the salt phase as a function of the enzyme loading
(Figure 3.2). In another experiment, the concentration of enzyme was
varied while the methylcellulose concentration and ammonium sulfate
saturation were fixed at 0.005% and 30%, respectively. A similar
loading profile was produced (Figure 3.2). The results indicate that a
loading of between 1 and 10 g A4M per g amylase is required to partition
the majority of a-amylase.
Effect of different salts

Ammonium sulfate and sodium chloride were evaluated to determine
their ability to partition a-amylase as a function of their
concentration. The concentrations of K4M and a-amylase were 0.05% w/v
and 90 units/ml, respectively. The starch reducing activity of the salt
phase was measured after partitioning with ammonium sulfate or sodium
chloride at 0, 15, 20, 25, and 30% of saturation. The starch reducing
activity was measured as the percentage of activity remaining in the
salt phase after partitioning as a function of salt concentration
(Figure 3.3). At less than 10% of saturation, ammonium sulfate had
little or no effect on partitioning a-amylase, but at 30% saturation,

nearly all of the amylase was separated from solution. At similar

all III

 

38

6039.38 323m 82:02.80 Lo Nmm Lo mcfioo.
oEcho o L to 203051 o no one: tom 9: E
mEEoEoL £26.00 to omoycoocon one ”Nam “Ln/.50....

@8356 32.} 3 02.9.9 ”12>sz

ON m— or .vp NP OP m w .v N o
_.Lr.»dplt_t_.e._.Ltnto

 

T
r

 

 

 

 

 

4 2% Hood 2 29252828 ”12>sz ez_>m<> 1
HS»? 536:2: om E zocstzmozoo 212 oz.>m<> one
_ p a a _ a m P blip — . m . _ P P P n - CO —

 

ESVHd .L'lVS NI ONINIVWEH All/\llOV Z

39

.mcorocfiom 02.820 8298 to 30:45
E:_coEEo of Lo c0305: 0 mo omega tom 9: E
mEEoEE 33500 .8 $3588 9:. ”m6 mmDOE

Him do ZO_._.<mD._.<m N

 

 

0%... MN GEN .vN N_N O_N m... on? JP N__. m _. m w W LN O
20F<m2<m m._.<h_.._3m EQZOEE< no 29.—.023."— < m< Ola -O
.. ZOP<MDF<m wag—MOAIO EDEOW m0 20:023.”. < m< I r
I TON
.. :94
I. Tom
T low
a _ F _ P F . — . n a P n _ n _ . h _ . '7 GOP

 

 

 

BSVHd INS NI ONINIVWHH All/\IIOV Z

4O

saturations, sodium chloride partitioned less a-amylase than ammonium
sulfate did.
Effect of pH

The effect of pH on partitioning a-amylase was investigated in
solutions containing 0.05% w/v K4M, and 90 units/m1 a-amylase. Each
solution was prepared with buffers of pH 4.0, 5.0, 6.0 or 7.0 prior to
partitioning with ammonium sulfate to 30%of saturation. After
partitioning, the starch reducing activity was measured and reported at
each pH as a percentage of that remaining in an identical, unpartitioned
solution (Figure 3.4). In addition, the pH of each solution was
measured following partitioning with ammonium sulfate to get a true
measure of the solution pH. The results showed little difference in
partitioning at the different pH's tested with maximum partitioning
between 5 and 6.
Effect of temperature

The effect of temperature on partitioning was investigated in
solutions containing 0.05% K15M, 9O units/m1 a-amylase, and ammonium
sulfate at 30% of saturation. The solutions were held at constant
temperature from 30 minutes prior to partitioning until 30 minutes after
partitioning, when the activity of the salt phase was sampled. The
temperatures investigated were 1, 23, 46, and 60°C. The percentage of
a-amylase activity remaining in the salt phase at each temperature
(Appendix) are within 1% of each other, leading to the conclusion that
temperature does not play an important role in partitioning a-amylase

into methylcellulose.

41

62 833 32.3 825880 do
Now yo I Lo c0323 6 mo omocd tom 05 E
mc_c_oEo._ 33:00 1.0 o Scented one ”En mmDOE

 

 

 

 

Basso rise: cm 25 23. are To
_ . _ .

 

on

ESVHd .L‘lVS Nl ONINIVWBH All/\llOV %

42

3.4.2 Purification

The purification of a-amylase from contaminating proteins is
demonstrated by indicating the percent total protein as well as the
percent amylase activity remaining in the salt phase after partitioning.
The two incompletely purified a-amylase solutions, described in the
experimental section, were used to test for purification by partitioning
as a function of (i) methylcellulose concentration and (ii) ammonium
sulfate saturation. In addition, the purifications obtained by
methylcellulose partitioning as a function of ammonium sulfate
saturation were used for comparison against ammonium sulfate
precipitation.

The experiments on ammonium sulfate precipitation and two-phase
partitioning were done with 3 m1 solutions, to which 100% saturated
ammonium sulfate was added. The concentration of methylcellulose,
ammonium sulfate saturation, enzyme activity, and total protein of these
solutions are given in Table 3.1. The total protein was measured after
membrane dialysis, using 12 to 14 kilodalton molecular weight cutoff
bags, against 50 mM acetate buffer, pH 6.0, containing 5 mM CaCl2

The results shown in Figures 3.5a or 3.5b indicate the percentage
of activity or total protein remaining in the salt phase after
partitioning of, extracellular or periplasmic extract a-amylase,
respectively. The results shown in Figures 6a and 6b indicate the
percentage of activity or total protein remaining in the salt phase
after precipitation of extracellular or periplasmic extract a-amylase,

respectively. The above results are shown with respect to ammonium

sulfate saturation. However, the results shown in Figure 3.7 are with

43

Table 3.1. The a-amylase activity and total protein of
solutions prepared for ammonium sulfate precipitation and
methylcellulose/salt partitioning (partitioning as a
function of (i) ammonium sulfate saturation and as a

function of (ii) methylcellulose concentration).

, [K15M] Salt Activity Tot. Prot.
Solution (% w/v) (% Sat.) (units/ml) (ug/ml)
Extracellular
Precipitation o.oo var.a 120 830
Partitioning (i) 0.05 var.b 52 440
Partitioning (ii) var.C 30 28 230

Periplasmic Extract

Precipitation 0.00 var. 1850 300
Partitioning (i) 0.05 var.b 950 96
partitioning (ii) var.c 30 102 10

aAmmonium sulfate saturations varied from 0 to 35%.

bAmmonium sulfate saturations varied from 0 to 70%.

ICKISM concentration varied from 0 to 0.01% w/w.

44

.omoSEo to_:__ooobxo. co mchEtoo
t3 o 39.3 tom of E 9:50an 530.5

.6 o._>_o.oo to @0350th or: .om. m ”HEDGE

m._.<n_n5m 222022< “.0 ZO_._.<mD._.<m Hzmommm
¢MN_nonmNoN¢NNNONmF 0P *9 NF ow w m ¢ N o

Db-b bLFPL-LFb—Dbllli—rhn—I—b—D—hgbblPBbl o
wm<I¢ ._.-.<m z. >_._>_._.0< Ola

r mm<In_ Him 2. z_w._.omm 45.0... I .

 

 

   

 

 

_.bL_pLL+

_._L_bhlrb u.L.

 

SNINIVWBEJ NIELLOHd GNV MIAIiOV 34

45

 

.omoSEo yoobxo o_Emo_a._._mo co mchEtoa
LB o $93 :3 of E mEEoEm: £396
.5 E58 to $3588 9: "new #50:

m._.<...33m 2220224 no ZO_._.<m3._.<m HZMOKMQ
*nmnonmmmmtsmmmommw of: NP 0F m o .v N o

.u.LF..LiuicF.._.LL_._.Llr_t_.LL_. O
mmSi .5<m z. >._._>_._.o< Ola

 

mm<Im .5<m z. Z_m._.omn_ 45.0... I

 

 

~._.h._._.P.___p_.—r_._._.b..._._.

