MI!
“Ir-1"",

(I 'rvf

 

 

{._4/ WT"?
fl. I‘D " .. '.
., 1:344! ‘5‘"?-
{.15 p/I'p.‘.‘ (hf-fly};
7’7‘}? 'I--:-’ 'I‘ -
. . p .11 ~-« "‘"'
‘4‘“. W’-
fr I"

‘.o"’ J'?
.l‘
'3’

. -

. .'

,mfl'

-. ‘\
.

’
. - 5
‘.“-‘£

11..
01.1'
Gnu'v“ 'I

4"
If:.' “‘I

. 5.}, ."
-"'.'
— 13:”;
f 6...: ,' ""’:{g;

.r' 4-."
1:} ‘ .1?‘

,.. a" '.
oil‘
0 .r':

{I
4%”:- 6’ Jul"
'0

1,5,."

 

2‘72”".

“{4}”
1K

 

‘ O
’9'"

v'o

 

46 "
.22."? '
”211'; ' I.
.59???" 'LI‘
’3!)

4'
r.
3.1" {£3 ' ’1)
I.I}}n'.'"r'} "
4? ”if: y‘wlfl

‘
u a ' ‘ ‘ 3:5
'4‘}

P23:

.

. ’ '
w ‘

O . I
gag-3‘1"
W ,‘t‘

:’ n . :i‘
o;- u} ,“"l ' '3
5:6?“
.5; '4';

., w" p,-
rr'dh'rl"”!"
rib: n A .175
. "fit-55"3'5'w ' '
I” 'V‘LI I‘l' :1. -?'!" r”"’
.r» ~:-- 22%"-

'9

fr};

I
l. .- I ”7
3a,:afijffi" 1"(-‘I3-"'.f.
-a ‘ ' J
[1' a}. g'y{f 4.14.8" y if;
11-”- '

r

1:1;3323 .’ ‘i :3 :1},
.3377 5‘5? If" I'v‘
h???” ":4 1': 17" a?" '
fixrfifiz‘gw a I

", ' If"

u
-

_,..
2-2.3“?-
,.
z -
‘ £3:

1:
‘1;

1:?
',. o. 3"

r. " m
.4."'~"_’}?:-9”’%' .x
.5"; r'" 2""
“1-1:: {‘53: ' ' m?

--- "I

‘1 If"

it's-l; ’ '
gin-#1:)

 

.‘I_.-'-'
w?

s?"-
'53.? £3":-
$.85; ggfi‘?

.5 ',,.J
....-(.J , :"?~.J§'f
2.,”l/r"
“~11 .«fir

.a'F'
..-‘
b

(2’

[3?“

.M’ $Im"""
”I (fi’ 6’-

.- In'
’1'» ,8“

k
I ‘ firgl 1.
i;-
m?"

If. It'l': 'lr")

12:511.?! V"
,/

;;, ‘r'm

t
.o' ., '.
'-.‘ "iJ:—‘.’
2: ':
3‘
(:121
ray-:54 £753.! '
4" . . 5‘- $9".
(I ' 413: err [1,}!

“1“
.1?

3.,
F“.-
.

“gr" “1.2;:‘xlr
.r¢-'r.f."~"' :3"!
r

 

up >..
'55:.ng 711‘
v

.r- a!

{:5

31170HOC
|fi.

ill Milli

llllllllll l ill

3 1293 00577 4140
I-IDKAK 1

“Mdmmnflmu:
University

J
“my. m3 )Il
I

This is to certify that the
thesis entitled

Recovery of Clostridium thermosulfurogenes B—Amylase
From Fermentation Broth By Two-Phase Partition

presented by

Everson Alves Miranda

has been accepted towards fulfillment
of the requirements for

Masters degree in Agricultural Engineering

fl'd W

Major professor

Kris A. Berglund, Ph.D.
Date /2 ‘ f/

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

 

 

MSU

LIBRARIES
.2—

 

 

RETURNING MATERIALS:

Place in book drop to remove this
checkout from your record. FWES
will be charged if book is returned
after the date stamped below.

 

 

 

RECOVERY OF WW
[3 - AMYLASE FROM FERMENTATION BROTH BY

TWO - PHASE PARTITION

By

Everson Alves Miranda

A THESIS

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

MASTER OF SCIENCE
in
Agricultural Engineering

Department of Agricultural Engineering

1988

ABSTRACT

RECOVERY OF WWW B-AMYLASE
FROM FERMENTATION BROTH BY TWO — PHASE PARTITION

By

Everson Alves Miranda

Enzymes produced in commodity amounts, like amylases, require
simple and low cost recovery and purification for commercial-
competitiveness. A procedure for recovering a very thermoactive and
thermostable WWB-amflase from
fermentation broth by partition to a hydroxypropyl methylcellulose phase
has been developed. First, the enzyme is partitioned to a gel-like polymer
phase with ammonium sulfate.This B-amylase containing polymer phase
is separated by centrifugation. The resulting pellet can be redissolved and
the polymer recovered by a second salt precipitation. The process is not
dependent on polymer-enzyme solution pH, but it is affected by
temperature, polymer nominal molecular weight and loading, polymer-
enzyme solution dielectric constant, fermentation carbon source, and
dilution and purity of the enzyme preparation. The partition mechanism
seems to be adsorption of B-amylase to the insolubilized polymer in

the presence of ammonium sulfate.

This work is dedicated to
my wife, Chia Miao
my mother, Dona N eide

my country, Brazil

ACKNOLWLEDGEMENTS

I would like to express my deep appreciation and gratitude to Dr.
Zivko Nikolov of Iowa State University. I am in debt to him for providing
support and opportunity at the commencement of my graduate studies at
Michigan State University. This thesis is an achievement that would not
have been accomplished without him.

Deep appreciation and gratitude is expressed to my major professor, I
Dr. Kris A. Berglund, for his invaluable input, encouragement, and
friendship throughout my graduate program.

I also would like to express my gratitude to Dr. Elankovan
Ponnampalam and Mr. Lloyd LeCureux for their friendship, cooperation
and discussion in the field of protein fermentation and recovery.

To the Michigan Biotechnology Institute — MBI - my appreciation for
the financial support to carry out the project; to the Conselho Nacional de
Desenvolvimento Cientifico e Tecnolégico — CNPq — Brazil, my recognition
for their financial support through part of my program.

Finally, I would like to thank all of those who helped in one way or

another during the course of this work.

iv

TABLE OF CONTENTS

Page
LIST OF TABLES ............................................................................. vii
LIST OF FIGURES ............................................................................ ix
INTRODUCTION ................................................................................ 1 -
LITERATURE REVIEW ...................................................................... 4
Extracellular B—amylase from 4
Wm WW .................................................
FWMH’W” .................................................. 8
Protein Precipitation by Non-Ionic Polymers ................................ 1 7
Protein Partition in Aqueous Two-Phase
Systems ................................................................................... 21
MATERIALS AND METHODS ........................................................... 28
C, W Fermentation for
B-amylase Production ................................................................ 28
Cell Removal and Broth Concentration ........................................ 29
B-amylase Preparations ............................................................. 29
Polymers .................................................................................. 30
Assays ..................................................................................... 31

Page

Partition Experiments................................................................32
Small Scale Experiments ............................................ 32
Batch Experiments .................................................... 33
RESULTS ......... . ................................................................................ 35
Preliminary Experiments ........................................................... 35
Experiments with B-amylase Preparations ................................... 35
Experiments with Fermentation Broth ......................................... 47
DISCUSSION ..................................................................................... 58 '
CONCLUSIONS ................................................................................. 63
RECOMMENDATIONS ....................................................................... 65
APPENDICES .................................................................................... 67
Raw Data Used on the Preparation
of Figures and Tables ................................................................. 67
LIST OF REFERENCES ..................................................................... 80

Table

10.

11.

Page
Effect of nominal molecular weight of HPMC on B-amylase
partition from pre-purified preparation ....................................... 45
Effect of ethanol on B-amylase partition from pre-purified .
preparation .............................................................................. 50
Raw data of Figure 8 ................................................................. 67
Raw data of Figure 9 ................................................................. 68
Raw data of Figure 10 ................................................................ 69
Raw data of Figure 11 ................................................................ 70
Raw data of Table 1 ................................................................... 71
Raw data of Figure 12 ................................................................ 72
Raw data of Figure 13 ................................................................ 73
Raw data of Figure 14 ................................................................ 74
Raw data of Table 2 ................................................................... 75

LIST OF TABLES

Table Page

12. Raw data of Figure 15 ................................................................ 76
13. Raw data of Figure 16 ................................................................ 77
14. Raw data of Figure 18 ................................................................ 78
15. Raw data of Figure 19 ................................................................ 79

LIST OF FIGURES

Page

Simplified chemical structures of methylcellulose
and hydroxypropyl methylcellulose [from Handbook on
Methocel Cellulose Ethers Products] ............................................ 9

Gelation of 2% aqueous solution of Methocel A100 on
heating at 0.25 °Clmin [from Sarkar,1979] ................................... 12

An illustration of the change of the 545 nm light transmission

for an aqueous solution of a hydroxypropyl methylcellulose

sample having degree of substitution and molar substitution

(see footnote for definitions) of 1.45 and 0.11, respectively, as

function of temperature at different concentrations [from

Sarkar, 1979] ............................................................................ 13

Temperature of sol-gel transformation for aqueous solutions
of Methocel F50 as function of concentration {from Sarkar,
1979] ........................................................................................ 14

Incipient gelation temperature (IGT), incipient precipitation
temperature (IPT), and gel strength of 2% methylcellulose
(Methocel A150) solution as function of NaCl concentration

(see footnote for definitions) [from Sarkar, 1979] ........................... 15

Particle agglomerate (A) and helix-like conformation (B)
(200K) ...................................................................................... 36

 

Figure

10.

11.

Page
Droplets inside cluster of filaments (400X) ............................. 37

Effect of ammonium sulfate concentration on B-amylase

partition at different HPMC concentrations (0, 0.500% w/v;

O, 0.200% w/v; A, 0.005% w/v) and on precipitation of B-amylase

(A) and HPMC ( O, 0.200% w/v ). Enzyme used: different

pre-purified B-amylase preparations. Conditions of experiments:
enzyme activity, from 33.0 to 80.0 U/mL; protein concentration,

from 0.7 to 2.4 mg/mL ............................................................... 38

Comparison between the two types of polymers, MC (0, 0.200%

w/v; A, 0.005% w/v) and HPMC (O, 0.200% w/v; A , 0.005% w/v)

in partitioning B-amylase as function of ammonium sulfate
concentration. Enzyme used: pre-purified B-amylase.

Conditions of experiments: enzyme activity, 41 .1 U/mL; protein
concentration, 0.7 mg/mL .......................................................... 40

B—amylase and MC dissolution from a [iamylase containing

pellet ( A and 0 , respectively) and expected B«amylase

dissolution from a pellet of B-amylase precipitated with 30%
ammonium sulfate saturation ( A ) as function of ammonium
sulfate concentration of the "equilibrium solutions”. B-amylase
containing pellets were obtained by adding saturated

ammonium sulfate solution to a MC solution, mixing, and

then adding the enzyme solution ................................................ 42

Bcamylase partition as function of HPMC loading. Enzyme

used: pre-purified B-amylase. Conditions of experiments:

enzyme activity, 36.7 U/mL; protein concentration,

1.2 mg/mL; ammonium sulfate concentration, 30% saturation ...... 44

12.

13.

14.

15.

Page

B~amylase partition as function of concentration of pure

Boamylase. Enzyme used: pre-purified B—amylase ( A , 17.5

U/mng, 38.9 U/mg ) and purified B-amylase ( 0 , 608.8 U/mg).
Conditions of experiments: HPMC concentration, 0.200% w/v;
ammonium sulfate concentration, 25% saturation. The lines

shown are linear regression lines .............................................. 46

Boamylase partition as function of enzyme preparation purity.
B-amylase preparations at different specific activities were

obtained by mixing pre-purified B—amylase ( 17.5 U/mg, 7.1

mg/mL ) and purified B-amylase ( 608.8 U/mg, 1.1 mg/mL ).
Conditions of experiments: enzyme activity, 40.0 U/mL; HPMC
concentration, 0.200% w/v; ammonium sulfate concentration,

25% saturation ......................................................................... 48

B-amylase partition as function of HPMC-B-amylase solution

pH. Buffers for pH lower or higher than 7.0 were acetate or Tris,
respectively, both at 50 mM with 5 mM CaClz. Enzyme used:

pre-purified B-amylase. Conditions of experiments: enzyme

activity, 38.3 U/mL; protein concentration, 2.4 mg/mL; HPMC
concentration, 0.200% w/v; ammonium sulfate concentration,

25% saturation ......................................................................... 49

Activity distribution between precipitate (O , O) and supernatant

(A , A) on Boamylase partition as function of broth concentration

for maltose broth ( open symbols ) and starch broth (close
symbols ). Conditions of experiments: enzyme activity, from 5.0

to 29.2 and from 13.1 to 127.5 U/mL for maltose and starch broth,
respectively; protein concentration, from 2.4 to 2.6 mg/mL for
maltose broth; HPMC concentration, 0.200% w/v ; ammonium
sulfate concentration, 30% saturation ......................................... 52

Figure

16.

