Biochemical and Biophysical Research Communications 266, 366 –370 (1999)
Article ID bbrc.1999.1804, available online at http://www.idealibrary.com on
Lysozyme Association with Nucleic Acids
L. K. Steinrauf, 1 David Shiuan, Wen-jen Yang, and Michael Y. Chiang
Departments of Biological Science and Chemistry, National Sun Yat-sen University,
70 Lien-hai Road, Kaohsiung 80424, Taiwan, Republic of China
Received October 6, 1999
Lysozyme is well known for the ability to hydrolyze
the cell wall of bacteria. Based on the similarity of
structure between lysozyme and histones as seen from
the results of X-ray crystal structure determinations,
we have postulated that binding to nucleic acids may
be another biological function of lysozyme. We have
therefore begun a systematic study of the interactions
of lysozyme and related molecules with nucleic acids,
and present here a preliminary report. Binding to
DNA and RNA has been demonstrated from gel electrophoresis, enzyme activity, and coprecipitation
studies. We suggest that this function of lysozyme will
provide an explanation why Lee-Huang et al. (1999)
[Proc. Natl. Acad. Sci. USA 96, 2678 –2681] were able to
call lysozyme a “killer protein” against the AIDS virus,
and may provide a new avenue of research on AIDS
therapy. © 1999 Academic Press
phy in 1962 (7, 8, 9). HEW lysozyme has been crystallized with several different anions (10), and recently two crystal structure determinations at
atomic resolution have been reported (11, 12) in
which the anions, acetate, nitrate, and iodide, have
been well identified. The recent report by Lee-Huang
et al. (13), that lysozyme and ribonuclease were
“killer proteins” against HIV prompted us to investigate the possibility that lysozyme might be able to
interact with DNA and/or RNA. A comparison using
computer graphics of the crystal structures of lysozyme with those of histones suggested to us the
possibility that lysozyme might have some histonelike properties. As the first part of our investigation
we have used gel electrophoresis, enzyme activity,
and co-precipitation to demonstrate the action of
lysozyme on nucleic acids.
MATERIALS AND METHODS
Ever since the discovery by Fleming (1) in 1922,
lysozyme was been one of the most intensely studied of
proteins. With a molecular weight of about 14,400, the
tertiary structure is compact with several helices surrounding a small beta sheet region. The protein is very
stable and soluble, the many arginine residues giving a
net positive charge. Lysozyme is well known to hydrolyze the cell wall material of Gram negative bacteria as
described in a 1996 review (2) in which there is no
mention of any other enzymatic functions. Scattered in
the literature there are hints of other functions such as
binding to lactoferrin (3), crosslinking to DNA by
N-acetoxy-2-acetylaminofluorene (4), sedimentation
with DNA (5), and DNA-membrane association (6).
Lysozyme is produced by the cytoplasmic granules
of most cells and is present in all body fluids, blood at
0.5–2.0 mg/ml and saliva at 2.0 –5.0 mg/ml (2). Hen egg
white (HEW) lysozyme was the third protein and the
first enzyme to be determined by x-ray crystallogra1
Professor emeritus, Indiana University School of Medicine, Indianapolis, Indiana, and visiting professor, Office of Research Affairs,
National Sun Yat-sen University, Kaohsiung, Taiwan. E-mail:
[email protected].
0006-291X/99 $30.00
Copyright © 1999 by Academic Press
All rights of reproduction in any form reserved.
Materials. Chicken egg-white lysozyme was purchased from
Worthington Biochemicals, Inc. (Lakewood, NJ) and from Sigma
Chemical Co. (St. Louis, MO). Human lysozyme from macrophage,
a-lactalbumin, bovine serum albumin, histone type II-S from calf
thymus, poly-lysine (15,000 –500,000 mw) hydrobromide, and lyophilized cells of Micrococcus lysodeikticus were also purchased from
Sigma, and all were used without further purifications. The proteins
were made up in stock solutions of 10 mg/ml and kept frozen until
used. Serial dilutions were made by a factor of 2 or 3 to give solutions
down to 0.1 mg/ml. Hen egg white was taken from fresh eggs obtained locally.
Plasmid pUC18, approximately 2.7 kb, had been prepared according to published procedures (14). The HindIII, BamhI, and DdeI
restriction enzymes; the Lambda DNA-HindIII digest ladder, and
the RNA ladder were purchased from New England Biolabs, Inc.
(Beverly, MA). The single strand DNA ladder from calf thymus was
purchased from Sigma.