9NiNlVW38 NIEiOHd GNV MINlOV %

46

.omano to_:mo_ooobxo
to cozotflooto toto EEoEot £303
to 5.358 to 33:09.3 on» How...” manor.

m._.<.._.._Dm 22203.2(. “5 ZO_._.<mD._.<m szommm
Om. om cum 0% mm ON 9.. o

 

 

 

\- i- b IP b! [- LD b
if . umsi Ham 2. tsco< To 0
.. Hug—Ln— H._<m Z. Z_m._.0mn_ AKPOH I u
1 FUN
T r
.. no...
I loo
. 1. Tom
_.. j
l. p L b Lt HIOOP

 

 

 

ONINIVWBB NIBiOHd GNV MlALLOV %

.omoSEo uoobxm o_Emo_o_._oo
to 8533695 toto 055025.. £8.05

.5 3358 to 33:09.3 9:. 59m HEDGE

m._.<n._._2m 2220.226. “.0 ZO_._.<m_D._.<m Hzmommm
on 9v on ON 0.. o
L L L L

n — [b [D h n L
mm<Im m30m30< Z. >:>_._.o< ale 0

lb

 

47

 

mm<1m m30m30< z. Z_M._.omn_ .._<._.O._. I ..

 

 

00F

SNINIVWEH NIBlOl-Jd CINV MlALLOV %

48

603955050 22v. .6 coroca o mo .tom .yom
.6 Ron 5;, oscoona .38 82a :8 2: 5
mEEoEmt 33:8 to 33:09.3 9:. ”5n manor.

A52; .5 8(2me 3 53. m

Nolm F noflmm o

u . . J.
as}: 2 .255: moo unfit? 8&8 2.....me 01..
.. as}: onu 4.33.2: as mafia: «Sajmoéfi To _.

 

 

 

 

 

OO—

HSVHd .L'WS NI SNINIVWBH MINiOV %

49
respect to the methylcellulose concentration, and are indicated as the
percentage of activity remaining in the salt phase after partitioning
extracellular and periplasmic extract a-amylase with 30% of saturation
ammonium sulfate.

The specific activities of the extracellular a-amylase before
either purification step is around 0.13 units/ug. However, after
partitioning or precipitating with 30% or 50% saturated ammonium
sulfate, respectively, the specific activity is approximately 3.8 or 1.1
units/ug, respectively (Figures 3.5a and 3.6a). Similarly, the specific
activities of the periplasmic a-amylase before either purification step
is around 9 units/ug. After partitioning or precipitating with 30 or
50% saturated ammonium sulfate, respectively, the specific activity is
approximately 13.6 or 7.2, respectively (Figures 3.5b and 3.6b). The
specific activity of a-amylase after the purification step was
determined by assaying only the salt phase; therefore, the
concentrations in the polymer or precipitate phases had to be
extrapolated.

The results suggest that the purification (i e. the ratio of
specific activities) of extracellular a-amylase was nearly 30 fold by
partitioning, and nearly 10 fold by precipitation. These approximate
ratios indicate that partitioning extracellular a-amylase into a
methylcellulose phase with 30% saturated ammonium sulfate is about three
times more effective in purification than precipitating the a-amylase
with 50% saturated ammonium sulfate. The periplasmic extract a-amylase
had a specific activity 50 times larger than the extracellular amylase
prior to the partitioning and precipitation experiments. Possibly for

this reason, neither the partitioning nor the precipitation step

SO
resulted in any substantial degree of additional purification.
3.4.3 Nature of the Partitioning Process
Equilibrium

The equilibrium of a-amylase between the salt and methylcellulose
phases was investigated as a function of ammonium sulfate concentration
in solutions containing 0.05% w/v AAM and 1200 units/ml a-amylase. Two
experiments were carried out, one to test the equilibrium of a-amylase
partitioned into a polymer phase with 30% sat. ammonium sulfate and its
subsequent equilibration when the salt phase is replaced with solutions
of 0, 10, 20 and 30% sat. ammonium sulfate. The second experiment tests
the equilibrium of a-amylase, in solutions of 0, 10, 20 and 30% sat.
ammonium sulfate, when a polymer phase that contains zero amylase is
added. The polymer phase is formed by addition of 30% sat. ammonium
sulfate after the subsequent salt phase had been removed. Equilibrium
was assumed after 24 hours.

The percentage of initial activity remaining in the salt phase
after attaining equilibrium, as a function of the equilibrium salt
concentration, indicates that a-amylase is partitioned into the polymer
phase with the addition of salt (Figure 3.8). In addition, the a-
amylase resolubilizes out of the polymer phase with a reduction in salt
concentration. The partitioning equilibrium of a-amylase in the
methylcellulose/salt system is therefore only a function of ammonium
sulfate concentration.

Hydrophobicity

The interaction between a-amylase and methylcellulose may be

hydrophobic. It is already known that methylcellulose is slightly

hydrophobic due to the degree and kind of substitution of its hydroxyl

51

.cozotBom 333m EEcoEEo .6 c0305.:
0 mo 82.5533 ooEoto mo; 9:8:an coto
055080.. 3328 .6 omoucmotoo cc.— ”mfi. ”$50....

m._.<n_.5m 23.20—22.41 “.0 ZO_H<W_D._.<m Hzmommn.
mN JN ON 0% NW 0 e. 0

Pl P P P .— lxr P P h h I...) h
mwicfion. “.0 .50 0250: m< mm5S< “.0 23.xm350m ale 0

- mwzhom 2.2. 0250.2 m< mmj>2< no :3.mm350m ole -

 

 

 

 

WWHBWHOH .LV SNINIVWEH ALI/\IiOV %

52
26,27 . .
groups . To test the degree of hydrophobic interaction
corresponding to a-amylase, its adsorption, as a function of ammonium
sulfate concentration, to two known hydrophobic gels, octyl and phenyl
sepharose Cl 48, was tested. Batch type adsorption experiments were
performed separately after the sepharose gels were equilibrated with 50

mM Acetate, pH 6.0, containing 5 mM CaCl and the appropriate ammonium

2
sulfate concentration. The adsorption of extracellular a-amylase out of
the aqueous phase was determined by measuring the dissappearance of
starch reducing activity from that phase. The total protein remaining
in the aqueous phase was measured. After removal of the ammonium
sulfate, the amount of a-amylase desorbing into the aqueous phase from
the sepharose was measured also. The adsorption/desorption results
using octyl and phenyl sepharose-Cl 4B are shown in Figures 3.9a and
3.9b, respectively.

The results indicate that a~amylase is very strongly adsorbed to
hydrophobic gels, with more affinity towards phenyl than octyl
sepharose. The experiment demonstrated that a—amylase is a hydrophobic
enzyme that can be strongly adsorbed to a very hydrophobic polymer. A
strong affinity makes it difficult to recover the enzyme. However,
these data do not provide insight to accurately draw conclusions about
the nature of interaction between the a-amylase and methylcellulose in
the precense of ammonium sulfate. Any explaination for the selectivity
of a-amylase partitioning into methylcellulose requires a comparison of
the interaction of several enzymes with methylcellulose and also a
hydrophobic polymer, such as octyl sepharose (Chapter 4).

3.4.4 Methylcellulose Recovery

After purification, methylcellulose can be recovered and a-amylase

% ACTIVITY OR PROTEIN

53

H 25 TOTAL PROTEIN LEFT IN AQUEOUS PHASE
H X ACTIVITY DESORBED FROM GEL
o—o X ACTIVITY LEFT IN AQUEOUS PHASE

 

 

 

 

 

 

 

1 1 r

r ' r f r a 1 QFF-T
0 4 8 12 16 20 24- 28 32 36 40

PERCENT SATURATION OF AMMONIUM SULFATE

FIGURE 3. 90: The adsorbtion desorbtion of
amylose and adsorbtion 0 protein using
octyl sepharose-CI 4B.

Z ACTNITY OR PROTEIN

54

H 3 TOTAL PROTEIN LEFT IN AQUEOUS PHASE
H 3 ACTIVITY DESORBED FROM GEL
0—0 3 ACTIVITY LEFT IN AQUEOUS PHASE

100 .,.

 

 

 

40-I . -

20- ..

.I .

 

 

 

o the

o 4 81216 20 24 28 3'2'36 40
PERCENT SATURATION OF AMMONIUM SULFATE

FIGURE 3.9b: The adsorbtion desogbtion of
amylase and adsorthon o proteIn usmg
phenyl sepharose-CI 4B.