17.

18.

19.

Page
B—amylase purification achieved by partition from maltose
broth with 0.200% w/v HPMC and 30% ammonium sulfate
saturation as function of broth concentration ............................... 53

SDSpolyacrylamide gel electrophoresis of (1) maltose broth, (2)
supernatant, and (3) dissolved B—amylase containing pellet
obtained by partition from maltose broth (16-fold concentrated)
with 0.200% w/v HPMC and 30% ammonium sulfate saturation.
Concentrations were adjusted to have the same intensity in each
band ........................................................................................ 54.

Activity distribution and total recovery in B—amylase partition

from starch broth (8-fold concentrated) as function of

temperature ( A , pellet; 0 , supernatant; A , total recovery).
Conditions of experiments: HPMC concentration, 0.200% w/v;
ammonium sulfate concentration, 30% saturation ....................... 55

Efi‘ect of ammonium sulfate concentration on B-amylase
partition ( O , O ) and HPMC precipitation (A , A) at 21 °C
(open symbols) and at 56 °C (close symbols) from dissolved
B-amylase containing pellet obtained by partition from maltose

broth (16-fold concentrated) with 0.200% w/v HPMC and 30%
ammonium sulfate saturation ................................................... 57

Flow diagram of the developed process for B-amylase recovery
from fermentation broth by partition to a HPMC gel-like phase ...... 62

INTRODUCTION

Maltose containing syrups are experiencing increased demand due
to their low hygroscopicity, low viscosity in solution, resistance to
crystallization, low sweetness, reduced browning capacity, and good heat
stability. These syrups are used as moisture conditioners, crystallization
inhibiters, stabilizers, or bulking agents in brewing, baking, soft drink, -
canning, and confectionery industries. In Japan , the Hayashibara
Company alone produces over 20,000 tons of crystalline maltose per year
(Saha and Zeikus, 1987).

There are three types of maltose containing syrups: high maltose
syrup, extremely high maltose syrup, and high conversion syrup. Their
production is governed by the use of an appropriate enzymatic system in the
saccharification step for each specific syrup. In high maltose syrup
production the enzyme used is either B-amylase or fungal a-amylase. In
order to produce extremely high maltose concentrations, a debraching
enzyme such as isoamylase or pullulanase is used. For high conversion
syrup, glucoamylase is substituted for the debraching enzyme.

Low temperature starch saccharification has drawbacks like starch
retrogradation, high solution viscosity, and low hydrolysis rates. Slow rates
may increase reversion product formation and microbial growth. Most of
the B-amylases used industrially are of plant origin, such as barley,

soybean, wheat, and sweet potato. They are expensive and thermally

I

2

unstable. On the other hand, most microbial B-amylases are not
thermostable and thermoactive enough to substitute for the plant enzymes.
Therefore, it is desirable to have a B-amylase which can operate at high
temperatures, like the extracellular B-amylase from Wm
WM reported by Hyun and Zeikus (19853). This enzyme
presents an excellent opportunity for development of new processes for
maltose production at higher temperatures than the ones currently used.

In addition to enzyme characteristics, downstream processing plays
a key role in the development of commercial enzymes, since the recovery
process often represents the major manufacturing cost. An optimum
balance between yield and purification must be found by considering
product specification and production costs, since high yield and
purification are rarely achieved in a large scale process. Two examples are
precipitation and affinity processes. Precipitation techniques, for example
by ammonium sulfate or ethanol, are the most common recovery processes
in protein recovery due to their operational simplicity and high yields
obtained. However, purifications achieved by these processes are not high
and sometimes very poor. Affinity purification processes, like affinity
precipitation, chromatography, or ultrafiltration, often result in very large
purification with reasonable recovery. The high costs of process
development and ligand production make affinity separation processes
applicable only to very expensive products, unlike the amylases currently
under consideration.

Recovery process development for amylases must search for
processes whose optimum cost, yield, and purification fall between those for
precipitation and affinity methods. Partition in an aqueous two-phase

system is a promising candidate for such processes. The simplicity and

3

mild conditions of this method produce enzymes at relatively low cost and
high purity. Precipitation by non-ionic polymers is another suitable
process. Even though there is clear distinction between precipitation and
aqueous two-phase partition, when non-ionic polymers are used in both the
distinction is not so obvious. In fact, some authors believe that, for some
systems, they are the same basic phenomenon (Albertsson, 1986 and Fried
and Chun, 1971).

The objective of this present work was to develop the basic features of
a process for recovery of the extracellular C. W B-
amylase from its fermentation broth by partition in an aqueous two-phase
system. A polymer/salt system was chosen rather than a polymer/polymer
system in order to produce simpler process which required removal of only
one polymer for recycling and further B-amylase purification. The non-
ionic polymers selected for this study were methylcellulose (MC) and
hydroxypropyl methylcellulose (HPMC). These compounds form a gel
phase when salted-out by ammonium sulfate that can be pelleted by
centrifugation. This characteristic could be exploited as a way to separate
the enzyme containing phase and/or to carry out polymer recovery from the
purified product for recycling.

MC and HPMC showed no harmful effects in B-amylase. These
polymers have food-grade status, ready availability and modest cost. Their
carbohydrate back-bone resembles the amylose structure opening the
possibility for specific interaction between the polymers and the Bcamylase.

The basic variables of the partition process (pH, polymer loading,
temperature, fermentation carbon source, protein concentration and broth
purity) were studied regarding yield and purification. Polymer recovery

was also evaluated.

LITERATURE REVIEW

Extmoellular B-amylase from Clostridinm WW

B-amylase' from (2, mm was first isolated by Hyun
and Zeikus (1985a). The enzyme is produced in high yield as a primary
metabolite during growth on starch, maltose, or glucose and it is excreted

extracellularly. It was purified 2.4-fold by sequential precipitation with 7

ethanol, ammonium sulfate, and acetone. Some physicochemical and
kinetics properties of the Bcamylase were studied.

Shen‘ and co-workers (1988) purified this B-amylase 811-fold to
homogeneity by ultrafiltration, precipitation by ethanol, and DEAE-
Sepharose CL-6B and Sephacryl S 200 chromatographies. The
physicochemical characteristics and catalytic and general molecular
properties of this purified B—amylase were studied. The enzyme was found
to be a tetramer of 210,000 daltons composed of a subunit of 51,000 daltons.
The tetramer molecular weight is similar to the sweet potato B—amylase
(200,000 daltons), but it is glycosylated. Moreover, immunological studies
indicated that these two enzymes do not crossreact. The B-amylase from C,
W is rich in acidic and hydrophobic amino acids but low

in arginine and histidine. The sequence of the 20 first N-terminal amino

 

* EC 3.2.1.2, a-l, 4-D-Glucan maltohydrolase, saccharogenic amylase is an
exo-acting carbohydrase which cleaves alternative a-l , 4 glucosidic
linkages of starch from the nomreducing end producing B-maltose.

4

 

5

acids showed only 45% homology when compared with the same sequence
of B-amylase from mm. The specific activity of purified B-
amylase was high (4215 U/mg) when compared with activities of B-amylase
from other microorganisms (400-560 U/mg). A relatively high turnover
number (400,000 minutes-1) and a low Km value (1.68 mg/mL) may explain
this high specific activity. Unlike the behavior of the B-amylase preparation
reported by Hyun and Zeikus (1985a), this more pure B-amylase was
inhibited by Schardinger dextrins (a- and B-cyclodextrin). The enzyme
isoelectric point is 5.1 and its optimum activity pH is 5.5. However, at least
90% of the optimum activity is present in the pH range of 5.0 to 6.0. The
enzyme thermoactivity and thermostability are supported by its optimum-
activity temperature, 75 °C, and its stability at 70 °C, respectively.
Moreover, the enzyme is still stable at 80 °C if either Ca2+ (5 mM) or
substrate (1 %) is provided. Shen and co-workers (1988) concluded that the B-
amylase purified from Q Wes, is novel when compared to
other described B-amylase because of the high specific activity, extreme
thermostability, and glycosylation.

Saba and co-workers (1988) developed a purification process for this
fi-amylase based on the ability of the enzyme fermented with maltose to bind
raw starch. First, B-amylase previously purified by ultrafiltration and
ethanol precipitation was put in contact with raw starch to which it bound.
The B-amylase containing starch was separated by centrifugation. Soluble
starch was added to the pellet and the enzyme was desorbed from the raw
starch. The soluble starch-B-amylase solution was passed through an octyl-
Sepharose column pre-equilibrated with acetate buffer containing 30%
ammonium sulfate. B-amylase was adsorbed to the matrix, but the soluble

starch passed through. After washing the column with buffer, the enzyme

6

was eluted with the same buffer with no ammonium sulfate. The eluted
active fraction was dialysed and showed a single protein band when
analyzed by SDS-polyacrylamide gel electrophoresis. The purified B-
amylase had a specific activity of 4188 U/mg of protein which is very close to
the B—amylase purified by Shen and co-workers cited above.

Elankovan (1988) investigated the recovery of the (L
W B-amylase from fermentation broth. He studied two
operations (ultrafiltration and precipitation) and the effect of two types of
fermentation carbon sources (maltose and starch/maltodextrin) on these
separations. Both B-amylase fermented with maltose and with starch were
efficiently concentrated by ultrafiltration with a 30,000 molecular weight.
cut-off membrane with 88 and 85% activity recovery, respectively. However,
only the latter was eficiently concentrated in a 100,000 molecular weight
cut-off membrane. In this case, activity recoveries were 93% for B-amylase
fermented with starch and 51% for B-amylase fermented with maltose. B-
amylase fermented with starch was assumed to have a effective molecular
weight higher than the B-amylase fermented with maltose. This
assumption was confirmed by HPLC gel filtration. It was postulated that B-
amylase subunits form a biospecific complex with soluble starch or non-
stable clusters among themselves which are stabilized by the presence of
soluble starch.

Precipitation studies were carried out with two different precipitants
and their combinations (ammonium sulfate and ethanol) at different
temperatures (in the range of 4 to 50 °C). The optimum temperature and
salt concentration for precipitation by ammonium sulfate was determined
to be 20 °C at 50% saturation. Precipitation of B-amylase fermented with
starch showed a poor activity recovery, from 57 to 84%. B-amylase

7

fermented with maltose was precipitated more efficiently with activity
recovery of 98%. In both cases, purifications were low, 7-fold maximum.
Precipitation by ethanol was efficient only for B-amylase fermented with
starch due to the formation of a biospecific complex between the B-amylase
and starch. The higher activity recovery for this case was correlated to the
larger mean particle size obtained after precipitation for the enzyme
produced from starch than for the enzyme produced from maltose. The
optimum temperature and ethanol concentration were 20 °C and 20%,
respectively, and purification was around 6.2—fold at maximum. The use of
3% ammonium sulfate in conjunction with the B-amylase precipitation by
ethanol resulted in an an increase in purification to 31-fold at the same.
activity recovery. This purification is significantly higher than is usually
the case for precipitation separations (Bonnerjea em" 1986).

New processes for maltose syrup production have been developed
with this unique B—amylase. Saba and Zeikus (1988a) studied the synergistic
action between the B—amylase and two other amylases (isoamylase and c:-
amylase). They reported that isoamylase stimulated the starch hydrolysis of
B-amylase 3.6-fold with approximately 97% of the low molecular weight
products made up by products with degree of polymerization equal to two.
On other hand, a-amylase had its activity stimulated by the presence
of B-amylase. These same authors (Saha and Zeikus, 1988b) developed
processes for high maltose, extremely high maltose, and high conversion
syrups production from liquefied and raw starch taking advantage of the
thermactivity and thermostability of this B-amylase. Environmentally
compatible highly thermostable enzymes were used together with the
B-amylase allowing starch conversion at higher temperatures than used in

the past.

8

Miranda and Nikolov (1987) succeeded in immobilizing Q,

WW5, B—amylase on porous silica activated with
glutaraldehyde. The activity of the B-amylase containing silica was around

140 units per gram, but the half-life time was low, estimated to be 14 days at
60 °C.

Methylcellulose and Hydroxypropyl Mefliylcellulose
Methylcellulose (MC) and hydroxypropyl methylcellulose (HPMC)
belong to a family of cellulose ethers with applications ranging from

components of construction products to pharmaceuticals. The basic

molecular structure of these non-ionic polymers are shown on Figure 1.‘

MC can be considered one of the five most important commercial gums and
its produced worldwide under difl'erent brand names. It was first prepared
by Suida in 1905 and its commercialization in the United States started in
1938. The production of the MC in this country in 1978 was 25,000 tons
which corresponded to a value of more than 40 million dollars (Greminger,
1978; Greminger and Krumel‘, 1980; and Whistler, 1978).

MO is prepared by reacting a cellulosic raw material, like cotton
linters or wood pulp, with aqueous sodium hydroxide and methyl chloride.

The main reactions that take place are:

ReenOH + NaOH —> RceuOH-NaOH (complex)
ReellOH-NaOH z ReellONa 4' H20
RcenONa + CH3CI —> RceuOCH3 + NaCl

 

* This is an excellent review on MC and HPMC. Most of the information
presented in the literature review was taken from it.

