Prevention of thermal inactivation and aggregation of lysozyme
by polyamines
Motonori Kudou
1
, Kentaro Shiraki
1
, Shinsuke Fujiwara
2
, Tadayuki Imanaka
3
and Masahiro Takagi
1
1
School of Materials Science, Japan Advanced Institute of Science and Technology, Ishikawa, Japan;
2
Department of Bioscience,
School of Science and Technology, Kwansei Gakuin University, Hyogo, Japan;
3
Department of Synthetic Chemistry
and Biological Chemistry, Graduate School of Engineering, Kyoto University, Kyoto, Japan
Proteins tend to form inactive aggregates at high tempera-
tures. We show that polyamines, which have a relatively
simple structure as oligoamids, effectively prevent thermal
inactivation and aggregation of hen egg lysozyme. In the
presence of additives, including arginine and guanidine
(100 m
M
), more than 30% of 0.2 mgÆmL
)1
lysozyme in
sodium phosphate buffer (pH 6.5) formed insoluble aggre-

developed to prevent the formation of protein aggregates.
One of the major approaches used to prevent protein
aggregation is the addition of small molecules to the
solution. This is a relatively simple method compared with
using chaperon systems [6–8].
The small molecular additives used to prevent the
formation of protein aggregates are classified as protein-
denaturing reagents or others. Denaturants, typically
guanidine and urea, weaken the hydrophobic intermole-
cular interaction of proteins [9,10]. Detergents, such as
Triton-X100 and SDS, are stronger protein-denaturing
reagents than denaturants [10,11]. Not only do these
reagents dissolve aggregates and inclusion bodies but they
also unfold the native structure of proteins. Accordingly, the
concentration at which this type of reagent is effective at
preventing the aggregation and inactivation of proteins is
hard to determine.
3
Arginine (Arg) is a nondenaturing reagent that has been
used widely as an additive to prevent protein aggregation
[9–12]. Arg does not destabilize the native structure, having
only a minor effect on protein stability [11,13], and enhances
the solubility of aggregate-prone molecules. Because of its
beneficial properties, Arg has been used for various proteins
and situations. However, the effect of Arg and other
nondenaturing additives does not completely solve the
aggregation problem. The development of better additives
for preventing protein aggregation has been long awaited.
In this article, we focus on naturally occurring poly-
amines [putrescine, NH

2
)
3
NH
2
] as small molecular additive candidates
for preventing heat-induced aggregation and inactivation of
proteins. There are a large number of different polyamines
4
in hyperthermophiles [14–16], which suggests that poly-
amines have a biophysical role in the adaptation of
hyperthermophilic proteins to high temperature environ-
ments. Although it has been reported that polyamines bind
to biomolecules (DNA, RNA, and platelets) by electrostatic
interactions [17–19], at present no research has been
published regarding the role of polyamines on thermal
aggregation and inactivation of proteins.
Materials and methods
Materials
Hen egg white lysozyme and betaine/HCl were purchased
from Sigma Chemical Co. All amino acids, guanidine/HCl,
urea, putrescine/2HCl, spermidine/3HCl, and spermine/
4HCl were purchased from Wako Pure Chemical Industries
(Osaka, Japan). Micrococcus lysodeikticus for the kinetics
Correspondence to K. Shiraki, School of Materials Science,
Japan Advanced Institute of Science and Technology,
1-1 Asahidai, Tatsunokuchi, Ishikawa 923-1292, Japan.
Fax: + 81 761 51 1655, Tel.: + 81 761 51 1657,
E-mail:
Abbreviations: DCp, heat capacity change; DH, enthalpy change;

heat treatment at 98 °C, the samples were centrifuged at
15 000 g for 20 min. The absorbance of the supernatants
was monitored by using a Jasco spectrophotometer model
V-550 (Japan Spectroscopic Company, Tokyo, Japan) to
determine the concentration of soluble lysozyme, using an
extinction coefficient of 2.63 cm
)1
per mgÆmL
)1
7
[20].
Residual activity of lysozyme after heat treatment
The bacteriolytic activity of lysozyme was estimated as
follows [21]. A 1.5 mL volume of 0.5 mgÆmL
)1
M. lys-
odeikticus solution prepared in 50 m
M
sodium-phosphate
buffer (pH 6.5) was mixed with 20 lL of the heat-treated
samples containing 0.2 mgÆmL
)1
lysozyme and 100 m
M
additive. The decrease in light scattering intensity of the
solution was monitored by measuring the absorbance (A)at
600 nm. The rate constant of inactivation was determined
by fitting the data to a linear extrapolation.
CD spectra
Far-ultraviolet CD spectra were measured using a Jasco