55
subsequently freed from most of the polymer. The recovery and liberation
of the polymer and enzyme, respectively, is due to the reversibility of
partitioning and the unique partioning profiles of methylcellulose and
a-amylase upon addition of ammonium sulfate (described below).

The findings (Figure 3.10) suggests that extracellular a-amylase
would require a purification step occuring at 30-35% sat. of ammonium
sulfate, as over 94% of the enzyme was partitioned into the polymer
phase there, while approximately 97% of the contaminating proteins were
left in the salt phase. The contaminants could then be removed by
disposing of the salt phase. In addition, the a-amylase can be
reversibly resolubilized from the methylcellulose, as Figure 3.10
indicates, at an ammonium sulfate saturation of 15% and after
equilibrium between amylase and methylcellulose has been reached. Under
these conditions none of the a-amylase is partitioned into the
methylcellulose phase while over 80% of the methylcellulose is
partitioned out of the salt phase. After disposing of the
methylcellulose precipitated at 15% salt saturation, the majority of the
remaining methylcellulose can be precipitated by increasing the ammonium
sulfate saturation to 20%. However, removing methylcellulose at an
ammonium sulfate saturation of 20% results in a loss of a-amylase.

3.5 Conclusions

Recombinant DNA technology made it possible to produce a robust,
thermostable a-amylase in E. 921;, and to direct its excretion
extracellularlyzg. The excretion of a-amylase extracellularly
simplifies its downstream refining steps because it contains fewer
contaminating proteins. An aqueous two phase partition step was studied

to determine its potential for purifying a-amylase.

56

.omflxEo 3.2.3835 Lo mEcoEtoo
rota omoc .
.33 .3338 to $359.8 9: “Someone...

tom 05 E @5535. .2me ._o 5395

WEE-Sm 3.3—202‘: “.0 ZO_._.<mD._.<m hzmomun.

VMNnonmNoNVNNNONLWPmP¢pNFop m w .v N

be-PbLPLbprrlP.

o

 

wgm Him 2. ESE: Ola
mm<In_ Him 2. onmm I
um<Im ._.._<m 2. In; T!

 

 

 

.L—.—.-L—Ph.-n—._._ . -L—p

 

 

ONINIVWBH 'OBHVO HO NIBiOEId 'AJJALLOV %

57
The aqueous two-phase partioning step utilized a unique system
consisting of methylcellulose and ammonium sulfate. Only one polymer
phase was necessary for selective partitioning, an improvement over
other aqueous two-phase partitioning techniques described using PEG and

dextran2’35.

Methylcellulose, designated MC or HPMC, were found to be
equally effective especially at a molecular weight of 15 kilodaltons.
Nearly complete partioning of a-amylase occurs at ammonium sulfate
saturations of around 30-35% and methylcellulose concentrations of at
least 1 g methylcellulose per g a-amylase. There is no pH optimum for
partitioning, however, optimum pH for a-amylase activity is between 5.0
and 6.0. Similarly, there is no optimum temperature for partioning,
however, ambient would be the most economical.

'Preliminary studies indicated that a single aqueous two-phase
partitioning step at an ammonium sulfate saturation of 30%, purified
(according to specific activity) a fairly crude extracellular solution
of a-amylase by 30 fold; 94% of the enzyme was partitioned out of the
salt phase. Preliminary studies also indicated that approximately 80%
of the methylcellulose can be separated from the a-amylase and nearly
all of the enzyme recovered by resolubilizing the a-amylase at an
ammonium sulfate saturation of 15%.

A scaled up purification process for the isolation of a-amylase was
tested with approximately one liter of fermentation broth. The process
consisted of insolubles removal by centrifugation, concentration by
ultrafiltration, soluble contaminant removal by thermal precipitation,
and aqueous two-phase partitioning steps for purification and

methylcellulose recovery (Figure 3.11). The results are indicated in

Table 3.2, and demonstrate a very good purification. However, the

Centrifugation -------- > Insolubles

Thermal Precipitation ----> Denatured
------------------------- Contaminants

First Partition
(add 0.05% KISM and
35% ammonium sulfate)
Centrifugation ---~> Salt Phase
------------------------- Contaminants
Resolubilize Polymer
Phase in buffer

Second Partition
(add 35% amm. sulf.)

Centrifugation ----> Salt Phase
------------------------ Contaminants

Third Partition
(add 35% amm. sulf.)

Centrifugation ----- > Salt Phase
------------------------ Contaminants

Resolubilize Polymer
Phase in buffer

Fourth Partition
(add 20% amm. sulf )

Centrifugation ----- > Polymer Phase

oooooooooooooooooooooooo

a-amylasc

Figure 3.11. The downstream purification of u-amyiase.

59

Table 3.2: The downstream refining of a-amylase.

Activ.

(units)

Protein Carbohyd.

(ug)

(ug)

Process Volume
(m1)
Centrifugation 970
Ultrafiltration 200
Heat Precipit. 200
First Partition 400
Second Partition 200
Third Partition 100
MC Precipitation 50

3112

3034

2318

1960

1596

1084

510

2,648,000
1,380,000
1,255,000
18,000
2,000

300

52.0

35.2

30.6

19.2

16.2

11.7

1.

2.

147

1320

8800

9

1

97.5

74.5

63.0

51.3

34.8

16.4

60
process did not excell at recovery of a-amylase, around 16%. If
small quantities of methylcellulose could be tolerated, the recovery of amylase
could be increased to around 70%, with approximately 20 ug protein/ml
remaining, by limiting the process to removal of only the insolubles
followed by two partioning steps. Each partitioning step
results in a loss of 10-15% of the a-amylase.

The concentration and purification success of a-amylase upon
aqueous two-phase partitioning with methylcellulose and ammonium sulfate
provides what appears to be a successful purification scheme for a-
amylase. Moreover, methylcellulose has a low cost and a good standing
with the FDA who have often approved its use in the food industry26
3.5 Literature Cited
1. Kroner, K. H., H. Hustedt, S. Granda, and M.-R. Kula. 1978. Technical
aspects of separation using aqueous two-phase systems in enzyme

isolation processes. Biotechnology and Bioengineering. 20:1967-1988.

2. Hustedt, H., K. H. Kroner, U. Menge and M.-R. Kula. 1985. Protein
recovery using two-phase systems. Trends in Biotechnology. 3(6):139-144.

3. Johansson G. 1987. Dye-ligand aqueous two-phase systems. Pages 101-
124. In: Y. D. Clonis, T. Atkinson, C. J. Bruton, and C. R. Lowe, edts.

Reggtivg dyes in ngtgig and enzyme technology. Stockton Press, New
York, New York.

4. Mattiasson, B. 1983. Applications of aqueous two-phase systems in
biotechnology. Trends in Biotechnoloyg. 1(1):l6-20.

5. Mattiasson, B, and R. Kaul. 1986. Use of aqueous two-phase systems
for recovery and purification in biotechnology. American Chemical
Society. 314:78-92.

6. Bell, D. J., M. Hoare, and P. Dunnill. 1983. The formation of protien
precipitates and thier centrifugal recovery. Advances in Biochemical
Engineering. 26:1-72.

7. Oriel, P. 1987. Personal communication. Dept. of Microbiology and
Public Health, Michigan State University, E. Lansing, Michigan.

8. Kennedy, J. F. 1978. Chemical synthesis and modification of
oligosaccharides, polysaccharides, glycoproteins, enzymes and
glycolipids. Carbohy. Chem. 10:427-497.

61

9. Butler, L. G. 1975. Enzyme immobilization by adsorption on
hydrophobic derivatives of cellulose and other hydrophobic materials.
Arch. Biochem. Biophys. 171:645-650.

10. Hjerten, S., J. Rosengren, and S. Pahlman. 1974. Hydrophobic
interaction chromatography: The synthesis and the use of some alkyl and
aryl derivatives of agarose. J. Chromatog. 101:281-288.

11. Schwacha. A. 1983. Affinity chromatography: Principles and methods.
Pages 12-110. WW. Piscataway. New

Jersey.