'1'}; J1

W

 

H O H
Ho
OCH, H
H
” o
CWOCH;
H OH
Ho
OCH, H
H
H 0
0420C":

 

 

 

o
H
OH H
H OCH,
Methylcellulose
CH,o<:H, _
O 0,.
H
OH H
H
H OCH,CHCH,
t
0H

 

Hydroxypropyl Methylcellulose

 

H OCH,
H
H
0 OH
CH,OH

Figure 1. Simplified chemical structure of methylcellulose and
hydroxypropyl methylcellulose [from Handbook on Methocel Cellulose

Ethers Products].

10

where Reell represents the cellulose radical. If propylene oxide is added to

the reaction mixture, HPMC is produced according to the following

reaction

\
O OH

The reaction mixture composition (sodium hydroxide:cellulose,
methyl chloride:cellulose, and propylene oxidezcellulose weight ratios and
sodium hydroxide concentration) controls the level of substitution. The
importance of this control is that differences in substitution markedly affect ‘
the properties of the cellulose derivatives like solubility, solutions viscosity,
and biodegradation. The measurement of the substitution is given by the
concept of degree of substitution (DS, the average number of hydroxyl
groups substituted on the anydroglucose unit of cellulose) which is three at
maximum. Molar substitution (MS) is the term used when side-chain
formation is possible and its value can exceed three.

There are two difficulties in the production of MC or HPMC. First,
there is the crystalline nature of cellulose which requires a precise pre-
treatment of the raw material with sodium hydroxide. Secondly, the
reactions are exothermic and cellulose pulps do not exhibit good heat
transfer properties. Therefore, reactor design for these transformations
must have special attention to heat dissipation.

MC and HPMC are insoluble in water at high temperatures (80 -
90 °C). However, they do not dissolve by just adding polymer powder
preparations to cold water. Under these conditions, the particles gelatinize

on the surface and agglomerate among themselves prohibiting the

11

hydration and dissolution of the entire particle. In order to achieve
complete polymer dissolution, the powder must be dispersed in hot water
(80-90 °C) by agitation to wet the particles. Cold water or ice is added as the
suspension is cooled under mixing. The particles dissolve and a marked
increase in solution viscosity takes place.

Aqueous solutions of MC and HPMC behave as pseudoplastic
nonthixotropic fluids at room temperature. Solutions with more than 5% of
polymer may show some thixotropy due to weak chain-to-chain interaction.
Dilute solutions are very nearly Newtonian fluids. All the rheological
properties are strongly affected by polymer molecular weight and
molecular weight distribution .

MC and HPMC can function as polymeric surfactants since they
reduce surface tension and interfacial tensions of aqueous systems.
Thereby, they can act as emulsifiers for two-phase systems or as foaming
agents. They are efi'ective even at very low concentrations, like 0.001%, due
to their polymeric structure.

One of the most remarkable characteristics of these polymers is their
thermal gelation properties (Sarkar, 1979). In contrast to the majority of
other polymers, MC and HPMC gel when the temperature is raised. This
gelation is usually preceded by precipitation. Figures 2 to 5 illustrate these

phenomena', which are probably due to a decrease in hydration of the

 

* Some variables used to quantify gelation and precipitation are "Incipient
Gelation Temperature" (IGT) and "Incipient Precipitation Temperature"
(IPT), and " Cloud Point" (CP), respectively. IGT is defined as the
temperature at which a sharp increase in viscosity takes place when
polymer solution temperature is increased. IPT and CP are defined as the
temperatures at which light transmission through polymer solution
reaches 97.5 and 50%, respectively, when such solutions are heated.

 

 

200
Rate Of Shear = 86 sec"
160 "'—
x
C
00/1);
Q:
~ 1..
g. 120
a)
o
O
.52 ‘
>
«03
G
o
B
a 80 "'
<1 ’79
6’70
.9
\
40 r- Incipient Gelation
Temperature
0 1 1 a 1 1 1

 

 

0 1 0 20 30 4O 50 60 70
Temperature, °C

Figure 2. Gelation of 2% aqueous solution of Methocel A100 on heating at
0.25 °C/min [from Sarkar, 1979].

13

 

   
  
  

 

 

 

100
Incipient —'
Precipitation
Temperature (IPT)
80 _
C
.9
a 60 -
E 0.25%
C
‘3 Cloud Point (CP
'—
fw 40 ._
.1
39
0.5%
Rate of Temperature
20 — Increase = 0.25 °C/min
1%
0 l l l l l l
30 40 50 60 70 80 90 100

Temperature, °C

Figure 3. An ilustration of the change of the 545 nm light transmission for
an aqueous solution of hydroxypropyl methylcellulose sample having
degree of substitution and molar substitution (see footnote for definitions) of
1.45 and 0.11, respectively, as function of temperature at different

concentrations [from Sarkar, 1979].

14

 

65

60

 

Incipient Gelation
Temperature

  

Temperature, °C
U1
on

  

Incipient
Precipitation
Temperature

 

 
   

 

 

 

l l l l l l
0 l 2 3 4 5 6 7
Concentration, Wt. %

 

45

Figure 4. Temperature of sol-gel transformation for aqueous solutions of
Methocel F50 as function of concentration [from Sarkar,1979].

15

 

 

 

 

50
‘ —- 7000
A
IGT
40 _ IPT
.0 8
‘5 E’.’
5 a
e a
3 .3
E .e
f: A —« 6000 a
3
30 _ Gel Strength A
I
20 1 1 1 1 1 5000
O 1 2 3 4 5 6
NaCl Concentration, %

Figure 5. Incipient gelation temperature (IGT), incipient precipitation
temperature (IPT), and gel strength of 2% methylcellulose (Methocel A15C)
solution as function of NaCl concentration (see footnote for definitions)
[from Sarkar, 1979].

16

polymer molecules and an increase in hydrophobic interaction among
themselves when polymer solutions are heated.

The IGT and IPT of commercial MC and HPMC, which have wide
molecular weight distributions, are not very dependent on their average
molecular weights. This is due to the presence of high molecular weight
fractions, which precipitate out first and strongly influence the IGT and
IPT values. As a consequence, commercial polymers exhibit similar IGT
and IPT.

The degree of hydroxypropyl and methoxyl substitutions are
important in determining the gelation properties of these cellulose ethers.

Methoxyl groups are responsible for gelation. However, the presence of '

hydroxypropyl groups in the polymers alters the gelation characteristics
significantly, like increasing precipitation temperature and decreasing gel
strength (Sarkar, 1979).

Additives can strongly alter the gelation properties of MC solutions.
A decrease in IGT usually takes place when electrolytes, dextrose, and
glycerol, are added to MC solutions. However, some components, (e.g.,
ethanol, and propylene glycol), have a opposite efi‘ect, increasing the IGT
due to their large solvent power. Figure 5 shows the effect of salt on the
gelation and precipitation of MC. The salting out phenomena is not
explained simply on the basic of competition between the polymer and the
electrolyte for the water molecules, since the relative efficiency of the
insolubilization by difi'erent salts is function of the polymer (Sarkar, 1979).
Albertsson (1986) used this property to separated MC from a dextran/MC
solution by salting out it with ammonium sulfate.

Both polymers, MC and HPMC, are very stable in the pH range of 3
to 11 , are very resistant to microorganisms, and have food grade status. The

A .. ' - 'Jl'l‘n'llr

 

17

Food and Drug Administration lists MC as Generally Recognized as Safe
(GRAS). HPMC compounds whose methoxyl and hydroxypropyl
substitutions ranges from 19 to 30% and 4 to 12%, respectively, meet the
requirements of Food Additive Regulations 182.1480 and 172.874 as
miscellaneous and/or general purpose food additives for nonstandardized
foods.

The variation of the MC and HPMC properties over a wide range give
these polymers an extensive spectrum of uses. They are used commercially
as compounding and formulating ingredients in adhesives, agricultural
chemicals, chemicals specialities, construction industry products,
cosmetics, food products, latex painting, paint removers, paper products, '
pharmaceuticals, printing inks, resins, elastomers, textiles, and tobacco
sheet. In addition, MC and HPMC are commercially used as processing
aids in ceramics, leather, and polyvinyl chloride industries. MC has also
been used as phase forming agent in partition in aqueous two-phase
systems applied to separate biomolecules and cell particles (Albertsson,
1986 and Rule £31., 1982).

Protein Precipitation by Non-Ionic Polymers

Non-ionic polymers have been successfully used as protein
precipitants based on their ability to stabilize proteins and room
temperature use. As they are soluble in water, they can be added to proteins
as aqueous solutions in concentrations generally less than twenty percent
(Bell fig]... 1983).

Cornwell and Kruger (1961) discussed the use of sulfated
polysaccharides, including some non-ionic, for the isolation and

characterization of plasma proteins by precipitation. However, Polson and

18

co-workers (1964) were the first to publish a report of protein precipitation
with non-ionic polymers. In their pioneer work, they studied the
precipitation of plasma proteins with five different non-ionic polymers :
dextran, polyethylene glycol, nonylphenolethoxylate, polyvinyl alcohol, and
polyvinyl pyrrolidone. After preliminary experiments, polyethylene glycol
was the only one submitted to further studies because the latter three
produced significant denaturation of the protein precipitates and dextran
solutions had very high viscosity.

Martin and Landon (1975) included dextran in the antibody
precipitation technique to improve the separation of antibody-bound and
free antigen fraction in radioimmunoassays. They found dextran markedly '
increased the rate of immunoprecipitation of the antibody-bound antigen
fraction. However, it was not clear whether dextran coprecipitated with the
antibody or a partition to a second phase occurred.

Polyethylene glycol has been the only non-ionic polymer extensively
studied as protein precipitant (Bell eLaL, 1983; Ingham, 1984; Patel, 1985
and Scopes, 1987). Examples of proteins precipitated by polyethylene glycol
are plasma proteins (including immunoglobulins), alcohol oxidase,
fumarase, alcohol dehydrogenase, carboxyhaemaglobin, and invertase.
Precipitation conditions vary in polyethylene glycol concentration from a
few percent to 70% w/v and in ionic strength from 0.1 M to 2.5 M.
Precipitation pH varies from acid to alkaline ( from 3,0 to 9.0). At very high
values of ionic strength, a two-phase liquid system is formed instead of a
precipitate. Protein purification varies from crystalline forms to
precipitates with no significant purification. Protein mixtures of low
protein concentration can be fractionated by polyethylene glycol

precipitation, but good yield results only when working with highly

 

19

concentrated protein mixtures ( Janssen and Ruelius, 1968; Fried and
Chun, 1971; Juckes, 1971; Foster e131” 1973; Ingham, 1978 and Hao eLaL,
1980).

The amount of polyethylene glycol needed for a specific precipitation
a is function of pH and ionic strength. pH affects not only polymer-protein
interaction, but also protein-protein interaction by changing the net charge
on the molecules. Similar effects are provided by certain ions, the sulfate
anion in particular. These salt effects correlate roughly with the
Hofmeister series and their effective concentrations are far below the
concentrations known to work as salting-out agents (Ingham,1978). Miekka
and Ingham (1978) attributed the ammonium sulfate enhancement in the
precipitation of ehymotrypsin and chymotrypsinogen by polyethylene glycol
to the ability of these proteins to undergo self-association in the presence of
this salt.

Haire and co-workers (1984) described hemoglobin precipitation by
polyethylene glycol as a two step phenomenon. First, an amorphous solid
phase is formed by the protein solution upon polyethylene glycol addition.
This phase is very hydrated with a water mass fraction from 0.5 to 0.7.
After 17% polyethylene glycol concentration is reached, the amorphous
solid phase becomes metastable and a new phase of crystalline hemoglobin
appears.

Protein precipitation with non-ionic polymers is said to be closely
related to the formation of liquid-liquid two-phase systems from mixtures of
aqueous polymer solutions. However, its mechanism is not fully
understood. Some authors explain it as either coprecipitation or partition
between polymer/water phases. Others believe in the exclusion theory by
which the polymer excludes the protein molecule from part of the solution

2)

thereby reducing the amount of water available for solvation and bringing
the protein solution to its solubility limit (Fried and Chun, 1971; Atha and
Ingham, 1981 and Bell £31., 1983).

The exclusion volume theory has been supported by experimental
data that show insensitivity of some precipitation characteristics to changes
in solution conditions (pH, temperature or salts) and the observation that
polymer-protein interaction depends primarily on the size and shape of
polymer and protein molecules. Moreover, there is an apparent absence of
specific interactions between water soluble polymers with proteins or with
groups that commonly occur on the surface of proteins molecules.
Thermodynamic studies of the polymer-protein interaction based on the ’
exclusion volume theory by computer simulation agreed very well with
experimental data (Knoll and Hermans, 1983 and Baskir M, 1987).

The removal of non-ionic polymers from a protein precipitate is
sometimes a necessity. The polymer may be considered an undesirable
contaminant due to toxicity, adverse effects in further processing, viscous
product solutions, and interference with analytical procedures. Moreover,
if the polymer is present at high concentrations in the precipitate, its
recycling is of major interest in large scale processing. Polymers have a
linear non-compact configuration that give them a large exclusion radius,
making them difiicult to remove from globular proteins of the same or even
larger molecular weight using conventional molecular sieve methods.
Since they are nonvolatile, they can not be removed by lyophilization, a
common operation in the pharmaceutical industries. Thus polymer
removal is dificult and results in increased processing costs (Ingham and
Busby, 1980 and Smith, 1985).