Methods. Gel electrophoresis was carried out on 0.8% agarose
slabs, using 13 HEA buffer, pH 8.0 (0.04 M Tris-acetate, 0.001 M
EDTA): some runs were made at pH 4.0. The apparatus used was the
Run One electrophoresis cell (Embl Tech, San Diego, CA). The gel
loading buffer was 30% glycerol with 0.1% bromo phenol blue. Runs
were at room temperature at 100 volts for about 20 min. The finished
electrophoresis gels were stained with 1% ethidium bromide for 10
min, and examined by UV light in a Ultra Violet Transilluminator,
Model MUVB-20D (Ultra-Lum, Inc., Paramont, CA). Centricon filters were obtained from Amicom, Inc. (Beverly, MA).
366
Vol. 266, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
The enzyme assay for lysozyme followed the procedure of Shugar
(15). Dried cells of Micrococcus lysodeikticus were suspended in 0.1
M Tris buffer, pH 8.0, to give an OD of approximately 0.7. To 1 ml of
cell suspension, 2 to 10 mg of lysozyme was added. The activity was
determined by following the decrease of turbidity at 450 nm on a
Model 7800 spectrophotometer (Jasco Spectrophotometer Co., Hachioji City, Tokyo, Japan).
The results were plotted with turbidity as a function of time. The
points were fitted satisfactorily by a modified first order kinetic
equation, expressing three contributions to the turbidity, 40% from
cells that were rapidly removed, 40% from cells that were slower by
a factor of 3, and 20% from cells that were removed very slowly or not
at all.
OD 5 OD0 / 2 3 exp~20.5 3 k 3 t!
1 OD0 / 2 3 exp~21.5 3 k 3 t! 1 TR,
where OD 0 is the original turbidity less the residual turbidity, T R, k
is the rate constant, and t is time.
The lysozyme-DNA complex was obtained from an excess of lysozyme added to 0.2 mg/ml pUC18/HindIII DNA in 0.5 ml of 0.02 M
Tris buffer, pH 8.0. Turbidity appeared almost immediately. The
material was centrifuged at 1000 3 g for 5 min. The supernatant was
removed and the excess lysozyme was measured by OD at 280 nm.
The pellet was washed with one volume of distilled water and again
centrifuged. The OD of the wash was negligible. The packed precipitate readily dissolved completely in one volume of 300 mM sodium
chloride solution. The UV spectrum was a mixture of protein and
nucleic acid. The solution was passed through a Centricon 50 filter,
but the OD of the filtrate was negligible.
RESULTS
Electrophoresis
DNA. When a constant amount (usually 1.0 mg) of
DNA was mixed with increasing amounts of lysozyme
and electrophoresed, the bands of the DNA began to
fade at about an equal weight of lysozyme, and were
completely gone at higher amounts, with nothing appeared under UV light upon staining with ethidium
bromide. Two important points were noticed. As the
FIG. 1. (A) Gel electrophoresis of intact pUC18 plasmid DNA in
the presence of decreasing amounts (8.0 mg/ml to 0.12 mg/ml) of
lysozyme. (B) Effects of lysozyme on the gel electrophoresis of plasmid pUC18, linearized with restriction enzyme HindIII.
FIG. 2. Effects of lysozyme on the gel electrophoresis of l DNA
digested with the restriction enzyme HindIII, (A) 1.0 mg of l DNA,
(B) 0.3 mg of l DNA.
amount of lysozyme in the mixture was increased, the
bands of the DNA had the same migration as without
lysozyme. Moreover, all of the bands were removed by
the same amount, that is, lysozyme did not show any
selectivity. The loss of the DNA bands are shown in
Figs. 1A, 1B, 2A, and 2B.
One explanation for the loss of the DNA without
any selectivity might be that the positively charged
lysozyme was neutralizing the negative charge on
the DNA, resulting in a loss of solubility so that the
DNA would precipitate rather than migrate in the
electrical field. Indeed, when DNA solutions with an
excess of lysozyme were examined, they appeared
slightly turbid. However, simple calculations (details
given below) showed that complete electrical neutralization of 1.0 mg of DNA should occur at about 5.5
mg of lysozyme.