12. Vretblad, P. 1974. Immobilization of ligands for biospecific
affinity chromatography via thier hydroxyl groups: The
cyclohexylamylose-fl-amylase system. FEBS Letters. 47(1) 86-89.

13. Cuatrecassas, P. 1972. Affinity chromatography of macromolecules.
Advan. Enzymol. 36:29-89.

14. Horejsi, V., and J. Kocourek. 1973. Studies on Phytohemagglutinins.
XII. o-glycosyl polyacroamide gels for affinity chromatography of
phytohemagglutinins. Biochim. Biophys. Acta. 297:346-351.

15. Matsumoto, Isamu, and Toshiaki Osawa. 1972. The specific
purification of various carbohydrates binding hemaglutinins. Biochemical
and Biophysical Research Communications. 46(5):1810-1815.

l6. Kristiansen, T., L. Sundberg and J. Porath. 1969. Studies on blood
group substances II. coupling of blood group substance A to hydroxyl-
containing matrices, including animoethyl cellulose and agarose.
Biochim. Biophys. Acta. 184:93-98.

17. Gordon, J., et a1. 1972. Purification of soybean agglutinin by
affinity chomatography on sepharose derivatives. FEBS Letters.
24(2):193-196. '

18. Barker, R., et al. 1972. Agarose derivatives of uridine diphosphate
and N-acetylglucosamine for the purification of a galactosyltransferase.
J. Biol. Chem. 247(22):7135-7147.

l9. Junowicz E., and J. E. Paris. 1973. Affinity chomatography by
enzyme-subsrate interaction. purification of some rat liver
glycosidases. Biochim. BiOphys Acta. 231:234-245.

20. Rafestin, M. E., et a1. 1974. Purification of N-acetyl D-
glucosamine-binding proteins by affinity chromatography. FEBS Letters.
40(1):62-66.

21. Hayes, C. E., and I. J. Goldstein. 1974. An a-D-galactoseyl-binding

lectin from ngdgiraea simplicifolia seeds: Isolation by affinity
chromatography and characterization. J. Biol. Chem. 249(6):1904-1914.

62

22. Ellingboe, J., et a1. 1970. Liquid-gel chromatography on lipophilic-
hydrophobic sephadex derivatives. J. Lipid Res. 11:266-273.

23. Rosengren, I., et a1. 1975. Hydrophobic interaction chromatography
on non-charged sepharose derivatives. binding of a model protein,
related to ionic strength, hydrophobicity of the substituent, and degree
of substitution (determined by NMR). Biochim. Biophys. Acta. 412:51-61.

24. Keillich, G., et a1. 1972. Optical rotary dispersion and circular
dichroism of benzoyl polysaccharides. Biopolymers. 11:1997-2013.

25. P.-A. Albertsson. 1958. Nature. 182:702.

26. Greminger, G. K. Jr., and K. L. Krumel. 1980. Alkyl and
hydroxyalkylalkylcellulose. Pages 3-1 to 3-25 In: R. L. Davidson, edt.

Hagdbggk 9f Wage; sglgblg gyms agd resigs. McGraw Hill, New York.

27. Neely, W. B. 1963. Solution properties of polysaccharides. IV.
Molecular weight and aggregate formation in methylcelluse solutions, J.
Polymer. Sci. 1(1):311-320.

28. Chemical Marketing Reporter. August 1. 1988. Schnell Publishing
Company, New York, New York.

29. Oriel, P., and A. Schwacha. 1988. Growth on starch and extracellular

production of thermostable amylase by E. cgli. Enzyme Microb.
Technol.,10:42-46.

30. Nossal, N. G., and L. A. Heppel. 1966. The release of enzymes by
osmotic shock from E. 9211 in exponential phase. J. Biol. Chem.
241(13):3055-3062.

31. Davis, R. V., Botstein, D., and J. R. Roth. 1980. Advanced Bacterial
figgggigg. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York,
New York. 201 pp.

32. Bernfeld, P. 1955. Amylases, alpha and beta. Pages 149-154. In: S. O.
Colowick and N. 0. Kaplan, eds. Method n enz o o . vol 1. Academic
Press, New York, New York.

33. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randan, R. J.,
1951. J. Biol. Chem., 193:265.

34. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F.
Smith. Colorimetric method for determination of sugars and related
substances. Analyt. Chem. 28(3):350—356.

35. Kroner, K. H., H. Hustedt, and M.-R. Kula. 1982. Evaluation of crude
dextran as phase-forming polymer for the extraction of enzymes in
aqueous two-phase systems in large scale. Biotechnology and

Bioengineering. 24:1015~1045.

CHAPTER 4

EFFECT OF A GENETICALLY ATTACHED SECRETION SEQUENCE ON

PARTITIONING ALKALINE PHOSPHATASE FROM E. COLI

4.1 ABSTRACT

Alkaline phosphatase was previously fused genetically to a sequence
of a-amylase. Purification of the fusion protein via aqueous two-phase
partitioning (a technique successful for the purification of a-amylase)
was investigated and some enhancement over regular precipitation was
found. Investigation of the hydrophobicity, precipitation, and
partitioning of a-amylase, alkaline phosphatase, and the
amylase/phosphatase fusion protein indicate that partitioning of a-
amylase into methylcellulose is not governed by hydrophobic
interactions. Further partitioning experiments with other starch
binding enzymes indicates a specific a-amylase/ methylcellulose
interaction possibly similar to interaction between amylase and starch.
4.2 INTRODUCTION

Extracellular excretion of an enzyme simplifies its subsequent
collection and purification by eliminating traditional problems of
recovery from a heterogeneous mixture of periplasmic proteins or
cellular debris. There are very few enzymes known to be naturally
excreted by E. gglil’z. This lack of extracellular enzymes in gram-
negative bacteria is thought to be due to transport limitations imposed
by the lack of a general outer membrane secretion mechanismz. However,
foreign gene products are known to be secreted, at least if they carry

bacterial signal peptides whose function is similar for many

63

64

organisms These signal peptides function in transporting the
protein into the periplasmic space3 in E. ggli and are removed during
transport. The excretion mechanisms for transporting proteins from the
periplasmic space to the culture medium are unknown3; however, the
structure of the protein is known to play an important roleh.
Techniques exist for coupling protein synthesis to protein excretionz.

Recently, Oriel and Alexanders discovered a peptide sequence in a-
amylase transformed from figgillgs ggggrgthermophilus into E. 991; that
enabled approximately 28% of this enzyme to be excreted extracellularly.
The enzyme is excreted throughout the growth phase; however, the
excretion is not accompanied by general periplasmic rupturings. The
structural regions which confer a-amylase its excretion properties were
investigated to determine if fusions containing these regions could
confer extracellular release to other proteins that are not normally
excreted3. One such enzyme of commercial interest is alkaline
phosphatase from E. 2211, which is naturally secreted across the inner
membrane to the periplasmic space, but not secreted extracellularly '
With this in mind, they took N-terminal regions containing approximately
10 and 65 percent of the amylase gene and fused them in frame to an
alkaline phosphatase gene in which the upstream control and leader
sequence had been removedA. They found alkaline phosphatase activity in
both gene fusions produced at the predicted molecular weights.

Both fusion products were produced under culture conditions similar
to those that generated the 25 percent extracellular excretion of a-
amylaseh. Oriel and Alexander found that the amount of fusion protein

excreted under these conditions to be 17% and 7% of the larger and

smaller alkaline phosphatase fusions, respectivelya. These results

65

indicated that extracellular release of normally periplasmic proteins
can be conferred by fusion to sequences of extracellular proteins, with
longer sequences of the amylase being more effectivea. The successful
excretion is thought to be a result of the a-amylase sequence conferring
desirable conformational changes on the alkaline phosphatase domaina.

With a similar enhancement of purification in mind, recovery studies
of the a-amylase and alkaline phosphatase gene fusion were conducted to
determine if the fusion protein had separation properties which could be
exploited commercially. This is of importance because a simple recovery
and partial purification process has been demonstrated for a-amylase
(Chapter 3) and a similar technique could be applicable to the fusion
protein if the fused portion of amylase confers enough structural
similarity.