21

However, there are some methods available for polymer removal.
Polyethylene glycol can be separated from other purified proteins by
ultrafiltration when the differences in molecular weight are large enough.
Other methods are ion exchange or affinity methods since polyethylene
glycol molecules have no charge or a ligand can be chosen with no
specificity towards the polymer. Precipitation of either polyethylene glycol
by salts or of proteins by any suitable method can be carried out. Formation
of a two-phase system by adding salts, like ammonium sulfate or
phosphates, under appropriate conditions can be exploited for the
separation of polyethylene glycol from the proteins, which tend to partition
into the salt-rich phase (Ingham and Busby, 1980).

Protein Partition in Aqueous 'IVvo-Phase Systems

Liquid two-phase partition, also called liquid-liquid extraction or
solvent extraction, is a very well developed technique. Such processes are
based on the distribution‘ of components of a mixture into phases of a
multiphase system, usually two-phase systems, due to differences in
aflinity of these components to each phase. Initially a tool of analytical
chemistry, solvent extraction has emerged into a major unit operation at
the industrial scale. However, common applications of this technique to
biochemical processes have been very limited (basically separation of
proteins and nucleic acids in phenol/water two-phase system) because

proteins were precipitated or denaturated in the organic phases or

 

* Quantification of distribution is given by the partition coeficient K
defined as the ratio between top and bottom phase concentration of the
partitioned substance .

 

22

concentrated only in the aqueous phases (Albertsson, 1958; Kroner M,
1978).

Some of these difficulties have been overcome by using two-phase
systems containing water, salts, and organic solvents soluble in water, like
ethanol and glycol ethers. However, partition using these systems require
careful temperature control to avoid changes in phase composition which
affect separation.

Albertsson (1958) introduced aqueous two-phase systems containing
high polymers and opened a new era for two-phase partition in
biotechnology. The high water contents of both phases (from 80 to 99
percent) make them an excellent environment for cell organelles or .
biologically active proteins that distribute differently between the two phases
( Albertsson, 1962 and Kula $31., 1982). These two-phase polymer systems
are still sensitive to temperature changes, but they can be made far from
the plait point so temperature changes lead to little concentration variation
and consequently little change in partitioning.

When solutions of two different polymers are combined, phase
separation occurs with each polymer concentrating preferentially between
the phases. The phase formation is governed by interactions between
molecules of the two polymers. The interaction energy among polymers is
proportional to polymer molecule size since it is the sum of interactions of
repetitive groups throughout the molecule. Thus, this interaction energy
per mol of polymer is very high, usually much higher than the entropy per
mol due to mixing, which is based on the number of molecules involved in
the mixing process. This additive phenomenon allows second phase
formation,even though the polymers are very similar to each other or are

present in very low concentration. This theory is in good agreement with

23

experimental data that shows that the higher the molecule weight of the
polymer, the lower its concentration necessary for phase separation
(Albertsson, 1986).

In some cases salts of low molecular weight compounds are part of
the system. Addition of salts to some polymer solutions may cause phase
separation with one phase polymer-rich and the other salt-rich, as in the
polyethylene glycol/ammonium sulfate system. Depending on the polymers
and conditions used, a gel or solid-like phase may be formed. Examples are
dextran/potassium phosphate, methylcellulose/ammonium sulfate, and
low molecular weight polyethylene glycol/dextran systems (Albertsson, 1986
and Kroner 94:31., 1982). The presence of salts can reduce the water
available for polymer hydration, as in protein precipitation by non-ionic
polymers. Such a reduction in the water available increases polymer-
polymer interaction and results in polymer segregation in a gel phase. At
the same time, ions like HPO42", SO42- or Cl" can form complexes with
polymer chains leading to the formation of gel structure.

Another interesting and useful feature of salts in aqueous two-phase

systems is the unequal partition of small ions. This creates a small
electrical potential (p across the interface given by

(p=RT/[(Z++Z-)F]ln(IL/K..)

where R is the gas constant, T the absolute temperature, F the Faraday
constant, 2* and Z“ the net charge of the cations and anions, respectively

and K... and IL the partition coefficient of the cations and anions,

respectively. Charged proteins will have their partition altered by the

24

interfacial electrical potential (1) according to the following expression

mxpeanop+[ZpF/(RT)]<1>

where 2,, is the net charge of the protein, KP the protein partition
coefiicient, K0p the partition coemcient of the protein at condition of zero

interfacial potential or at its isoeletric point where Zp becomes zero.

Therefore, pH has an important role in such partition because it can
determine, not only the protein net charge, but also the interfacial potential
by establishing multivalent ions dissociation (Albertsson, 1986 and Kula e1;
31., 1982).

Among 63 different aqueous polymer two-phase system listed by
Albertsson (1986), 20 were composed of two non-ionic polymers, and 38 have
one of the polymer components non-ionic. Polyethylene glycol and dextran
appear as components in 11 and 9 systems, respectively. Other polymers
were polypropylene glycol, polyvinyl alcohol, polyvinylpyrrolidone,
methylcellulose, ethylhydroxyethylcellulose, hydroxypropyldextran,
methoxypolyethylene glycol, sodium dextran sulfate, sodium
carboxymethyldextran, sodium carboxymethylcellulose, ficoll, and DEAE
dextran-HCI. Despite the number of polymers available for aqueous two-
phase partition, large scale processes have been developed using only
polyethylene glycol/dextran or polyethylene glycol/salts (Kula eLal., 1982).

Partition of macromolecules and cell particles is a very complex
phenomenon. The mechanism by which molecules (or particles) are
preferentially distributed between the difi‘erent phases is largely unknown.
Hydrogen bonds, charge interaction, van der Waals forces, hydrophobic
interactions, and steric efl'ects are expected to take part in the interaction

25

between the partitioned substance and the components of each phase.
Therefore, partition depends on the size and chemical properties of both the
phase forming polymers and partitioned molecules (or particles).

This complex situation makes it difficult to predict behavior of a
partition system or to interpret precisely experimental results. On the other
hand, the versatility of the approach is great.

Like in precipitation, ultrafiltration, and chromatography, chemical
modification of separating agents can greatly increase specific interactions
resulting in specificity of the separation. Hydrophobicity of one of the phases
can be increased by attaching hydrophobic groups to polymers, like fatty

acid esterified groups to polyethylene glycol. Separation of serum albumin '

in a system of such modified polyethylene glycol/dextran was increased by
approximately 56-fold (Shanbhag and Axelsson, 1975). Similar
improvement in separation can be achieved when the attached group is an
affinity ligand. Those ligands can be very specific to a certain product or
have afinity to to a variety of compounds. The formers are ligands such as
enzyme substrates and inhibiters; the latters are such as triazine dyes.
These specific partitions are called hydrophobic and affinity partition,
respectively (Kula, 1982).

Partition can be used as a research tool to detect and quantify
interactions of protein-protein, protein-small ligand, nucleic acid-small
ligand, protein-nucleic acid, and protein-lipid since these interactions
change the partition of the individual substances. However, the main
application of partition in biotechnology is on large scale concentration,
purification of enzymes, and extractive bioconversion.

Concentration of a desired product (biomolecules, viruses and cell

particles) is achieved when the system used is such that the product

“or

%

partitions to a phase that has a small volume compared to the original
volume. Examples of virus concentration by this method are bacteriophage
T2, influenza, parotitis, New Castle disease, vaccinia, poliovirus, and
tobacco mosaic virus (Albertsson, 1962; Albertsson and Frick, 1960 and
Albertsson, 1960).

A great variety of proteins have been purified by partition in
laboratory or pilot plant scale. Some examples are interferon, amylases
(e.g., pullulanase and glucose isomerase) and dehydrogenase (e.g., glucose
dehydrogenase, leucine dehydrogenase and formate dehydrogenase).
Overall purification and yield were as high as 33-fold and 98 percent,
respectively, for a maximum of 4 partition steps (Kula $131., 1982) . '

Albertsson (1986) lists the following advantages for large scale
enzyme purification by partition:

1 . Scale-up can easily and reliably be predicted from small laboratory
experiments.

2. Rapid mass transfer and equilibrium is reached by relatively little
input of energy in the form of mechanical mixing.

. It can be developed as as continuous process.
. The polymers stabilize the enzymes.

. Separation can be made selective and rapid.

cameos

. Because of the rapid separation it may be carried out at room
temperature instead of in the cold, thus at lower investment costs.

7.It is more economical than other purification methods.

Extractive bioconversion is accomplished in a two-phase system
where an enzyme and its substrate are contained in one phase while the

reaction products segregate to the other phase. The renewing of the phase

27

containing the products by repetitive partitions or a continuous extraction
may increase conversions and favor kinetics by avoiding product inhibition
and reverse product formation, respectively (Albertsson, 1986).

Removal of the polymers from the concentrated or purified products
is subject to the same problems and solutions discussed for removal of
polymers from products concentrated or purified under precipitation by
non-ionic polymers. In the case of products that are large particles such as
cells or cell fragments, centrifugation of the phase followed by repeated
washing can produce pellets of polymer-free products (Albertsson, 1986).

MATERIALS AND METHODS

C, W Fermentation for B-amylase Production

The catabolic repression resistant mutant H-12-1 (Hyun and Zeikus,
1985b) derived from (2. WW wild type strain 4B (ATCC
33743) was used. The culture was grown at 60 °C in 26 ml anaerobic
pressure tubes, Bellco Glass Inc., NJ, USA, containing 10 mL of TYE '
(trypticase and yeast-extract) medium (Hyun and Zeikus, 1985b)
supplemented with 0.5% soluble starch and a N2-002 (95 : 5) gas
headspace. The culture was expanded twice, to 500 mL in a round bottom
flask and to 16 L in a 18 L carboy. All the transfers were made under
anaerobic conditions and the medium used was TYE with 0.5% of maltose
or partially hydrolysed starch (maltodextrin) added as the fermentation
carbon source (substrate). A last transfer for final growth, still under
anaerobic conditions, was made to a 250 L fermentor containing 180 L of the
same medium used previously.The fermentor content was stirred at
constant rate (400 rpm), N2-002 was purged for a few hours to remove any
dissolved oxygen present and temperature was kept at 60 °C. Antifoam
emulsion FG-10, Dow Corning Corp., MI, USA, was added to suppress
excessive foaming due to high production of gases during fermentation.
Growth was followed by hourly turbidity measurements and the

fermentation was stopped after detecting that the stationary phase was

28

29

achieved. The culture was cooled down to approximately 15 °C before cell

removal and concentration.

CellRemovalandBroth Concentration

Cell removal and concentration was done under aerobic conditions.
In the cases where maltose was used as the substrate, the cells were
removed by ultrafiltration of the culture broth in an Amicon Hollow Fiber
Ultrafiltration Cartridge, 100,000 MWCO membrane, model H26MP01-43. A
batch of 20 L of this broth free of cells (called maltose broth) was then
concentrated to 800 mL using an Amicon Spiral Ultrafiltration Cartridge,
30,000 MWCO membrane, model S10Y30. Three aliquots of 500 mL were A
taken during the concentration. These aliquots were denoted by their
concentration ratio: ratio between the initial volume (20 L) and the volume
of the concentrated at the time sampling was performed. For the
maltodextrin case, the cell removal was done by centrifugation at 8,000 rpm
for 10 minutes and the concentration of the resulting broth (called starch
broth) was performed the same as for the maltose broth case.

B-amylase Preparations

Two B-amylase preparations were used for the first set of
experiments at small scale. They will be denoted pre-purified and purified
enzymes (or preparations) to take into account that one has a higher
specific activity than the other due their different processing as described
below.

B-amylase contained in 1 L of maltose broth was precipitated by
adding 500 mL of saturated solution of ammonium sulfate at room

temperature. After slow mixing for 1 hour the suspension was left

30

overnight at 4 °C. The precipitate formed was separated by centrifugation at
8,000 rpm for 30 minutes, suspended in 200 mL of 50 mM acetate buffer
pH=6.0 with 5 mM CaClz, and dialysed overnight at 4 °C against this same
buffer. The dialysed enzyme solution was frozen in aliquots of 10 mL. This
is denoted as pre-purified B-amylase. Prior to use, the enzyme solution was
thawed, centrifuged for 10 minutes in a clinical centrifuge, and filtered
through Gelman Acrodisc 0.45 pm filters.

Chilled ethanol was added dropwise to pre-purified B-amylase while
stirring to give a final alcohol concentration of 20% v/v. The precipitate
formed was suspended and dialysed against the acetate buffer above. This
is denoted as purified Bcamylase. SDS-polyacrylamide gel electrOphoresis of '
this preparation showed only one protein band.