It was necessary to address the question whether the
lysozyme was digesting the DNA into fragments too
small to show by the electrophoresis procedure. The
other possibility was that the lysozyme may be contaminated by a DNase. In order to disprove these possibilities, the DNA ladder must be recovered and analyzed
after treatment with lysozyme. Therefore, the DNA
and lysozyme were mixed together, and 5 min later a
1⁄3 volume of saturated guanidinium chloride was
added and the mixture was electrophoresed. The DNA
now appeared (data not shown). The DNA could also be
recovered by urea plus heating or by the addition of a
1⁄3 volume of saturated sodium chloride. Also, electrophoresis at pH 4.0, which would be out of the range of
pH activity of DNase, gave much the same results as a
pH 8.0.
Another possibility was that the lysozyme was destroying the fluorescence from the DNA with ethidium
bromide. To investigate this, the lysozyme was not
mixed with the DNA but was introduced from a well at
the opposite end of the electrophoresis gel slab. Since
367
Vol. 266, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
lysozyme has a very positive charge, it would run in the
opposite direction as DNA. Indeed, in the middle of the
gel slab, the lysozyme met the advancing DNA and a
strongly fluorescent band was formed after staining
with ethidium chloride (data not shown).
In addition to the intact plasmid pUC18 DNA, the
plasmid was cut with the restriction enzymes, BamhI,
HindIII, and DdeI, and lambda DNA cut with HindIII
were used. The pattern for plasmid pUC18 is shown in
Fig. 1A, pUC18 plasmid linearized with Hind III (one
cut giving a single band of about 2.7 kb) in Fig. 1B. The
lysozyme concentration at which all of the DNA pattern was lost differed by about a factor of 2. Figures 2A
and 2B show the gel electrophoresis of different concentrations of lambda phage DNA cut with HindIII
(several bands from about 0.5 to 23 kb), and 1.0 mg
lambda DNA is shown in Fig. 2A and with 0.3 mg of the
same DNA is shown in Fig. 2B, in which the ratio of
lysozyme to DNA where the pattern disappears is essentially the same.
Results for single strand DNA, for RNA, and for
pUC18 cut with BamHI, and DdeI (data not shown),
were very similar to those for pUC18 cut with HindIII.
Other possible DNA binding materials, such as bovine serum albumin and alpha-lactalbumin were tried,
but had no effect on the electrophoresis of the DNA
ladder. Lysozyme, which had first been denatured by
urea, was without effect, which showed that positive
charge alone is not enough. On the other hand, polylysine would remove the DNA from electrophoresis
quite efficiently (data not shown). At pH 8.0 poly-lysine
would probably show some alpha helix and have some
resemblance to lysozyme.
Human lysozyme was tested with DNA, and gave the
same results as did HEW lysozyme. Human histone
was tested and found to bind DNA in much the same
way as did lysozyme. Egg white was also able to remove DNA at up to 1/30 dilution (data not shown).
Enzyme Activity
Lysozyme, 10 ml of a 1.0 mg/ml solution in distilled
water, was added to 1 ml of a cell suspension in 0.1 M
Tris buffer, pH 8.0. The OD at 450 nm fell from about
0.8 to about 0.2 in three minutes. When an equal
weight of lambda DNA was mixed with the lysozyme
before addition to the cells, the OD did not change.
When the DNA was added to the cell suspension, and
then the lysozyme added, again there was no change of
the OD. Figure 3 shows the results of adding increasing amounts (up to 5 mg) of lambda phage DNA to the
cell suspension before the addition of 10 mg of lysozyme. The 5 mg of DNA seemed to give about a 50%
inhibition of the 10 mg of lysozyme. The points were
reasonably well fitted by the two-term first order kinetic equation given above. These results suggest but
do not prove that DNA is binding very quickly to the
FIG. 3. Decrease of the optical density (OD) with time (in seconds) as cells are digested with lysozyme which had been inhibited
by increasing amounts of DNA of l/HindIII.
active site of lysozyme. With 5 mg of pUC18 plasmid
DNA, either the intact plasmid or the linearized plasmid, were added to the cell suspension and then the 10
mg of lysozyme added, there was very little effect from
the DNA. It can be estimated that the plasmid is from
10 to 15 times less effective in inhibiting lysozyme than
is lambda DNA. Thus it appears that enzyme activity
is a more sensitive method for following the interactions between DNA and lysozyme.
RNA, which is very rapidly degraded, was not tested
in the enzyme activity assay.