), and midpoint temperature of thermal unfolding
(T
m
) were determined by a conventional method, as
described previously [22].
Results
Heat-induced aggregation of lysozyme
Hen egg white lysozyme (pI ¼ 11) was used as a model
protein because its mechanism of refolding and misfolding
has been extensively studied [12,13,23–27]. Lysozyme can
preferentially refold into its native structure from thermally
unfolded states, while under neutral pH it tends to form
irreversible aggregates during heating [23–25].
The amount of heat-induced aggregates produced from
0.2 mgÆmL
)1
lysozyme,whenheatedto98°C for 30 min,
was measured in the presence of various additives at
pH 6.5 (Fig. 1). The amount of aggregates gradually
Fig. 1. The amount of heat-induced aggregates produced in the presence of various additives. Solutions containing 0.2 mgÆmL
)1
lysozyme (pH 6.5)
and various concentrations of additives were heated at 98 °C for 30 min. After heat treatment, the amount of aggregates was calculated by
determining the soluble concentration of lysozyme by centrifugation. (A) Arginine (Arg), (d); glycine (Gly), (s); guanidine, (h). (B) Betaine, (d);
trimethylamine-N-oxide, (s); putrescine, (m); spermidine, (h); spermine, (j). (C) NaCl, (d); KCl, (s); urea, (j).
4548 M. Kudou et al. (Eur. J. Biochem. 270) Ó FEBS 2003
decreased with increasing concentrations of Arg or
guanidine from 0 to 0.5
M
(Fig. 1A). In contrast,

initial concentration of lysozyme. In the absence of any
additives, the concentration of soluble proteins reached a
plateau at  0.07 mgÆmL
)1
. Further increase of the protein
concentration resulted in a gradual increase in the soluble
concentration of lysozyme, from 0.07 mgÆmL
)1
to
0.14 mgÆmL
)1
. In the presence of 100 m
M
spermidine or
spermine, no aggregates were observed at a protein
concentration of < 0.4 mgÆmL
)1
. With an increasing con-
centration of lysozyme, the curve reached a plateau at
 0.7 mgÆmL
)1
. Interestingly, putrescine was clearly less
effective than spermidine, whereas spermine was as effective
as spermidine. This implies that an important factor
required for polyamines to prevent protein aggregation is
the presence of a secondary amine, rather than the number
of cations or molecular mass.
Aggregation by cooling
We examined the heat-induced aggregation of lysozyme
during cooling (Fig. 3A). Protein solutions of 0.2 mgÆmL

Heat inactivation of lysozyme
The recovery of enzymatic activity after heat treatment is
another criterion used to estimate the effect of additives
because it is the most reliable measure of whether additives
prevent irreversible misfolding as well as aggregation.
Figure 4 shows the residual activity of lysozyme after heat
treatment at 98 °C. In the absence of additives, the
inactivation curves of 0.2 and 1.0 mgÆmL
)1
lysozyme
fitted well to single-exponential equations (Fig. 4A). The
inactivation rate constants for 0.2 and 1.0 mgÆmL
)1
lyso-
zyme were 0.067 and 0.21 min
)1
, respectively. The heated
samples, containing 1.0 mgÆmL
)1
lysozyme (black circles in
Fig. 4A), were resolved by the addition of guanidine/HCl
(to a final concentration of 4.0
M
)
10
and vortexing for 15 min.
These samples were diluted 10-fold by 50 m
M
sodium-
phosphate buffer (pH 6.5), after which the residual activities

molecules, under the experimental conditions used in this
study
11
, were mainly stabilized by covalent bonds (probably
disulfide exchanges) rather than by noncovalent inter-
actions.
In the presence of 100 m
M
putrescine, the inactivation
curves of 0.2 and 1.0 mgÆmL
)1
lysozyme depended on the
protein concentration, as shown by single-exponential
equations
12
(Fig. 4B). The inactivation rate constants for
0.2 and 1.0 mgÆmL
)1
lysozyme with 100 m
M
putrescine
were 0.023 and 0.11 min
)1
, respectively, which were two- to
threefold slower than those in the absence of additives.
Interestingly, in the presence of 100 m
M
spermidine and
spermine, the inactivation curve of 1.0 mgÆmL
)1

recovered by the addition of 100 m
M
spermidine and
spermine, respectively. On the other hand, the residual
activity was < 5%, for most of the other additives (even
100 m
M
Arg),
13
which is an order of magnitude lower than
for spermidine and spermine. Also, at a protein concentra-
tion of 1.0 mgÆmL
)1
and heat treatment for 10 min,
polyamines prevented heat inactivation of lysozyme more
effectively than the other additives (Table 1).
DSC analysis
To reveal the effect of additives on protein stability,
thermodynamic parameters were determined using DSC.
Representative DSC curves of lysozyme in the presence of
100 m
M
additive are shown in Fig. 5, and the thermo-
dynamic parameters derived from nonlinear least-squares fit
of the DSC data are listed in Table 2. DSC curves showed
full reversibility at pH 4.4, but not at pH 6.5, so enthalpy
change (DH) and heat capacity changes (DC
p
) are listed only
at pH 4.4.