Polymers

The MC and HPMC used were the commercial products Methocel‘
donated by Dow Chemical Company, Midland, USA. MC was type A4M
with nominal molecular weight of 85,000 daltons and methoxyl degree of
substitution of 1.6-1.9. Three types of HPMC were used: K15M, K4M and
K100 M Premium with nominal molecular weight of 15,000, 85,000 and
250,000 daltons, respectively. These HPMC's have methoxyl degree of
substitution and hydroxyprOpyl molar substitutions in the range of 1 .1 to 1.4
and 0.1 to 0.3, respectively. The solutions were prepared according to the
manufacturer's instructions : wetting and dispersing the particles in hot
acetate buffer (temperature in the range of 80 to 90 °C), then adding cold
buffer to dissolve the wetted particles. Stock solutions of the various

 

* Trademark of the Dow Chemical Company, USA.

31

polymers were prepared (1 %) and kept at 4 °C. Dilutions to lower

concentrations were made by weight.

Assays

A reaction mixture containing 800 1.1L of 2.5% w/v boiled starch and
200 uL of enzyme solution was used for the routine enzyme assay for [3-
amylase activity. Both solutions were prepared in 50 mM acetate buffer
pH=6.0 with 5 mM CaC12 and the assay temperature was 60 °C. Reducing
sugars released by enzymatic hydrolysis of starch were determined by the
dinitrosalicylic acid method (Bernfeld, 1955). One unit of B-amylase was
defined as the number of umol of reducing sugars as maltose produced per .
minute under the above conditions.

Total carbohydrates content was determined by slight modification of
the phenol-sulfuric method of Dubois eLal. (1956). Aliquots of 200 uL were
added to 200 uL of 5% phenol solution in a cuvette. After mixing, 1 mL of
95.5% sulfuric acid was added and after 5 minutes the content of the
cuvettes were again thoroughly mixed. The absorbance at 470 nm was
measured after 15 minutes. MC A4M was used as the standard. Protein
was determined by the method of Lowry M. (1951) with bovine serum
albumin as the standard. SDS-polyacrylamide slab gel electrophoresis was
performed as described by Laemmli (1970). Molecular weight standards
were bovine serum albumin (66,000 daltons), egg albumin (45,000 daltons),
glyceraldehyde 3 - phosphate dehydrogenase (36,000 daltons), carbonic
anhydrase (29,000 daltons), trypsinogen (24,000 daltons), and a-
lactoalbumin (14,200 daltons).

The effects of ammonium sulfate or HPMC present in samples on

activity, total carbohydrates, and protein assays were investigated. The only

 

32

significant interference for the practical concentration ranges observed was
ammonium sulfate on the activity assay for concentrations higher than
12% saturation. Therefore, all samples containing ammonium sulfate were

diluted at least to 10% prior to the activity assay.

Partition Experiments

Quantitative experiments were done in two ways. First, screening
experiments were performed to study the use of MC and HPMC in the
partition of B-amylase. These were small scale experiments carried out in
1.5 mL Eppendorf tubes and the enzyme preparation used was pre-purified
or purified B-amylase. Subsequently, experiments were performed with.
broth in a batch fashion.

In both the small scale and batch experiments, the parameter used
for evaluation regarding activity partition was "percentage partitioned".
This is defined as the percentage of the activity missing from the
supernatant compared with the initial value determined in a control
experiment. This is an indirect measurement of how much activity was
partitioned to the pellet. "Percentage precipitated" was defined in a similar
way to account for the polymer or B-amylase salted-out by ammonium
sulfate.

Percentage of saturation was used as the concentration unit for
ammonium sulfate assuming that all solutions were at 21°C. The solubility

for ammonium sulfate at this temperature is 757 g/L.

Small Scale Experiments
Small scale experiments were conducted in 1 .5 mL Eppendorf tubes
using 500 11L of enzyme preparation and 500 uL of HPMC or MC solution at

33

twofold the final desired concentration. Polymer solutions and enzyme
dilutions were prepared with the acetate buffer cited before, unless specified
otherwise. Very viscous solutions of these polymers were measured by
weight. After homogenization in a vortex for 10 seconds, the appropriate
volume of saturated ammonium sulfate solution was added and the content
of the tubes was mixed again in a vortex for another 10 seconds. The
reaction mixture was kept at 21 °C for approximately 10 minutes and then
centrifuged at 12,000 rpm for 10 minutes. The supernatant was separated
from the pellet and filtered through Millipore Millex HV 0.45 pm filters.

The filtrate was used for enzyme activity and total carbohydrates assays.

Batch Experiments

Batch experiments were done with cell-free maltose and starch
broths, with different levels of concentration fixed by ultrafiltration
, (concentration ratios from 1 to 17). HPMC and ammonium sulfate
concentration were 0.200% w/v and 30% saturation, respectively. Broth pH
was in the range of 5.1 to 5.2 and after addition of HPMC very little change
was observed.The batch experiments were carried out in a 50 mL beaker at
room temperature, unless specified otherwise. Agitation was provided by a
magnetic stirrer. No significant difference in activity recovery was noticed
when mixing was provided by paddles indicating the stir bars did not
denature the protein. Broth and HPMC aliquots, 20 and 5 mL, respectively,
were homogenized for 5 minutes. 10.7 mL of ammonium sulfate saturated
solution was then added dropwise and mixing was continued for more 10
minutes. The pellet and supernatant were separated by centrifugation at .
8,000 rpm for 20 minutes and the pellet was dissolved in 100 mL of buffer.
This solution was used to study HPMC recovery and for electrophoresis

34

after concentration in 10,000 MWCO disc membrane using an Amicon
Ultrafiltration Cell, model 8010. The same solution was also used for
enzyme activity, total carbohydrates, and protein assays. The volume of the
supernatant was measured and the supernatant was used for activity
assay. One aliquot of the supernatant was dialysed and used for
electrophoresis after concentration using the same disc membrane and

ultrafiltration cell as above.

RESULTS

PreliminaryExperiments

The continuous addition of ammonium sulfate solution to enzyme-
MC or enzyme-HPMC solutions, under mixing, produced a cloudy
suspension. As the ammonium sulfate concentration achieved a certain
level, the particles agglomerated among themselves in a filamentous form. .
Observing this suspension with a microscope, one can see particles and an
arrangement of them in linear form which twist around each other
forming a helix-like structure (Figures 6). Droplets, probably gelified
polymer, were detected inside clusters of the filaments (Figure 7).
Therefore, the insoluble phase seems composed of two and not just one
phase. A decrease in solution viscosity was noticeable during the
insolubilization, since the polymer was been removed from the solution.
Centrifugation of this mixture produces a gel-like compact pellet. This
pellet gradually lost its white color while dissolving slowly in water. It

became a transparent gel before complete dissolution.

Experiments with B-amylase Preparations
Figure 8 shows the ability of HPMC to promote partitioning of B-amylase to
the polymer phase. B-amylase activity recovery of 98% was achieved for

ammonium sulfate concentration of 30% saturation.

35

 

Figure 6 — Particle agglomerate (A) and helix-like conformation (B) (200K).

 

Figure 7 - Droplets inside cluster of filaments (400X).

 

 

 

 

 

5100- -
e .

:3

.E‘

so- -
CL. .

5-1

0

"o 60' '
q) .

c:

O

14?:

t3 40' '
c6 .

O-1 .
80

c0 20' -
+2 .

c:

G)

O

8 a
m 0 - ulnnnnlnnnnl-n-nln4k

O 10 Z) 30 4O

Ammonium Sulfate Concentration
(Percentage of Saturation)

Figure 8 - Effect of ammonium sulfate concentration on B-amylase partition
at different HPMC concentrations (0 , 0.500% w/v ; O , 0.200% w/v ; A ,
0.005% w/v ) and on precipitation of B-amylase ( A ) and HPMC ( O , 0.200%
w/v ). Enzyme used : different pre-purified B-amylase preparations.
Conditions of experiments : enzyme activity, from 33.0 to 80.0 U/mL; protein
concentration from 0.7 to 2.4 mg/mL.

39

Approximately 10% less ammonium sulfate is necessary for
partition as compared to precipitation. However, this occurs at a
ammonium sulfate concentration approximately 14% higher than the one
necessary to precipitate HPMC from a 0.200% w/v solution with no enzyme
present.

Comparison of the two polymers, MC and HPMC, as partitioning
agents showed that they are equally effective in partitioning B-amylase at
30% ammonium sulfate saturation (Figure 9). However, HPMC is more
effective at lower ammonium sulfate concentrations.

B-amylase partition takes place at a ammonium sulfate
concentration intermediate to the concentrations necessary to precipitate .
the enzyme and to segregate HPMC from its solution. There are two
possibles explanations for this. First, it may be that polymer and B—amylase
interact with each other and form a complex that is insoluble at a specific
ammonium sulfate concentration. Second, the enzyme may partition to the
polymer phase during its formation, since low ammonium sulfate
saturation, for example 15%, is enough to separate more than 95% of the
HPMC from its solution.

To check this second possibility, a pellet of MC was prepared in a
Eppendorf tube from 1 mL of 0.400% w/v MC solution by adding
ammonium sulfate solution up to 30% saturation. The pellet was washed
twice with 1 mL of 30% ammonium sulfate saturated solution and 1 .43 mL
of pre-purified B-amylase with 30% ammonium sulfate saturation was
added to it. The tube was shaken for 30 seconds, let stand at room
temperature for 10 minutes, and after centrifugation the supernatant was
filtrated and analyzed for activity. Only 9% of the initial activity was found
on this supernatant. If only B-amylase precipitation by ammonium sulfate

 

 

 

 

 

100' it "
"aeo— -
a .

.3

+9

E h

“60' '

D4 .

C)

an s

+340" '

Cl i

Q) o

o

i-I

0°:
m1- -
Onn4a114nalannn1

O 10 Z) 30

Ammonium Sulfate Concentration
( Percentage of Saturation )

Figure 9 - Comparison between the two types of polymers, MC ( O , 0.200%
w/v ; A , 0.005% w/v) and HPMC ( O , 0.200% w/v ; A , 0.005% w/v) in
partitioning B-amylase as function of ammonium sulfate concentration.
Enzyme used : pre-purified fi-amylase. Conditions of experiments : enzyme
activity, 41.1 U/mL; protein concentration, 0.7 mg/mL.

41

had occurred, the supernatant would have around 21% of the initial
activity. This indicates that B-amylase partitions to the polymer phase
rather than coprecipitates with it.

Hydrophobic interaction is known to be the mechanism responsible
for protein binding to some polysaccharide columns, such as Sepharose 4B
and cellulose, just as in hydrophobic chromatography (Arakawa, 1986). To
check if this is the case of the process under investigation, partition
experiments were performed with a protein well known as hydrophobic,
bovine albumin. The experiment conditions were 30% ammonium sulfate
saturation and 0.200% w/v HPMC (loading of 10 milligrams of polymer to
one milligram of protein). No more than 30% of the bovine albumin was .
partitioned, dismissing hydrophobic interaction as the major mechanism
on 8amylase partition.

The influence of ammonium sulfate concentration on the B- amylase
separation to the polymer pellet was checked by a "reverse experiment". 8-
amylase-containing pellets were prepared by adding 429 uL of saturated
solution of ammonium sulfate to 500 uL of 1% w/v MC solution,mixing, and
then adding 500 uL of enzyme solution. After 15 minutes under gentle
mixing, the supernatant was separated by centrifugation. Ammonium
sulfate solutions at different concentrations, called "equilibrium solutions",
were added to the pellets. The Eppendorf tubes containing the mixtures
were shaken for 30 seconds and allowed to stand at room temperature
overnight. The supernatant was separated by centrifugation and analyzed
for activity and total carbohydrates. The results in Figure 10 show that
polymer dissolution was negligible. The percentage of activity in solution

was a function of the ammonium sulfate concentration and it was lower

 

 

 

 

 

c:

.2 1m' ' try I * TI fil I 'n
g .

:1 .

H

O

m P

.E 80' '
o I

2 1

5 60j i
.3, .
.E L

*5 40 ‘
< p «1
.,_. .
o q
8’.) 20,“ '
cs

4.)

c:

§ Gunfic.-.“
ii 5 10 15 20 25 so 35

Ammonium Sulfate Concentration
( Percentage of Saturation )

Figure 10 - B-amylase and MC dissolution from a B-amylase containing
pellet ( A and O , respectively) and expected B-amylase dissolution from a
pellet of B-amylase precipitated with 30% ammonium sulfate saturation (A)
as function of ammonium sulfate concentration of the "equilibrium
solutions". B-amylase containing pellets were obtained by adding saturated
ammonium sulfate solution to a MC solution, mixing, and then adding the
enzyme solution.

43

than the percentage expected from dissolution of just [fr-amylase
precipitated by ammonium sulfate.

HPMC loading (mass of polymer per mass of protein) is shown in
Figure 11. A minimum of 0.8 is required for maximum partition. At the
final protein concentration used, 1.2 mg/mL, this loading correspond to a
HPMC concentration of 0.100% w/v.

Table 1 contains the results of experiments made with HPMC of
different nominal molecular weights. The HPMC type K4M, with nominal
molecular weight of 85,000 daltons, produced the highest partition among
the three polymers studied.