In addition, experiments were designed to determine the nature of
the interaction between a-amylase and methylcellulose. The adsorption
to octyl sepharose, precipitation, and partitioning into methylcellulose
of a-amylase, alkaline phosphatase, and the phosphatase/amylase fusion
protein were determined upon addition of ammonium sulfate. These
results were used to indicate if the a-amylase portion of the fusion
protein conferred any hydrophobicity to the fusion protein. They also
can give an indication of the nature of the interaction between a-
amylase and methylcellulose in the precense of ammonium sulfate.

4.3 Experimental
Materials

The methylcellulose and hydroxypropyl methylcellulose were donated

by Dow Chemical and have brand names A4M, K4M, K15M, and K100M

designating the type of cellulose derivative (A for 26-33 percent

66

methoxy substitution and K for 19-24 percent methoxy and 7-12 percent
hydroxypropoxyl substitution) and the approximate molecular weight (4,
15, and 100 corresponding to 85, 15, and 250 kilodaltons, respectively).
The octyl sepharose-Cl 4B was obtained from Sigma. Alkaline phosphatase
from E. £911 was obtained from Sigma (type III-N). The alkaline
phosphatase/a-amylase fusion protein and the a-amylase were obtained as
described below. All other chemicals were of reagent grade.
Amylase/phosphatase fusion preparation

The fusion of a 10% portion of a-amylase from B. stearothermophilus
to alkaline phosphatase from E. 9911 was performed by Dr. P. Alexander“,
and designated EC 243. EC 243 grown on agar plates (LB medium6) was
obtained from Dr. P. Oriel, Department of Microbiology and Public
Health, Michigan State University, E. Lansing, Michigan. EC 243 cells
were transferred to 5 ml LB medium and grown in a 37°C shaker incubator,
overnight. The innoculum was transferred to l 1 of LB medium contained
in a 2800 m1 large bottom Erlenmeyer flask and grown in a 37°C shaker
incubator, overnight. The cells were pelleted and both the supernatant
and cell pellet collected. The cell pellet was resuspended in
approximately 50 ml of 50 mM tris-Cl buffer, pH 7.5. The method of
Nossal and Heppel7 was used to osmotically shock the cells and recover
the periplasmic fusion phosphatase. The extracellular and periplasmic
extract alkaline phophatase solutions were made up to 1000 ml and 100
ml, respectively, to give final concentrations of 50 mM tris-Cl and
0.01 M MgSOa. These solutions were heated to 75°C for 15 minutes and
stored at 4°C overnight to precipitate contaminants by denaturation as
described by Schwartz and Lipmanns. The precipitates were pelleted and

discarded. The extracellular phosphatase solution was further

67

concentrated by ultrafiltration (10,000 molecular weight cut off) from
1000 ml to 15 ml. After refrigeration overnight, the concentrated
extracellular solution was cloudy with precipitate. The precipitate was
removed by centrifugation.
Amylase preparation

The a-amylase was prepared from EC 147 cultures, provided by Dr. P.
Oriel (Michigan State University, E. Lansing, MI 48824), by the same
method described for the phosphatase/amylase fusion preparation, with
slight modifications. The modifications consist of: using 50 mM acetate
buffer, pH 6.0, including 5 mM Ca++, instead of tris buffer; and, heat
treating at 80°C for 45 min in the presence of 15 mM Ca++ instead of at
75°C for 15 min. I
Methylcellulose preparation

The methylcellulose is prepared in 50 mM tris-Cl buffer, pH 7.5, or
in 50 mM acetate buffer, pH 6.0, including 5 mM Ca++, depending on
whether partitioning alkaline phosphatase and the phosphatase/amylase
fusion protein or a-amylase, respectfully. The methylcellulose powder
is dispersed and wetted in the appropriate buffer solution by the slow
addition of the powder to the hot buffer, around 80-90°C. The wetted
methylcellulose is then poured into an equal volume of cold buffer,
around 4°C, and well mixed. Methylcellulose solutions thus formed9 were
stored at 4°C between experiments for several months with no apparent
qualitative changes.
Partitioning

The term partitioning is used here to indicate the selective
localization of protein into either of the two phases formed. The exact

physical nature of the two phases is somewhat unclear. With ammonium

68
sulfate—methylcellulose partitioning, the polymer phase can appear as
liquid droplets under a microscope while resembling a precipitate in
other characteristics.

Experiments involving homogeneously pure alkaline phosphatase were
performed on a small scale using 1.5 ml Eppendorf tubes. However,
experiments involving a-amylase or the phosphatase/amylase fusion
protein were performed on a scale of several ml.

Equal volumes of methylcellulose and enzyme solution, prepared as
described above, were added in various concentrations and mixed
vertically at room temperature. After approximately 30 minutes with
occasional mixing, saturated solutions of ammonium sulfate were added to
the samples, then immediately mixed to produce the desired salt
concentration. The addition of ammonium sulfate to the enzyme and
dilute (0.5-0.0001%) methylcellulose solution created a precipitate like
polymer phase and a salt phase. After an additional 30 minutes, while
mixing occasionally, the precipitate was pelleted by centrifuging the
Eppendorf tubes at 13000 rpm for 20 minutes. The resulting supernatant
and/or pellet were then tested, after appropriate dilutions, for total
enzymatic activity and when possible total protein. A measure of total
protein was not possible on every sample because pure alkaline
phosphatase along with the dilutions used in most of the experiments
reduced the protein levels to less than that assayable by the Lowry et
a1. procedure10
Assays

Alkaline phosphatase was assayed as described by Torriannill.
Equal 1 ml volumes of substrate, 4.0 mM p-nitrophenol phosphate (pNPP)

in 1 M Tris-Cl (pH 8.0), and enzyme solution were added and allowed to

69
react at 37°C for 10 minutes. The reaction was stopped by the addition
of 0.2 ml 20% KZHPoa. Adsorbance was measured at 410 nm in solutions
diluted to give adsorbance changes of less than 0.1/minute. The molar
extinction coefficient of p-NP at 410 nm is 16.2 E3, and one unit of
phosphatase was defined as l nmole p-nitrophenol/min/ml.

Total protein was measured using the method of Lowry et. al10 using
bovine serum albumin as the protein standard. The contribution of the
methylcellulose to the total protein measurement was negligible.

4.4 Results and Discussion
4.4.1 Comparison of partitioning of alkaline phosphatase,
amylase, and the phosphatase/amylase fusion protein

It has previously been shown“ that the alkaline phosphatase/a-
amylase fusion protein was conferred the a-amylase property of
extracellular excretion. If the fusion protein was also conferred
isolation properties similar to those of a-amylase, it could be isolated
using a previously described and effective technique involving aqueous
two-phase partitioning (Chapter 3). The experiments described in this
section were designed to test the partitioning of the phosphatase/fusion
protein and determine if either fusion portions, i.e. the phosphatase or
the amylase, conferred observable selectivity to the fusion proteins
partitioning. The concentration of methylcellulose, ammonium sulfate
saturation, initial enzyme activity, and total protein used in these
experiments are listed in Table 4.1.

Separate solutions of alkaline phosphatase, a-amylase, and
phosphatase/amylase fusion protein were used to test for selectivity of
partitioning as a function of both ammonium sulfate saturation (Figure

4.1) and methylcellulose concentration (Figure 4.2). The results show

70

Table 4.1. The initial enzyme activity and total protein of
solutions prepared for ammonium sulfate precipitation and
methylcellulose/salt partitioning (partitioning as a
function of (i) ammonium sulfate saturation and as a
function of (ii) methylcellulose concentration).

[KlSM] Salt Activity Tot. Prot.