p
values were approximately the same
(within 5%) in the presence or absence of additives. These
results suggest that the thermodynamic equilibrium of
lysozyme is not influenced by 100 m
M
additive and that the
molecular mechanism of spermidine and spermine as
aggregation suppressors cannot be explained by the slight
change in the thermodynamic parameters.
Table 1. Residual activity of lysozyme after heat treatment.
Additive
Residual activity
(%)
a
Residual activity
(%)
b
Putrescine 15.1 ± 2.4 67.6 ± 4.8
Spermidine 52.4 ± 5.0 82.2 ± 6.1
Spermine 56.6 ± 4.3 90.5 ± 4.7
Lysine 2.3 ± 0.4 32.1 ± 6.6
Arginine 4.3 ± 1.3 30.9 ± 7.1
Glycine 1.5 ± 1.2 15.4 ± 2.8
Guanidine 0.8 ± 0.8 23.5 ± 2.9
Urea 0.8 ± 0.3 11.8 ± 3.1
NaCl 2.3 ± 0.4 13.2 ± 1.9
KCl 1.0 ± 0.7 11.8 ± 2.8
Glucose 0.9 ± 0.6 11.8 ± 1.3
Maltose 1.5 ± 0.7 13.1 ± 0.9

rate-limiting step of the heat inactivation of lysozyme is
involved in an intermolecular interaction and (b) the
resolved and refolded samples did not increase the activity
(Fig. 4A). These two facts imply that the heat-induced
inactivation of lysozyme is caused by covalent interactions
among molecules, probably disulfide reshuffling. Interest-
ingly, in the presence of spermidine and spermine, the
inactivation rates were not dependent on the protein
concentration (Fig. 4B,C). This implies that spermidine
and spermine prevent intermolecular interactions. More-
over, after heat treatment at 98 °C for 30 min, no aggre-
gates were observed in the presence of 100 m
M
spermidine
or spermine (Fig. 2A), while 50% of the molecules were
inactivated (Fig. 4B,C).
These facts propose the following mechanism, whereby
the heat-induced aggregation and inactivation of lysozyme
is considered to follow two steps of an irreversible reaction
at high temperatures:
U ! A Eqn ð1Þ
A þ A
n
! A
nþ1
Eqn ð2Þ
where, U represents the unfolded molecule that can
refold after heat treatment, A represents the irreversibly
denatured molecule and A
n

[29] and surfactants [30], were found to promote the
correct folding of proteins. When using these additives it
is important to use the appropriate nondenaturing con-
centration, but this may be difficult because the native
state is easily destabilized in the presence of these
additives.
18
On the other hand, Arg has a favorable
property – it is not a denaturant, yet it enhances the
solubility of the aggregate-prone form of unfolded protein
[10,12,31]. For this reason, Arg has been commonly used
as an aggregation suppressor. However, we report, in this
study, the new finding that spermine and spermidine are
more effective for preventing heat-induced aggregation
than other, well-known additives (Table 1).
Many researchers have reported the biological role of
polyamines in enhancing growth or cell proliferation [32,33].
Polyamines are relatively simple structures that are com-
posed of multivalent amines. The pK
a
values of the
secondary amines in putrescine, spermidine, and spermine
were 8.0–8.5, whereas those of the primary amines were
10.0–11.1 [34]. In biophysical aspects, polyamines can bind
with nucleic acids and phospholipids, and stabilize and
regulate their tertiary structures [17–19]. In this article, we
report that polyamines prevent aggregation of lysozyme, a
positively charged protein
19
(pI ¼ 11). Under acidic condi-

)1
)
b
DC
p
at pH 4.4
(kJÆmolÆK
)1
)
b
No additive 77.3 81.0 452 20.7
Putrescine 78.4 79.1 436 21.7
Spermidine 78.4 78.7 437 22.2
Spermine 78.6 78.3 448 22.8
Arginine 77.8 79.7 439 21.6
Glycine 77.7 81.3 442 20.6
Guanidine 77.1 79.1 437 22.0
NaCl 77.7 79.9 441 21.6
a
50-m
M
sodium-phosphate buffer (pH 6.5);
b
50 m
M
sodium-acetate buffer (pH 4.4).
4552 M. Kudou et al. (Eur. J. Biochem. 270) Ó FEBS 2003
a reduction of intermolecular interaction. It is worth
mentioning that hyperthermophiles, which can grow at
temperatures of > 90 °C, synthesize several kinds of