The effect of the enzyme dilution was investigated at three different _
levels of contaminant proteins concentration (different specific activities).
Two preparations of pre-purified B-amylase (specific activities of 17.5 U/mg
and 38.9 U/mg) and a purified Boamylase (specific activity of 608.8 U/mg)
were diluted with acetate buffer to different activities. These dilutions were
partitioned at 0.200% w/v HPMC and 25% ammonium sulfate saturation.
Results of these experiments were plotted based on the concentration of
pure B-amylase present in each dilution, assuming a specific activity of
4215 U/mg (Figure 12). The lower the concentration of pure B-amylase, the
lower the partition. However, this behavior was much stronger as the (5-
amylase preparation was more impure. Enzyme preparations with specific
activity of 608.8 U/mg is not affected significantly by dilution in the range
concentration studied (from 11.18 to 66.72 micrograms of pure B-amylase
per milliliter).

A better quantification of the influence of protein impurities on the
partition was made with enzyme solutions at different specific activities but

at the same concentration as pure B-amylase (19 ug/mL). Volumes of pre-

 

lm V j V V I V V V *‘T " V V V V I V V V *V I

 

 

Percentage Partitioned

 

 

l a

0.0 1.0 2.0 3.0 4.0

HPMC Loading
(mg HPMC/mg protein)

A A

 

Figure 11 - B-amylase partition as function of the HPMC loading. Enzyme
used : pre-purified B-amylase. Conditions of experiments : enzyme activity,
36.7 U/mL and protein concentration, 1.2 mg/mL ; ammonium sulfate
concentration, 30% saturation.

45

Table 1. Effect of the nominal molecular weight of HPMC on B-amylase
partition from pre-purified preparation.

 

 

HPMC Nominal molecular weight Percentage partitioned
K15M 15,000 65.5
K4M 85,000 84.6
K100M Premium 250,000 74.6

 

Conditions of experiments : HPMC concentration, 0.200% w/v ; HPMC

loading, 0.7 ; ammonium sulfate concentration, 25% saturation.

 

 

 

 

 

 

100 . ' 1 - r - .
O U
'U
c
a -1
.2
.4: ..
43
A
c6 «1
0-1
m
00
C6
49
G
Q) .-
o
H
o
0...
40 L a j 1 4 a a l L a n l
0 20 40 60
Concentration of Pure B-amylase
( ug/mL)

Figure 12 - p-amylase partition as function of concentration of pure
B-amylase. Enzyme used : pre-purified B-amylase ( A , 17.5 U/mg ;D , 38.9
U/mg) and purified B-amylase ( O , 608.8 U/mg). Conditions of experiments:
HPMC concentration, 0.200% w/v ; ammonium sulfate concentration, 25%
saturation. The lines shown are linear regression lines.

47

purified and purified B-amylase (specific activities of 17.5 and 386.7 U/mg,
respectively) were combined, diluted, and the resulting solutions were
partitioned at 0.200% w/v HPMC with 25% of ammonium sulfate
saturation. Results are plotted on Figure 13. The influence of contaminant
proteins, negligible for specific activity higher than 150 U/mg, becomes
increasingly important as the purity of the enzyme preparation decrease
from this point. At specific activity of 17.5 U/mg, 24% decrease in partition
is observed as compared to the value at 150 U/mg.

In order to verify the effect of B-amylase-HPMC solution pH on the
partition, pre-purified enzyme (7.1 mg/mL) and 1% w/v HPMC solutions,
both in 50 mM acetate buffer pH=6.0 with 5 mM CaClg, were diluted 3- and
2.5-fold, respectively, with buffers at pH's of 3.3, 4.0, 5.0, 6.0, 7.5, 8.5, and
10.0. Bufi'ers with pH lower than 7.0 were 50 mM acetate with 5 mM CaC12
and those with pH higher than 7.0 were 50 mM Tris with 5 mM CaClz. The
combination of 500 uL of enzyme solution diluted with each one of the above
bufi‘ers and 500 uL of enzyme solution diluted with the same buffer provided
final concentration of polymer of 0.200% w/v and pH's of 3.6, 4.2, 5.7, 6.0,
7.4, 8.5, and 9.3. The results of partition of these preparations showed that
the effect of pH is negligible (Figure 14). Differences in partition for
mixtures with pH's above and below 7.0 is probably due to the use of
diflerent buffers for each one of these ranges.

Introduction of ethanol to the system prior to ammonium sulfate

addition was deleterious to the B—amylase partition (Table 2).

Experiments with Fermentation Broths
Batch partition experiments with 0.200% w/v HPMC and 30%

ammonium sulfate saturation were carried out with both types of broth at

 

 

 

 

1m....,....,....,

8 P

:90 ~

o .

cs

05

$4

53

q,80- -

OD ‘ .

CO

49

Ci .

Q)

o

5 .

0.70.
maglnlJnnllnmnnlnlmn
0 100 200 300 400

Specific Activity ( U/mg )

Figure 13 - p-amylase partition as function of enzyme preparation purity.
B-amylase preparations at different specific activities were obtained by
mixing appropriate volumes of pre-purified p-amylase (1 7 .5 U/mg, 7.1
mg/mL) and purified B-amylase (608.8 U/mg, 1.1 mg/mL). Conditions of
experiments : enzyme activity, 40.0 U/mL ; HPMC concentration, 0.200%
w/v ; ammonium sulfate, 25% saturation.

 

I
l

70

Percentage Partitioned

 

 

 

3.0 4.0 5.0 6.0 7.0 8.0 9.0
pH

Figure 14 - B-amylase partition as function of HPMC-B-amylase solution pH.
Buffers for pH lower or higher than 7.0 were acetate or Tris, respectively,
both at 50 mM with 5 mM CaClz. Enzyme used : pre-purified B-amylase.

Conditions of experiments : enzyme activity, 38.3 U/mL ; protein
concentration, 2.4 mg/mL ; HPMC concentration, 0.200% w/v ; ammonium
sulfate concentration, 25% saturation.

50

Table 2 - Effect of ethanol on B-amylase partition from pre-purified

 

 

preparation.
Ethanol concentration“ Percentage partitioned
0 84.6
5 67.9
10 36.4

 

*in v/v before addition of ammonium sulfate
Conditions of experiments : HPMC concentration, 0.200% w/v ; HPMC -
loading, 0.7 ; ammonium sulfate, 25% saturation.

51

various levels of concentration (Figure 15). HPMC loading, measured for
the maltose broth case, was in the range of 0.78 to 0.83. Basically, no 8-
amylase denaturation took place. The behavior of maltose broth was the
same as in the small scale experiments, with more than 95% of the enzyme
recovered in the pellets even though broth specific activities were low, in the
range of 2.0 to 11.5 U/mg. However, B-amylase from starch broth was not
efficiently partitioned, with only 30% of the activity found on the pellets at
most.

Partition of the B-amylase from maltose broth achieved different
purification according to the concentration given by ultrafiltration (Figure
16). The purification for the maltose broth concentrated 17-fold in volume '
was confirmed by SDS-polyacrylamide gel electrophoresis (Figure 1 7).

Temperature increases were investigated to improve the partition of
B-amylase from starch broth. Since HPMC precipitates at relatively high
temperatures, solutions of this polymer were added dropwise to the broth to
which ammonium sulfate had already been added in sufficient amount to
give a final concentration of 30% saturation. Starch broth, ammonium
sulfate, and HPMC solutions were equilibrated at the temperature of the
experiment prior to addition of one to the other. The centrifugation was
done using a pre-warmed rotor. Even though high levels of partition were
not achieved, a reasonable amount of B-amylase activity was partitioned at
75 °C (approximately 25%) while partition was almost negligible at 21 °C
(Figure 18). The maximum enzyme denaturation was around 10% at 75 °C.

Separation and recovery of the HPMC from the B-amylase containing
pellet by a second precipitation with ammonium sulfate was studied at two
difi‘erent temperatures, 21 and 56 °C, in small scale. The pellet used for this
experiments was produced by partitioning 8amylase from 8-fold

 

 

 

 

 

 

.3100: '
.E .
43
o .
<11 80' -
a .
43
O
s 60- -
t...
o ‘1
CD - .
g” 40- -
a .
G
(D
a .
o " -
m 20-

.. I 3:2

0 nhl I . .. 1.4 . l.
0.0 5.0 10.0 15.0

Concentration Ratio

Figure 15 - Activity distribution between precipitate (O ,O) and supernatant
(A , A ) on B-amylase partition as function of concentration for maltose broth
(open symbols) and starch broth (close symbols). Conditions of experiments:
enzyme activities from 5.0 to 29.2 U/mL and from 13.1 to 127.5 U/mL for
maltose and starch broth, respectively ; protein concentration from 2.4 to 2.6
mg/mL for maltose broth ; HPMC concentration, 0.200% w/v ; ammonium
sulfate concentration, 30% saturation.

 

 

 

 

100’ j
. 1
o 80." '
3
cs .
a: .
c: 60? ‘
o . .
B
CO ’ t
8 ' .'
'1: 4o_ 1
a 1 .
O-c - 1
20: '
O a n a n l a a n a l A L a a l A L
0 5 10 15

Concentration Ratio

Figure 16 - B-amylase purification achieved by partition from maltose broth
with 0.200% w/v and 30% ammonium sulfate saturation as function of broth
concentration.

 

Figure 17 - SDS-polyacrylamide gel electrophoresis of (1) maltose broth, (2)
supernatant, and (3) dissolved p-amylase containing pellet obtained by
partition from maltose broth (16-fold concentrated) with 0.200% w/v HPMC
and 30% ammonium sulfate saturation. Concentrations were adjusted to
have the same intensity for each band.

 

 

 

 

 

 

 

1m . rfir 1+ - 15* u - fiwfi u -
*3
° 1 + -
33 “\~
0
<1 80- -
H L .
c6 .
Ii .
'5'
1—1 60' "
.—1 i ‘
Q ~ .
43
5 ' ‘

40' "
‘H , .
0
g E I
8 20- -
Ci .
G)
O
H
G) r ‘
n4 O‘.i.._L..u.AL..rrij'.ilij....ili

Z) 40 5O 60 70

Temperature (°C )

Figure 18 - Activity distribution and total recovery in B-amylase partition
from starch broth (8-fold concentrated) as function of temperature ( A ,
pellet ; O , supernatant ; A , total recovery ). Conditions of experiments :
HPMC concentration, 0.200% w/v ; ammonium sulfate concentration, 30%
saturation.

56

concentrated maltose broth. More than 90% of the polymer could be
precipitated along with only 5% of activity partition at both temperatures
(Figure 19). However, only 5% ammonium sulfate saturation was required
for the same separation at 56 °C while at 21 °C the salt concentration
required was 15% saturation. This is the result of the shift of the partition
and precipitation curves to the left produced by the raising in the
temperature. It is interesting to note that at 56 °C the partition and
precipitation curves are more separated from each other than at 21 °C. A
batch polymer separation was made and samples were analyzed to check
for further B-amylase purification by SDS-polyacrylamide gel
electrophoresis. No improvement in purification was observed (gel pictures _

not shown).

 

 

 

 

 

8

45100: ...__ '

4D

"-1

Q1

op‘ O

8 b

H 80' .

D-i .

h p

o 1-

"d - ..

o 60

a 1-

O ,

0H

4.3

"-1

1'3 . .

p., 1

c ' °

3°

+9 Z)’ "

a P I

Q)

U

H o

0)

n4 0-44..1L..41.1..14.m.
O 5 10 15 20 % 30

Ammonium Sulfate Concentration
(Percentage of Saturation )

Figure 19 - Effect of ammonium sulfate concentration on p-amylase
partition ( O , 0 ) and HPMC precipitation ( A , A ) at 21 °C (open symbols)
and at 56 °C (close symbols) from dissolved B-amylase containing pellet
obtained by partition from maltose broth (16-fold concentrated) with 0.200%
w/v HPMC and 30% ammonium sulfate saturation.

DISCUSSION

The small scale experiments showed that B-amylase partition to a
HPMC second phase is a very fast, simple and efficient technique, even
though its mechanism does not seem to be simple. The experiments
suggest that the polymer is first insolubilized and then the B-amylase is
adsorbed to it according to the ammonium sulfate concentration of the .
solution.

The slightly better behavior of HPMC for partition of B-amylase in
comparison to MC may be explained by comparison of the methoxyl degree
of substitution of HPMC ( in the range of 1.1 to 1.4) and MC (in the range of
1.6 to 1.9), since it is known that degrees of substitution lower than 1.4 give
products with low water solubility (Greminger and Krumel, 1980).

The minimum HPMC loading necessary for maximum partition,
0.8, is a relatively low loading if compared with some polyethylene glycol
loadings that can be as high as 31.1 (Foster e111" 1973). This low loading
contributes to low recovery cost and low viscosity of the HPMC containing
solutions.

The nominal molecular weight of HPMC has a significant influence
on the partition of B-amylase, contrary to what was expected since it is
known that IGT and ITP for commercial HPMC have little dependence on
this variable. Based on the results showed in Table 1, it seems that there is

an optimum HPMC molecular weight for the B-amylase partition.

58

59

Dilution and purity were found to be important variables with respect
to how much enzyme partitions to the HPMC phase while pH was not. A
competition of the contaminant proteins for the polymer, B-amylase, or
both could explain this behavior. Self-association of protein is known to
increase protein precipitation by polyethylene glycol. If such protein
interaction plays a significant role in the B-amylase partition, competition
of contaminant proteins with the B-amylase may be the mechanism by
which partition of preparations with low specific activity/high dilution
results in low recovery. Partition insensitivity to enzyme-polymer solution
pH may be due to the buffer efi‘ect of the relatively high ammonium sulfate
content used on the preparation (around 1 .7 M) when compared to acetate .
or Tris bufi‘er concentrations (50 mM).