Solution (% w/v) (% Sat.) (units/ml) (ug/ml)
va-Amylase

Precipitation 0.00 var.a 122 820

Partitioning (i) 0.05 var.b 52 440

Partitioning (ii) var.c 30 56 231

Alkaline Phosphatase

Precipitation 0.00 var. 7.06E-5 2.9

Partitioning (i) 0.05 var.a 7.06E-5 2.9

Partitioning (ii) var.c 30 1.54E-5 2.5
Fusion Protein

Precipitation o.oo var.a 4.65E-4 55

Partitioning (i) 0.05 var.a 4.61E-4 192

Partitioning (ii) var.C 30 3.13E-4 117

aAmmonium sulfate saturations varied from 0 to 70%.
bAmmonium sulfate saturations varied from 0 to 40%.

bK15M concentration varied from 0 to 0.03% w/w.

z ENZYMATIC ACTIVITY REMAINING

71

H FUSION PROTEIN: 4.658—4 unlta/ml. 55 ug pmtoIn/ml
o—O o-AMYLASE: 52.2 units/ml. 439 ug protein/ml
o—o PHOSPHATASE: 9.145—5 units/ml. 2.88 no protoIn/ml

 

 

 

 

 

 

1 00 ' ‘ I ——U I I r WI 1’ I i F -
'1 N H
80d .4
1 A a
60— -I
1
401 .1
J u
20- ..
.1
o . r . , e ———— 3 3 - .
0 1 O 20 30 4O 50 60 70

PERCENT SATURATION OF AMMONIUM SULFATE

FIGURE 4.1: .The percentageof phosphatase,
amylase .or fusIon pratem ac IVIty remaInIng after
partItIonIng upon addItIon of ammomum sulfate.

7. ENZYMATIC ACTTVITY REMAINING

72

a—m FUSION PROTEIN: 3.13E-4 units/ml. 117 ug prot/ml
H o-AMYLASE: 56 unIts/ml. 231 ug protoIn/ml
o—o PHOSPHATASE: 1.54E-5 unIta/ml. 2.5 ug protein/ml

 

 

 

 

 

 

 

 

 

100 4
+ .4
30- j
. ao- -
404 -
20- ._
- 1

o l— T

o 1E-02 25-02 35—02

7: K15M (g K15M/1OO ml WATER)

FIGURE 4.2: The perceptageof phosphatase,
amylase or fuqun proteIn actIVIty remaInIng In
the salt phase WIth respect to % K15M present.

73
the percentage of each of the three enzymatic activities remaining in
the salt phase after partitioning with respect to either ammonium
sulfate (Figure 4.1) or methylcellulose added (Figure 4.2).

In addition, the effect of ammonium sulfate saturation on
precipitation of the three proteins was tested to ensure that
precipitation and partitioning were not confused. The results show the
percentage of enzymatic activity remaining in solution after either
partitioning or precipitation of solutions of alkaline phosphatase
(Figure 4.3a), a-amylase (Figure 4.3b), or phosphatase/amylase fusion
protein (Figure 4.3a) with respect to ammonium sulfate added.

A comparison of partitioning and precipitation of each enzyme, upon
addition of ammonium sulfate, indicates that the exclusion of a-amylase
(Figure 4.3b) from the salt phase is greatly enhanced in the presence of
methylcellulose. Whereas, the exlusion of alkaline phosphatase (Figure
4.3a) or the fusion protein (4.3a) from the salt phase is not as
affected by the presence of methylcellulose. However, the fusion
protein partitions before it precipitates, similar to a-amylase (Figures
4.3b and 4.3a); and, the alkaline phosphatase precipitates slightly
before it partitions (Figures 4.3a and 4.3a). In addition, comparing
each enzymes partitioning (Figure 4.1) indicates that a-amylase is the
first to be partitioned out of the salt phase as ammonium sulfate is
added, followed by alkaline phosphatase and lastly the
phosphatase/amylase fusion protein. Similarly, at 30% of saturation
ammonium sulfate more a-amylase, by percentage, is partitioned out of
solution, with increasing methylcellulose concentration, than alkaline
phosphatase or the fusion protein (Figure 4.2). Again, the percentage

of phosphatase partitioning out of the salt phase is greater than that

Z ENZYMATIC ACTIVITY REMAINING

74

o—o PARTITIONING: 7.065-5 units/ml. 2.9 ug protein/ml
o—o PRECIPITATION: 7.065-5 units/ml. 2.9 ug protein/ml

 

100

 

0 fT T I ' r ' r '
0 10 20 ~ 30 4O 50 60 70

% SATURATION OF' AMMONIUM SULFATE

FIGURE 4.30: The 7: of phosphatase .activity
remaining in the salt phase” as a functan of salt
added, for either partItIonIng or ~preCIpItatIon.

 

 

 

 

Z ENZYMATIC ACTIVITY REMAINING

75

H PARTITIONING: 52.2 units/ml. 439 ug protein/ml
o—o PRECIPITATION: 121.8 units/ml. 824 ug protein/ml

 

100 T I I I— r —T r i I r 1 -

30- I —

Baa-I cu

 

 

 

 

o 10 db 30 40 so so 76
Z SATURATION OF AMMONIUM SULFATE

FIGURE 4.3!}: The percentage of a-amylgse activity
remaInIng In the salt phase” as a functan of salt
added, for either partItIonIng or' preCIpItatIon.

Z ENZYMATIC ACTIVITY REMAINING

76

H PARTITIONING: 4.85E-4 units/ml. 85 ug protein/ml
o—o PRECIPITATION: 7.81E-4 units/ml. 181.8 ug prot/ml

 

 

 

 

 

  

a T ‘I ‘I’ I~_ '7 T

a 1'0 2'6 35 40 5'0 so 70
% SATURATION OF AMMONIUM SULFATE -

FIGURE 4.3a: The percentage of phosphatase
actIVIty, due to thefusIon protein, remaInIn In
the salt phase wnth respect to-salt adde .

     

77
for the fusion protein (Figure 4.2).

These results indicate that the a-amylase portion (approximately
10% of the a-amylase) of the fusion protein conferred little of the
partitioning selectivity unique to a-amylase. These results seem to
indicate that the amylase portion of the fusion protein works against
partitioning into methylcellulose (Figure 4.2) and also against
precipitation (Figure 4.4). The results indicate that the partitioning
specificity of a-amylase into methylcellulose is probably not due to the
simple hydrophobic interactions assumed. It should be mentioned that
the quantitative comparisons of the results should be viewed with
caution due to the different sources and purities of the enzymes.

4.4.2 Comparison of hydrophobicity of alkaline phosphatase,
amylase, and the phosphatase/amylase fusion protein

Prior to the experiments just described, the nature of interaction
between a—amylase and methylcellulose was assumed to be hydrophobic in
nature (Chapter 3). Comparing the partitioning into methylcellulose of
alkaline phosphatase, a-amylase and the phosphatase/amylase fusion
protein upon addition of ammonium sulfate, however, indicated that the
selectivity for amylase partitioning was much more complex than the
simple hydrophobic interaction previously assumed.

An experiment was designed to determine quantitatively the degree
of hydrophobicity, if any, that was conferred to the fusion protein by
its amylase fusion portion. The resulting experiment tested the
hydrophobicity of the three enzymes by comparing their enzymatic
activity left in the salt phase after adsorbtion to octyl sepharose,
with respect to ammonium sulfate added (Figure 4.5). Adsorbtion to

octyl sepharose can be used to test for hydrophobicity because it is

Z ENZYMATIC ACTMTY REMAINING

 

78

H FUSION PROTEIN: 7.61E—4 units/ml. 192 ug prot/ml
H o-AMYLASE: 122 units/ml. 824 ug protein/ml
o—o PHOSPIMTASE: 7.08E—5 units/ml. 2.9 ug protein/ml

 

 

 

o r r I fl I I I I— T

0 1'0 20 30 40 so so 7'0

 

PERCENT SATURATION OF AMMONIUM SULFATE

FIGURE 4.4: .The percentage .of phosphatase,
amyIase 9r fuswn protein ac IVIty rernaInIng after
preprItatIon upon addItIon of ammomum sulfate.

Z ENZYMATIC ACTIVITY NOT ADSORBED

79

H FUSION PROTEIN: 8.88E—4 units/ml. 843 ug prot./ml
H PHOSPHATASE: 3.087E-4 units/ml. 2.8 ug protein/ml
o—o a-AHYLASE: 180 units/ml. 877 ug protein/ml

 

I ' T r l

Ioo~ ' ' 1

 

80- ' d

 

 

 

o ' II) 20 so
73 SATURATION OF AMMONIUM SULFATE

FIGURE 4.5: The adsorbtion of the fusion protein.
amylase or phosphatase to octyl sepharose as a
function of the salt added.