JST (Japan Science and Technology Corporation), and the Sasakawa
Scientific Research Grant from The Japan Science Society.
References
1. Fink, A.L. (1998) Protein aggregation: folding aggregates, inclu-
sion bodies and amyloid. Fold. Des. 3, 9–23.
2. Lilie, H., Schwarz. E. & Rudolph, R. (1998) Advances in refolding
of proteins produced in E. coli. Curr. Opin. Biotechnol. 9, 497–501.
3. Carrio, M.M. & Villaverde, A. (2002) Construction and decon-
struction of bacterial inclusion bodies. J. Biotechnol. 96, 3–12.
4. Bucciantini, M., Giannoni, E., Chiti, F., Baroni, F., Formigli, L.,
Zurdo, J., Taddei, N., Ramponi, G., Dobson, C.M. & Stefani, M.
(2002) Inherent toxicity of aggregates implies a common
mechanism for protein misfolding diseases. Nature 416, 507–511.
5. Chiti, F., Calamai, M., Taddei, N., Stefani, M., Ramponi, G. &
Dobson, C.M. (2002) Studies of the aggregation of mutant pro-
teins in vitro provide insights into the genetics of amyloid diseases.
Proc.NatlAcad.Sci.USA99, 16419–16426.
6. Rozema, D. & Gellman, S.H. (1995) Artificial chaperones: protein
refolding via sequential use of detergent and cyclodextrin. J. Am.
Chem. Soc. 117, 2373–2374.
7. Altamirano, M.M., Golbik, R., Zahn, R., Buckle, A.M. & Fersht,
A.R. (1997) Refolding chromatography with immobilized mini-
chaperones. Proc.NatlAcad.Sci.USA94, 3576–3578.
8. Izumi, M., Fujiwara, S., Shiraki, K., Takagi, M., Fukui, K. &
Imanaka, T. (2001) Utilization of immobilized archaeal chaper-
onin for enzyme stabilization. J. Biosci. Bioeng. 91, 316–318.
9. Buchner, J. & Rudolph, R. (1991) Renaturation, purification and
characterization of recombinant Fab-fragments produced in
Escherichia coli. Biotechnology 9, 157–162.
10. Rudolph, R. & Lilie, H. (1996) In vitro folding of inclusion body

Kinetic determinism of lysozyme folding at high temperatures.
Biochem. Biophys. Res. Commun. 56, 380–405.
21. Wetlaufer, D.B. & Saxena, V.P. (1970) Formation of three-
dimensional structure in proteins. I. Rapid nonenzymic reactiva-
tion of reduced lysozyme. Biochemistry 9, 5015–5023.
22. Hagihara, Y., Shiraki, K., Nakamura, T., Uegaki, K., Takagi, M.,
Imanaka, T. & Yumoto, N. (2002) Screening for stable mutants
with amino acid pairs substituted for the disulfide bond between
residues 14 and 38 of bovine pancreatic trypsin inhibitor (BPTI).
J. Biol. Chem. 277, 51043–51048.
23. Volkin, D.B. & Klibanov, A.M. (1987) Thermal destruction
processes in proteins involving cystine residues. J. Biol. Chem. 262,
2945–2950.
24. Tomizawa, H., Yamada, H., Tanigawa, K. & Imoto, T. (1995)
Effects of additives on irreversible inactivation of lysozyme at
neutral pH and 100 degrees C. J. Biochem. (Tokyo) 117, 369–373.
25. Tomizawa, H., Yamada, H., Wada, K. & Imoto, T. (1995) Sta-
bilization of lysozyme against irreversible inactivation by
suppression of chemical reactions. J. Biochem. (Tokyo) 117,635–
640.
26. Maachupalli-Reddy, J., Kelly, B.D. & Clark, E.D.B. (1997) Effect
of inclusion body contaminants on the oxidative renaturation of
hen egg white lysozyme. Biotechnol. Prog. 13, 144–150.
27. Ueda, T., Nagata, M. & Imoto, T. (2001) Aggregation and che-
mical reaction in hen lysozyme caused by heating at pH 6 are
depressed by osmolytes, sucrose and trehalose. J. Biochem.
(Tokyo) 130, 491–496.
28. Zale, S.E. & Klibanov, A.M. (1986) Why does ribonuclease irre-
versibly inactivate at high temperatures? Biochemistry 25, 5432–
5444.