The approach of decreasing the dielectric constant of the solution
prior partition by adding ethanol does not improve B—amylase separation.
Again, the concentration I of ammonium sulfate, a strong electrolyte, may be
the explanation for poor recovery since it may overcome the ethanol effect.
Moreover, ethanol is known to increase solubility of HPMC in water.

The batch experiments showed that partition with HPMC can be
applied to recover and purify Q themgsulfiimgenes B-amylase from the
fermentation broth with very high yield (95% or more). Purification
achieved (21- to 101 -fold) was very high when compared with purification
obtained by ammonium sulfate, ethanol, or ethanol/ammonium sulfate
precipitations (7-, 6.2- and 31-fold, respectively).

The behavior of maltose and starch broth in B-amylase partition is
similar to the behavior in precipitation by ammonium sulfate reported by
Elankovan (1988). In each case, fi-amylase fermented in maltose was

efficiently separated (activity recovery higher than 90%), while B-amylase

so

fermented with starch was poorly separated (activity recovery of 57% or
lower). These results suggest that interaction with starch makes B-amylase
unable to interact and partition with HPMC.

The conclusion above provides an explanation as to why partition of
B-amylase fermented in starch takes place only at high temperatures and is
a function of the temperature increase (Figure 18). The reason may be in
the kinetics of the starch hydrolysis by the enzyme, since increasing in the
temperature increases the reaction rate. As the temperature gets closer to
the B-arnylase optimum temperature, 75 °C, it is plausible to expect that
each enzyme molecule is more efficient in hydrolysing the starch
molecule(s) bound to itself or to other B-amylase molecules. Eventually, the ,
enzyme molecules would be free of the starch moiety and, then, able to
interact with HPMC molecules. Another contribution to the success in
partition at high temperature came from the HPMC solubility properties.
HPMC is less soluble at higher temperature, (e.g., 75 °C), than at 21 °C, and
B-amylase partition at higher temperatures is more efficient for the same
ammonium sulfate concentration, as one can see on Figure 19.

In a companion work, two other amylases, a thermophilic a-
amylase and a pullulanase, were also partitioned by the same HPMC
system. These facts suggest that there is a specificity involved in amylase
partition to HPMC and that this specific interaction may mimic amylases'
binding to starch.

The low solubility of HPMC at higher temperatures is probably the
reason why recovery of HPMC is more efficient at 56 °C. The shift in
partition and precipitation curves caused by temperature increase (Figure
19) shows a very important feature of this technique. Recovery of more than
95% of B-amylase activity took only 20% saturation in ammonium sulfate.

61

This is one third less than the value at 21 °C and 40% the salt needed by
precipitation with only ammonium sulfate. Therefore, the use of
temperatures higher than 21 °C can substantially reduce the amount of
ammonium sulfate needed, not only for polymer recovery, but for B-amylase
partition, also.

HPMC-B-amylase pellets were dissolved in beakers with magnetic
stirring for two to three hours. Even when this dissolution took place at
21 °C, no loss in activity was detected. HPMC may stabilize the B-
amylase , as it is the case of polyethylene glycol with some enzymes (Hao gt
al., 1980).

A process for recovery of Q. Wm B-amylase from _
fermentation broth is illustrated by the general scheme in Figure 20. If the
gain from purification and volume reduction due to ultrafiltration does not
overcome the cost and drawbacks of introduction of this processing step, the
operation should be omitted. The specific enzyme application will
determine what further processing steps (e.g., desalinization,
concentration, or precipitation) must follow partition in order to achieve

specifications.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Phase I

 

Separation i l

Pellet Supernatant

 

 

 

 

 

Ammonium
Permeate HPMC Sulfate
’ l l
Broth | '
Free of Cells , . .
—---—"' Ultrafiltration Partition
l
A Phase ' Polymer
T ' l ‘ Preci itation
Pellet Separation P
( HPMC ) l ?
Supernatant
( [iamylase )

Ammonium
SMQw

 

Pellet
Dissolution

 

l

Buffer

Figure 20. Flow diagram of the developed process for B-amylase recovery
from fermentation broth by partition to a gel-like phase.

CONCLUSIONS

1. Addition of ammonium sulfate to a final concentration of 30% saturation
to a solution of C. W B-amylase preparation and HPMC

or MC produces a gel-like phase which contains the enzyme. The
mechanism by which B-amylase is separated to the gel-like phase is _
partition to the polymer. First the polymer precipitates at approximately
15% saturation at 21 °C. The enzyme adsorbs to this polymer phase as the

salt concentration is increased from this point.

2. A minimum loading of 0.8 milligram of HPMC per milligram of protein

is necessary to assure maximum B-amylase partition from pre-purified

preparations.

3. HPMC is a better partition agent for B-amylase than MC for ammonium
sulfate concentrations lower than 30% saturation. The two polymers seem

to be equivalent to each other for partition at 30% saturation.

4. Bamylase partition is a function of the nominal molecular weight of the
HPMC in the range of 15,000 to 250,000 daltons.

64

5. Dilution and impurity of the B-amylase preparation afl‘ect its partition for
pure B-amylase concentrations from 0.3 to 26.7 ug/mL and specific activities
from 17.5 to 608.8 U/mg. The higher the dilution or impurity, the lower the
partition.

6. B-amylase partition does not depend on the HPMC-enzyme solution pH in
the range of3.6 to 9.3.

7. B-amylase partition is decreased by the introduction of ethanol to the
system prior to the addition of ammonium sulfate. 10% ethanol causes a

reduction of 57% in partition.

8. B—amylase produced from maltose partitions from the fermentation broth
to HPMC phase with high yield (activity recovery of 95% or more) and high
purification (at least 20-fold). 30% saturation in ammonium sulfate and

0.200% w/v are required conditions for the separation.

9. B-amylase does not partition efficiently from starch broth with HPMC
(maximum activity recovery of 25% at 75 °C ) due to its binding to starch.

10. HPMC can be recovered from a dissolved B-amylase containing pellet
solution by a second ammonium sulfate precipitation. Recovery of 90% of
the polymer along with only 5% of the total activity previously partitioned
can be obtained with 5% ammonium sulfate saturation at 56°C.

RECOMMENDATIONS

The following points are proposed topics for future research:

1. A better understanding of the partition mechanism is needed in order to
optimize and access the general applicability of this technique. One
approach is to apply this partition to other enzymes which have very well-
known structure or properties (like hydrophobicity, affinity bindings). The .
general applicability and clues to the partition mechanism may be provided

simultaneously by such experiments.

2. Optimization of partition and polymer recovery temperatures are needed,
since it was shown that the amount of ammonium sulfate required for a

specific partition yield decreases as the temperature increases.

3. Kinetics of HPMC phase formation and engineering aspects of pellet
separation from solutions in partition and polymer recovery steps must be
studied in order to optimize polymer separation and, consequently, activity
yield and polymer recovery. These pellets have a density approaching the
broth density indicating that large scale centrifugation may not be an
efficient separation technique. Filtration, macro or micro (membrane

separation) filtration, may be considered as centrifugation substitute.

65

$

4. Polymer recovery was shown to be feasible. However, it is necessary to
determine if the recycled polymer has the same partition properties as the
original one and how many times recycling would be possible.

5. Other salts (e.g., sodium chloride, magnesium chloride, and sodium
carbonate) should be considered as substitute for ammonium sulfate.
Besides processing cost, pellet physical characteristics, HPMC loading, and
enzyme purification may be improved by using other salting-out agents.

APPENDIX

67

.AERE v.« 3 he Ecuudemaabamocoo 5393 3.53 cdw 3 9mm 89¢ .5353 08.35 ”Esoatenxo we acomfivuoo
.Snoatoqxo 3.3... .8 2.5 .«e 3152 one me $394...

 

 

 

 

I «.3 I I I 9.
I «.5 I I I 9.
I haw odm 26 33 mm
«.mm has 35 «.8 v.8 8
«.8 can QS «.3 «.5 N
«.8 «.5 w.«m «do «.8 8
Ema oz: H.m« Q8 93 3
hwm «d v.5 «.3 v.5 3
Ga I 3: 3: an a
02mm 23 «$86 emu—«89¢ samocd £686 éd 55233
me $350qu
sacs—33689 c<3 cctaoflabaeocoo 02mm «anyway 3 coach—80:8
mo ommacooamm as 355ng emu—.953. me $350qu 3.53 5532.32

 

a 293m .3 3% 5m - a sage

nah—ME was avian. we acaauaoum 25 .3 com: 3.5 Bum N 56593

.AEBE he .uoswuucooaoo ESPE ”485 “.3 .5558 08.35 ”macoatoaxe mo macflmvaoo

 

 

 

 

Sb «.8 fima «.8 8
«.5 odw I I N
w.«m nah v.m« «.vm 8
H.m« 1mm I I ma
v.8 «.8 «.3 «.3 3
Q3 Q3 I I m
>3 «$86 £3 esogd £3 $95.0 33 $83.
53333
02mm 02 no 0935989

 

venomfifiwa $350qu

an damage—89:8
Sam—am 83.382

 

m 23E .e 3% Em - a. 2%...

.m. sea. 8 as. so... 353% 83.5 _.

358535 25 «o 85:2. 02 895 5022655 mScuthonuao 33 no om3n8umn «o $326 25 93 309 -5
$585598 53... «o 30:3 02 895 32835 .3353 «c 555533 «o 839' 98 « 59¢. .H au< -m

 

 

 

«.3 m. H I «.v I mm

«.5 m.« fiw «.m «mm 8

«.3. «.H «.3 «.«H «.3 N

«.«w «A «.3. «.9. Q: 8

«.m« «.m «.3 «8 «.m« «a

$9350qu :5 EUR $98.5. « ao< H ao< 555.358
$55.53 8:58:53 .3 omauaoouom

.3 «5333589 own—«Ema «c 55:3 55:3 02 895 330355 a35>53§ ma acmaahnmonou
895 028mm. 3 .3358 vBuoaxm .33 .8 .5353 .«o $35935 3815 8:58:32

 

S 25E as 3% 3am - m 2.3.

70

.35333 525 .ac535338
03.53 855883 “.553 NH 6335:330 5395 ”‘55: «.8 £55.53 08.33 ”Baofitoaxm mo 30553.5

 

 

 

Qmm «.on . «.mcH mug «mHH.

me QmoH «.3 hH..HoH «8H

H.«m H.«m «.mm chH mmwd

«.8 @on Qua H.«w «3.:

{ma Qmm msHm v.8 mwcd

mHm flaw H.wa v.3 N36

m.«w Qmm Hdw méw wed

owwum>< m 32:53me N acofitaxw H unmfitonnm

5393 we ma
Uganda

5285533 3339qu

 

: 23E .3 5% Ex . m 2.3.

833853 $3 dam—88888
3&1; 838.885 ”pd $5.32 02mm "$3 $25.: 83858388 055: ”mu—8081098 no 285980

 

 

 

mi. Qt. m5. 9563 83895 $83M
mam v.8 9mm Soda 2mg

wwauo>< N aco8€waxm H 8m8tqum
EH03 88838 3883A 0553

vmaosmfimn omficmoumm

 

H 2.12.. 5% 3am - u 2.3.

dcsmugam fimNdoSmuaawoaco Sam—am 88o88m "£3 §ocNd .coamuacmouco 02mm ”muso8tmmxm mo maomfivcoo

 

 

72

 

 

 

$3 I I 3.:
NS fiNm mNS mafia
NSm NAN ANNE Nuém
mNu mac 2:. and
can Emu QR. 8N
map flab mdw 8.:
mdw mdm v.2. NNSH
me me me vaN
wéw v.3 wdw 3.8
mdm I I omd
9mm I I and
ado I I wvé
«Na I I 8N
can I I 86
98 I I mad
«.5 I I 8.:
mdw I I En:
09834 N mum H mum $939.3. N gum a mum wwwuo>< N num— a am 38%.:
omd—h8?n 9:8 we
was 38 35 33 was 9: 8582850

 

85358 098% acouoamv he @2835an mafiaogom

 

NH 033% 3% 33H - a $35.6

73

dosages §mN .acflwuaaooaou

mama—am 83:25:“ ”£3 $ooNd 533.5528 035! "€53 adv .3353 08.35 ”3581095 no macmfivnoo

 

 

N. E. QS

Qmw 9cm

Nam 99:
v.8 QR:
Ema :68
6mm admN
v.8 odom.
th admm
fimm Emma

$308135 25 me $835 Egan—Sn mag—580m 38:: .3330: ammoam.

 

2 2am Is 5% Em - a 2.3.