80
known to interact hydrophobically with proteinslz. Also, the addition
of ammonium sulfate increases the van der Waals attraction between
proteins and nonpolar-hydrocarbon-coated adsorbentslB, i.e. the
hydrophobic effect. Proteins with a relatively large proportion of
nonpolar molecules exposed are attracted to hydrophobic adsorbents more
than proteins with relatively few nonpolar molecules exposed13.

Because the hydrophobic effect is increased with addition of
ammonium sulfate, and the electrostatic effect decreasedla, the relative
hydrophobicity of the enzymes is best indicated only after ammonium
sulfate was added. Without this assumption, the results (Figure 4.5)
would not give a clear indication if any hydrophobic interactive nature
was conferred upon the fusion protein (Figure 4.5). At ammonium sulfate
saturations of 108 and above, the results indicate that a-amylase has
more hydrophobic interaction with octyl sepharose than the other two
proteins, as assumed. The results also indicate that the fusion protein
is more hydrophobic in its interaction than alkaline phosphatase. Thus,
the fusion protein was conferred considerable hydrophobic nature from
its a-amylase portion.

4.5 Summary

Recent developments in recombinant-DNA technologies have made it
possible to direct the extracellular excretion of proteins produced in
E, ggliz, thus enhancing the proteins subsequent purification by
limiting contaminants. One such extracellular, recombinant-DNA
transformed protein is an alkaline phosphatase fusion to a portion of a-
amylase (10% of the amylase)“. Because a successful technique for
purifying a-amylase had already been developed, based upon aqueous two

phase partitioning into methylcellulose from a salt phase (Chapter 3), a

81
similar technique could be applicable to the fusion protein if the fused
portion of amylase confered enough structural similarity.

The results indicate that the phosphatase/amylase fusion protein
has been confered an increased hydrophobic nature but not a similar
selectivity for partitioning into methylcellulose. Therefore, the
conclusion can be made that it is not hydrophobic interactions that
govern the partitioning of a-amylase into methylcellulose. The highly
selective partitioning of a-amylase indicates that there must be a
specific interaction between the methylcellulose and the enzyme.
Somehow, possibly through an allosteric or active site, the a-amylase
may recognize some portion of the methylcellulose structure, while
existing in its precipitated state, as similar to its substrate. This
assertion must be confirmed by subsequent experiments.

4.6 Literature Cited

1. Holland, H. B., N. Mackman, and J. M. Nicaud. 1986. Biotechnology.
4:427-431.

2. Oliver, D. 1985. Protein secretion in E. coli. Ann. Rev. Microbiol.
39:615-648.

 

3. Miyake, et a1. 1985. Secretion of human interferon-a induced by using
secretion vectors containing a promoter and signal sequence of alkaline
phosphatase gene on Escherichia coli. J. Biochem. 97:1429-1436.

4. Alexander, P., and P. Oriel. 1988. Excretion of amylase/phosphatase
fusion proteins by 5. £211. Prepublication draft.

5. Oriel, P. and A. Schwacha. 1988. Growth on starch and extracellular

production of thermostable amylase by E. coli. Enzyme Microb.
Technol.,10:42-46.

6. Davis, R. V., Botstein, D., and Roth, J. R. 1980. Advanced Bacterial
genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York,
New York. 201 pp.

7. Nossal, N. G., and L. A. Heppel. 1966. The release of enzymes by
osmotic shock from Escherichia coli in exponential phase. J. Biolog.
Chem. 241(13):3055-3062.

8. Schwartz, J. H., and F. Lipmann. 1961. Phosphate incorporation into

82
alkaline phosphatase of E. coli. Biochemistry. 47:1996-2005.

9. Greminger, G. K. Jr., and K. L. Krumel. 1980. Alkyl and
hydroxyalkylalkylcellulose. Pages 3-1 to 3-25 In: R. L. Davidson, edt.

Handbook of Water soluble gums and resins. McGraw Hill, New York.

10. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randan. 1951.
J. Biol. Chem. 193:265.

11. Torriani, A. 1960. Repression of phosphatase by inorganic phosphate.
Biochim. Biophys. Acta. 38:460-469.

12. Izui, K., N. FUJita, and H. Katsuki. 1982. Phosphoenolpyruvate
carboxylase of 3, £911 hydrophobic chromatography using specific
elution with allosteric inhibitor. J. Biochem. 92:423-432.

13. Srinivasan, R., and E. Ruckenstein. 1980. Role of physical forces in
hydrophobic interaction chromatography. Separation and Purification
Methods. 9(2):267-370.

14. Bell, D. J., M. Hoare and P. Dunnill. 1983. The formation of protien
precipitates and thier centrifugal recovery. Advances in Biochemical
Engineering. 26:1-72.

CHAPTER 5

SUMMARY

5.1 Conclusions

The purpose of this research was to evaluate the concentration

and purification of a-amylase from E, coli using a methylcellulose/salt,

two-phase partitioning process. The following objectives were achieved

in this study:

1.

The partitioning of a-amylase in the methylcellulose/salt aqueous
two-phase system was optimized with respect to the type, molecular
weight, and concentration of cellulose derivative, the type and
concentration of salts added, the solution pH, and the temperature

during extraction.

. The partitioning of a-amylase was.reversed upon decreasing the

salt concentration, and a large percentage of the methylcellulose

recovered from the purified enzyme.

. A process was demonstrated for recovering a-amylase from crude broth

using methylcellulose/salt partitioning.

. The nature of interaction between the methylcellulose and a-amylase

was not hydrophobic in nature. The nature of interaction is most
likely some specific interaction similar in nature to the interaction
of starch and a-amylase. This conclusion is supported by work on

partitioning of fi-amylase with methylcellulose2

. Aqueous two phase partitioning with methylcellulose/ salt is not an

effective technique for the purification of a phosphatase/amylase

fusion protein containing a portion of a—amylase.

83

84

5.2 Proposals For Future Research

It would be fruitful to focus future research on several aspects of

the purification process:

1.

Investigate partitioning with other polymers such as water soluble
gums like curdlan or carboxymethyl amylase. In addition, specific

polymers could be tailor-madel.

. Investigate the exact interaction between methylcellulose and a-

amylase which this paper concluded could be similar in nature to the
starch-amylase interaction. Spectroscopic techniques seem
applicable.

Investigate the application of methylcellulose/salt partitioning with
other starch binding enzymes. (This has already been attempted with

fi-amylase, resulting in similar partitioning to a-amylase.)

. Investigate the purification of other enzymes that were designed

specifically to enhance their separation.

. Investigate the interaction of cellulose and a-amylase at different

ammonium sulfate saturations.

. Investigate further the function of temperature on partitioning,

especially at low ammonium sulfate saturations."

. Investigate the nature of the polymer phase, whether it is liquid,

precipitate or a combination of the two. Light scattering techniques

seem applicable.

5.3 Literature Cited

1.

Kennedy, J. F. 1978. Chemical synthesis and modification of

oligosaccharides, polysaccharides, glycoproteins, enzymes and
glycolipids. Carbohyd. Chem. 10:427-497.

2.

Miranda, E. 1988. M.S. Thesis. Michigan State University. (In

progress).

APPENDIX

ORIGINAL DATA

Effect of methylcellulose concentration, Figure 3.2:

The percentage of activity remaining in the salt phase as a function of
the enzyme loading.

30% ammonium sulfate; 90 units/m1 of initial amylase activity;
3.5 ug/ml of initial protein; T - 23°C

MC Final Activity Loading % Activity

(w/v) (units/ml) (g A4M/g amylase)

0.005 2.6 14 3 1.3
0.001 13.6 2.8 6.6
0.0005 33.6 1.4 16.2
0.0001 167.9 0 28 80.1
0.00005 184.3 0 14 88.9
0.0 207.4 0 0 100.0

Initial Final
Activity Protein Activity Loading % Activity
(units/ml) (ug/ml) (units/m1) (g A4M/g amylase)
0.0 100.0
2847 117.9 685.7 0.424 24.1
569 23.6 77.1 2.12 13.5
285 11.8 22.8 4.24 8.0
142 5.9 6.1 8.47 4.3
56.9 2.4 2.3 20.8 4.0

85

86

Effect of different salts, Figure 3.3:

The percentage of activity remaining in the salt phase as a function of
the ammonium sulfate or sodium chloride saturations.

0.05% w/v K4M; T - 23°C

*
% NaCl Final Activity % Activity % NS Final Activity % Activity

(units/m1) (units/m1)

0 0 129 100 O 0.0 26.0 100
15 0 102 78 9 15.0 22.9 88
20 0 84 65 0 20.0 13.6 52
25 O 87 67.3 25.0 3.6 14
30 0 90 69.7 30 0 0.6 2

* % of ammonium sulfate saturation
Effect of pH, Figure 3.4:

The percentage of activity remaining in the salt phase as a function of
pH.

0.05% w/v K4M; T - 23°C

0% NS 30% NS Final pH % Activity

53.7 2.4 4.23 4 5
102.7 1.0 5.17 1 0
88.3 1.6 5.95 l 8
112.2 3.2 6.76 2 8

Effect of temperature:

The percentage of activity remaining in the salt phase as a function of
temperature.

0.05% w/v K4M

0% NS 30% NS Temperature % Activity

113.1 1 62 1 1 4
118 7 1.46 23 1 2
120 6 1.94 46 1 6
101 3 2.80 60 2 7

87

Purification, Figures 3.5a, 3.5b, 3.6a, 3.6b and 3.7:

The percentage of activity or protein remaining in the salt phase after
partitioning of extracellular amylase.

0.05% w/v KISM; 52 units/m1 of amylase; 440 ug/ml of total protein

% NS Final Activity % Activity Final Protein % Protein

(units/m1) (ug/ml)
0 52.2 100.0 438.8 100
5 53.7 102.9 447.5 102
10 53 9 103.3 378.1 86
15 53 7 102.8 425 1 97
20 48.6 93.1 399.0 91
25 31.3 60.0 449.1 102
30 3 2 6.1 425.8 97

The percentage of activity or protein remaining in the salt phase after
partitioning of periplasmic amylase.

0.05% w/v K15M; 950 units/ml of amylase; 96 ug/ml of total protein

% NS Final Activity % Activity Final Protein % Protein

(units/m1) (ug/ml)
0 946.8 100.0 96.0 100
5 935.1 99.1 91.2 95
10 870.3 92.0 74.5 77
15 906.2 95.8 84.1 86
20 801.2 85.1 66.6 69
25 536.2 56.9 48.2 50
30 48.8 5.2 29.9 31
35 1.2 0.1 17.9 19

88

The percentage of activity or protein remaining after precipitation of
extracellular amylase.

0.00% w/v K15M; 120 units/ml of amylase; 830 ug/ml of total protein

% NS Final Activity % Activity Final Protein % Protein

(units/m1) (ug/ml)

0 121 8 100.0 832 9 100
10 121 8 100.0 --- ---
20 106 8 87.7 --- ---
30 104 3 85.6 804.4 98
40 81 0 66.5 787.1 96
50 10.5 9.0 734.4 89
60 0 3 0.3 702.2 85
70 0 1 0.1 721.3 87

The percentage of activity or protein remaining after precipitation of
periplasmic amylase.

0.00% w/v K15M; 1850 units/ml of amylase; 300 ug/ml of total protein

% NS Final Activity % Activity Final Protein % Protein

(units/m1) (ug/ml)

0 1846 100.0 298.1 100
10 1797 97.3 250.9 84
20 1854 100.4 226.6 76
30 1750 94.8 214 2 72
40 1038 56.2 142.0 48
50 92.3 5.0 55.6 19
60 106.9 5.8 7.0 2

89

The percentage of activity remaining in the salt phase as a function of
K15M concentration.

30% saturation of ammonium sulfate;

28 units/m1 of extracellular amylase;
230 ug/ml of extracellular total protein;
102 units/ml of periplasmic amylase;

10 ug/ml of periplasmic total protein

Activity % Activity Activity % Activity

0.0 102.4 100.0 28.0 100.0
0.0001 99.2 96.9 25.4 90.7
0.0005 84.3 82.3 20.3 72.5
0.001 75.2 73.4 17.2 61.6
0.003 51.6 50.4 10.4 37 1
0.005 41.1 40.1 9.4 33.4
0.008 26.2 25.6 6.8 24.1
0.01 5.1 5.0 1.4 5.0

Equilibrium, Figure 3.8:

The percentage of activity remaining after partitioning has attained
equilibrium as a function of ammonium sulfate saturation.

0.05% w/v A4M

Equilibrium of amylase as moving out of pe11et
(100% activity - 1193 units/m1)

0 1141 95 6
10 646.6 54 2
20 520.1 43 6
30 42.4 3 6

Equilibrium of amylase as moving into polymer
(100% activity - 1425 units/m1)

0 1425 100 0
10 1256 88 1
20 621.0 51 4

9O

Hydrophobicity, Figures 3.9a and 3.9b:

The adsorbtion/desorbtion of amylase and adsorbtion of protein using
octyl sepharose-C1 4B.

0.033% w/v K15M
(100% activity - 160.5 units/m1; 100% protein - 877 ug/ml)

0 82.4 51.3 8.08 5.0 863.7 98 5
10 7.3 4.5 20.8 13.0 --- ----
15 2.7 1.7 20.6 12.8 786.3 89 7
20 2.1 1.3 21.3 13.3 751.0 85.6
25 3 0 1 8 15.2 9.5 649.8 74.1
30 --- --- 12.8 8.0 607.0 69.2
35 2.2 1.4 8.0 5.0 566.1 64.6
40 0.35 0.2 5.4 3.3 446.4 50.9

The adsorbtion/desorbtion of amylase and adsorbtion of protein using
phenyl sepharose-Cl 4B.

0.033% w/v K15M
(100% activity - 160.5 units/m1; 100% protein =--877 ug/ml)

0 0.17 0.21 0.235 0.29 725.6 82.7
10 0.81 1.01 0.134 0.17 717.2 81.8
15 1.08 1.35 0.088 0.11 690.6 78.7
20 1.35 1.68 0 090 0.11 712.3 81.2
25 1.17 1.46 0 080 0.10 731.3 83.4
30 3.14 3.92 0.082 0.10 673.0 76.7
35 5.49 6.84 0.098 0.12 743.5 84.8
40 3.32 4.14 0.124 0.16 683.6 77.9

91

Comparison of partitioning of alkaline phosphatase, amylase, and the
phosphatase/amylase fusion protein, Figures 4.1, 4.2, 4.3a, 4.3b, 4.3c
and 4.4:

The percentage of phosphatase, amylase, or fusion protein activity
remaining after partitioning as a function of ammonium sulfate

saturation.

See Table 4.1 for initial conditions.

0 100.0 100.0 100.0
5 --- 102.9 ---
10 93.5 103.3 98.2
15 90.2 102.8 ---
20 81.3 93.1 93 1
25 81.2 60.0 ---
30 82.1 6.1 72.6
35 84.1 1 0 ---
40 56.6 --- 5.6
50 3.8 --- 3.3
60 0 --- 1.6
70 0 --- 1.5

The percentage of phosphatase, amylase, or fusion protein activity
remaining in the salt phase with respect to % K15M present.

See Table 4.1 for initial conditions.

0.00 100.0 100.0 100.0
0.0001 --- 90.7 ---

0.0005 98.9 72.5 88.3
0.001 99.4 61.6 82.5
0.003 97.1 37.1 81.8
0.005 100.0 33.4 85.7
0.008 105.7 24.1 88.3
0.01 103.4 5 0 86.4
0.02 101.7 --- 90.9
0.03 --- -—- 92.9

92

The percentage of phosphatase, amylase, or fusion protein activity
remaining in the salt phase after precipitation as a function of
ammonium sulfate added.

See Table 4.1 for initial conditions.

0 100.0 100.0 100 0
10 99.9 100.0 62 2
20 99.3 87.7 45 4
30 95.1 85.6 38 7
40 81.5 66.5 6.7
50 40.0 9.0 1.4
60 15.0 0.3 21 3
70 5.1 0.1 42.5

"ITIIIIIIIIIIIIIIII