74

damn—E33 8mm .cofiabcmocco menu—dc 83:25:“ ">15 $256

.aomuwbcwonoo 02mm SEES vd .noaganoonou 5395 "485 «.«« .3333 manna ”magmatogxo mo 2838.30

 

 

 

«A «.3. «.9. «.«b c«.«
«.6 «mu «.3. 5.3. and
«A c.«b «.3 kg 3.6
Nd 5.2. «$5 «.5 86
he «db «.2. «.2. ohm
«.o «.8 Nam v.8 84.
EN «db «.8 «.3. 8.«
_N acofitqum - a acwfitmaxa omwum>< N acwfitaxm a acmfitonxm
ma

3555.59 omflamuumm

 

E 255% 5% 3am - S 2.3.

75

3823 83.8888 3.6 8:382 USEE ”33 «$36

domawugwu «me $853335”.

538.5888 02mm ”38:53.8 .8 macmficaoo

 

 

v.m« ~.m« u.«« S
«Sm «.3 «d. w
«.v« «.v« «.3 o
8385. N unmatmaxm a “amatmaxm 882.; cm $38988

 

«2856.89 $3.80qu

858.5838 3.85m

 

a 2.3. 8.. 3% 3am - z 2.3.

76

.mgxméaxfiuq

dosggam ““me .coawuaaoocoo wean—um 53°88“ 53 §oONd

.cosmueamocco 0553 £83 2338 you 4:58 9N 3 v.N Scum 533.5588 53?:— 535898." £33 £033
Him mag—«E Sm HES mSNH 3 H.mH 89a 98 Has: NdN 3 9m 89: .3333 08.35 ”madmaflaxm .3 83:32.0

 

 

 

 

 

mN whm m.wm mdm oi ms ad ad NH
NH. mam mdw me QN Gm QHH hm H.
9m map adv H..Nw ed «.mN «me v.8 m 595 Acumen
o.N me wNm wdm vdH Qm H NdN wd N
mg. wHw 1mm QR. v.NH QNH ms 98 H
c.H NH. Hum Haw Nd QR: EEH QR: NH
c.H m6 «.6 H3 H3 H.wm Qua «.8 H.
«.6 Na Ham o6 o. H mfim 93 98 m 595 $332
o.H N6 hm Ev Nd wfioH N.wcH meH N
o.H Nd Hum NS EH wdm v.8 93 H
< ammuo>< N mum H Rum < $925. N gum H 9me
unaamEonsm magma—H85 3...“: 5323350 58m

 

89388.3 .5 55355 £33 33.

 

fl mama a. 3% 3am - m H as“...

 

Wow mdm flea 5

 

 

Qmm man 33 N.
mém adv .93 v
odw mush hum N
v43 fix: Ema fl
3334 N eamfitaxm a anmatmaxm 3.22
53338.80

.53 ccmuwomtsm

 

3 2:5 a. 3% 3am - 2 2.3.

don—83mm obom 83388888 38.23 83.8885 ”$3 «336 83358888 0353 ”3808185 mo gown—€85

 

 

 

 

Q8 98 03 Ema méN QNN fimN mu

flaw QR. céw QE. 9: an: 5.: mm

flow v.5 wdm va EN NA «an on

Q8 m.mw N.mw 0% EN N.N N.» mm

Qmm 6mm mNm «am HA NA md _N
$598}: 09295. N mum H mum mmauo>< N mum H 8Q
8358» A Dov

3350a :33. acfiagmaam ngfiufim 2333808

 

$93523 .5 52535 .358 3:5

 

2 055 a. 3% 3am - 3 ~38.

79

 

 

 

 

 

I I I m.wm m.wm ado. I I I mgvm mém aim 8
v.8 mam Qua Qua Qua Qua v.3 Nam mdm how «.3 Qmm 8
mdm mdm v.8 mdm Nam v.5 ”.8 Qmm 56w NdN EN EN 8
v.5 QB v.5 mHm mHm mHm HSN H.wm Q8 m6 3: c.H mH
me v.8 Q8 N.NH H.mH N.mH m.vH Nd Hd ad ad ad 3
mHm N.Nm me od ad ad Nd Qm 9N ad ad Q: m
Haw cdw Nah I I I Hum wé $6 I I I v
H. Hm me haw I I I hm H.m Nd I I I m
Gum QE Gum I I I QN md QN I I I N
ham H.wN mam I I I I I Nd I I I H

0w08>< NQNMH H86— 0w08>< Nana Hana 0ma8>< Nnme Hana 0u08>< Nam Ham
8030.880
0.. mm H 0.8888809 Do HN u 0.8889809 Down u 0.888980% 0.. HN n 085809809 .8 $0800.83

 

 

60838808 0353 .8 0w38008m

883383 0m38008m

00 803880088
30.28 8=m=o88<

 

2 2:08 a. 5% sum - 3 028.

LIST OF REFERENCES

LIST OF REFERENCES

Albertsson, P-. A. 1958. Partition of proteins in liquid polymer-polymer two-
phase systems. Nature, 182, 709-711.

Albertsson, P-. A. 1962. Partition methods for fractionation of cell particles
and macromolecules. Methods Biochem. Analysis. 10, 229-262.

Albertsson, P-. A. 1967. Two-phase separation of viruses. In : Methods in
Virology; Maramorosch, K. and Kopsowski, H., Eds.; Academic, New
York, pp 303-321.

Albertsson, P-. A. 1986. Partition of Cell Particles and Macromolecules.
John Wiley and Sons, New York.

Albertsson, P-. A. and Frick, G. 1960. Partition of virus particles in a liquid
two-phase system. Biochim. Biophys. Acta. 37, 231-237.

Arakawa, T. 1986. Thermodynamic analysis of the effect of concentrated
salts on protein interaction with hydrophobic and polysaccharide columns.
Arch. Biochem. Biophys. 248 (1), 101-105.

Atha, D. H. and Ingham. K. C. 1981. Mechanism of precipitation of
proteins by polyethylene glycols. J. Biol. Chem. 256 (23), 12108-12117.

Baskir, J. N., Hatton, T. A., and Suter, U. W. 1987. Thermodynamics of the
separation of biomaterials in two-phase aqueous polymer systems : effect of
the phase-forming polymers. Macromolecules. 20, 1300-1311.

Bell, D. J ., Hoare, M., and Dunnill, P. 1983. The Formation of protein
precipitates and their centrifugal recovery. In : Advances in
Bioengineering/Biotechnology, Downstream Processing; Fiechter, A., Ed.;
Spring-Verlag, Berlin, pp 1-72.

Bernfeld, P. 1955. Amylases on and [5. Methods Enzymol. 1, 149-150.

Bonnerjea, J ., Oh, S., Hoare, M., and Dunnill, P. 1986. Protein purification,
the right step at the right time. Bio/technology. 4, 954-958.

81

Cornwell, D. G. and Kruger, F. A. 1961. Molecular complexes in the
isolation and characterization of plasma lipoproteins. J. Lipid Res. 2 (2),
110-134.

Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F.
1956. Colorimetric method for determination of sugars and related
substances. Anal. Chem. 28, 350-356.

Elankovan, P. 1988. Ph. D. Thesis, Michigan State University, East
Lansing, USA.

Foster, P. R., Dunnill, P., and Lilly, M. D. 1973. The precipitation of
enzymes from cell extracts of W eerexisjee by polyethylene
glycol. Biochim. Biophys. Acta. 317, 505-510.

Fried, M. and Chun, P. W. 1971. Water-soluble nonionic polymers in
protein purification. Methods Enzymol. 22, 238-248.

Greminger, G. K. 1978. Cellulose derivatives, ethers. In. Kirk- Othmer
encyc10pedia of chemical technology; Grayson, M., Ed; John Wiley and
Sons, pp 143-163.

Greminger, G. K. and Krumel, K. L. 1980. Alkyl and
hydroxyalkylalkylcellulose. In: Handbook of water soluble gums and
resins; Davidson, R. L., Ed.; McGraw Hill, New York, pp 3/1-3/25.

Haire, R. N., Tisel, W. A., White, J. G., and Rosenberg, A. 1984. On the
precipitation of proteins by polymers, the hemoglobin-polyethylene glycol
system. Biopolymers. 23, 2761-2779.

Handbook on Methocel Cellulose Ethers Products (Form 192-702-78), Dow
Chemical, Midland, USA.

Hao, Y. L., Ingham, K. C., and Wickerhauser, M. 1980. Fractional
precipitation of proteins with polyethylene glycol. In : Methods of Plasma

Protein Fractionation; Curling, J. M., Ed.; Academic Press, New York, pp
57-74.

Hyun, H. H. and Zeikus, J. G. 1985a. General biochemical characterization
of thermostable extracellular B- -amylase from

themsnlmmgenea. Appl. Environ. Microbiol. 49 (5), 1162-1167.

Hyun, H. H. and Zeikus, J. G. 1985b. Regulation and genetic enhancement
of glucoamylase and pullulanase production in W
memehydmsmmm. J. Bacteriol. 164 (3) 1146-1152.

Ingham, K. C. 1978. Precipitation of proteins with polyethylene glycol,
characterization of albumin. Arch. Biochem. Biophys. 186 (1), 106-113.

82

Ingham, K. C. 1984. Protein precipitation with polyethylene glycol. Methods
Enzymol. 104, 351 -356.

Ingham, K. C. and Busby, T. F. 1980. Methods of removing polyethylene
glycol from plasma fractions. Chem. Eng. Commun. 7, 315-326.

Janssen, F. W. and Ruelius, H. W. 1968. Alcohol oxidase, a flavoprotein
from several basidiomycetes species. Biochim. Biophys. Acta. 151, 330-342.

Juckes, I. R. M. 1971. Fractionation of protein and viruses with
polyethylene glycol. Biochim. Biophys. Acta. 229, 535-546.

Knoll, D. and Hermans, J. 1983. Polymer-protein interactions. J. Biol.
Chem. 258 (9), 5710-5715.

Kroner, K. H., Hustedt, H., Granda, S., and Kula, M. -R. 1978. Technical
aspects of separation using aqueous two-phase systems in enzyme isolation
processes. Biotechnol. Bioeng. 20,1967-1988.

Kroner, K. H., Hustedt, H., and Kula, M.-R. 1982. Evaluation of crude

dextran as phase-forming polymer for the extraction of enzymes in aqueous
two-phase systems in large scale. Biotechnol. Bioeng. 24, 1015-1045.

Kula, M. -R., Kroner, K. H., and Hustedt, H. 1982. Purification of enzymes
by liquid-liquid extraction. In: Advances in Biochemical Engineering,
Reaction Engineering; Springer-Verlag, Berlin Heidelberg, New York, pp

73-117.

Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly
of the head of the bacteriophage T4. Nature 227, 680-685.

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

Martin, M. J. and Landon, J. 1975. The use of ammonium sulphate and
dextran together with second antibody at the separation stage in
radioimmunoassay. In: Radioimmunoassay in Clinical Biochemistry;

Pasternak, C. A., Ed.; Heyden, New York, pp 269-281.

Miekka, S. I. and Ingham, K. 1978. Influence of self-association of proteins
on their precipitation by polyethylene glycol. Arch. Biochem.Biophys. 191
(2), 525-536.

Miranda, E. A. and Nikolov, Z. 1987. Michigan Biotechnology Institute,
Lansing, USA. Unpublished results.

Patel, P. R. 1985. Enzyme isolation and purification. In: Biotechnology,
Applications and Research; Cheremisinoff, P. N. and Ouellette R. P., Eds.;
Technomic, Lancaster, pp 534-563.

83

Polson, A. Potgieter, G. M., Largier, J. F., Mears, G. E. F., and Joubert, F
J. 1964. The fractionation of protein mixtures by linear polymers of high
molecular weight. Biochim. Biophys. Acta. 82, 463, 475.

Saha, B. C., LeCureux, L. W., and Zeikus, J. G. 1988. Raw starch
adsorption-desorption of a thermostable B- -amylase from Clem

themeammmgenee. Anal. Biochem. Forthcoming.

Saha, B. C. and Zeikus, J. G. 1987. Biotechnology of maltose syrup
production. Process Biochem. June, 78-82.

Saha, B. C. and Zeikus, J. G. 1988. Synergistic action between thermostable

B-amylase from Winn} Lhemesnlfumgenee and other amylases on
raw starch degradation. Michigan Biotechnology Institute, Lansing, USA.
Working paper.

Saha, B. C. and Zeikus, J. G. 1988b. Improved method for preparing high
maltose conversion syrups. Biochem. Bioeng. Forthcoming.

Sarkar, N. 1979. Thermal gelation properties of methyl and hydroxypropyl
methylcellulose. J. Appl. Polymer Sci. 24, 1073- 1087.

Scopes, R. K. 1987. Protein purification, principles and practice; 2nd. ed.;
Springer-Verlag, New York.

Shanbhag, V. P. and Axelsson, C.-G. 1975. Hydrophobic interaction
determined by partition in aqueous two-phase systems. Eur. J. Biochem.
60,17-22.

Shen, G. -J., Saha, B. C. Lee, Y. -E., Bhatnagar, L., and Zeikus, J. G. 1988.
Purification and characterization of a novel thermostable B-amylase from

Clastfidiumthmsulfiimgencs Biochem J 254 835- 840

Smith, J. E. 1985. Biotechnology Principles. American Society for
Microbiology, Washington, D. C.

Whistler, R. L. 1978. Carbohydrates. Kirk-Othmer Encyclopedia of
Chemical Technology vol. 4. Grayson, M., Ed., John Wiley and Sons, pp
545.