Direct Precipitation
The DNA to lysozyme ratio was also estimated by
direct precipitation. When 0.05 mg of linearized pUC18
plasmid DNA and 1.0 mg of lysozyme were added to 0.5
ml of 10 mM Tris buffer, pH 8, the solution immediately became cloudy. After centrifugation at 1000 3 g
for 10 minutes, the supernatant remained clear even
after several hours. The supernatant, after appropriate
dilution, had the spectrum that would be expected from
approximately 0.71 mg of lysozyme. The precipitate
was washed with 0.5 ml of distilled water and centrifuged again, the supernatant giving negligible optical
density. The precipitate quickly and completely dissolved in 0.5 ml of 300 mM sodium chloride. The ultraviolet spectrum was that which would be expected from
a mixture of protein and nucleic acid. In theory centrifugation through a Centricon 50 filter should have
passed the lysozyme and retained the DNA. However,
only a negligible amount of protein came through. This
suggests that there is more than a simple electrostatic
association between the DNA and lysozyme.
Under the same conditions, the precipitation of intact pCU18 plasmid DNA with lysozyme gave different
results. The amount of precipitate was less and the
supernatant had an ultraviolet spectrum that would be
characteristic of a mixture of protein and DNA. Again,
the precipitate dissolved readily in 300 mM sodium
368
Vol. 266, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
TABLE 1
Summary of the Protein Binding of Nucleic Acids
DNA-binding
materials
Non-DNA-binding
materials
HEW lysozyme
Human lysozyme
Polylysine
Histone
Raw egg white
Denatured lysozyme
a-Lactalbumin
Bovine serum albumin
Nucleic acids used
l ladder DNA
pUC18 phage DNA
pUC18/HindIII DNA
pUC18/BamHI DNA
pUC18/DdeI DNA
Single strand DNA
RNA
chloride and the resulting solution would not pass
through a Centricon 50 filter.
DISCUSSION
The results of the gel electrophoresis show that the
interaction between lysozyme and nucleic acid is basically the electrostatic attraction between the positively
charged lysozyme and the negatively charged nucleic
acid. The enzyme activity observations show that not
all nucleic acids are the same with respect to lysozyme,
and the co-precipitation results show that the conformation of the DNA is important (Table 1).
It has been shown that the loss of DNA in the electrophoresis pattern is not the consequence of contamination by DNase, or from some intrinsic DNase activity of lysozyme. The interaction of lysozyme with DNA
must therefore be physical, since the DNA can be recovered by the denaturation of the lysozyme, or by the
addition of high salt concentration, a classical way to
dissociate two polymers of opposite charges. The simplest assumption is that the electrical charged groups
on lysozyme and DNA must balance and neutralize
each other. If it is assumed that the charge on DNA
comes only from fully ionized phosphate, then DNA
would have one negative charge per 350 molecular
weight. The charge on lysozyme is more directly observed from x-ray crystallography (11). Each molecule
of lysozyme has 8 excess positive charges per molecular
weight of 14,400, or one charge for 1,800 molecular
weight. This corresponds to one lysozyme per 4 base
pairs of DNA, which is getting rather crowded. Thus 1
mg of DNA would require about 5 mg of lysozyme for
complete neutralization.
The three methods used to determine the lysozyme
to DNA ratio were based on different properties, but
the results were fairly close. Gel electrophoresis gave 1
mg of DNA per 2 mg of lysozyme, the enzyme activity
required 1 mg of DNA to every 1 mg of lysozyme, and
the co-precipitation had 1 mg of DNA for each 6 mg of
lysozyme. From this one must conclude that the methods are measuring the association of lysozyme with
DNA, although perhaps not in the same way.
The precipitate of lysozyme with DNA could be
readily dissolved in 300 mM sodium chloride or magnesium sulfate. From the calculations, the amount of
lysozyme in the precipitate is almost enough to neutralize the charge on the DNA completely. However,
the association is not just electrostatic, because the
dissolved precipitate would not be separated by Centricon filtration.
If lysozyme were binding at random to DNA molecules, then the expected result would be to lessen the
effective negative charge on the DNA. The consequence
of this would be that all of the DNA molecules would
migrate, albeit slower. However, it does not happen
this way. It seems that lysozyme will completely remove a DNA molecule from migration, an all-or-none
situation. One possible explanation could be that the
first molecule of lysozyme to bind to a DNA molecule
provides a cross-link to the second DNA, which would
provide easier binding of the second molecule of lysozyme. By such a cooperative effect an insoluble polymer would be formed, which would be eliminated from
movement by electrophoresis.
At present it is not possible to say whether lysozyme
makes a profound change to the physical state of the
DNA, such as changing the curvature, such that the
DNA molecule is now much more susceptible to the
binding by more lysozyme, and has greatly increased
frictional properties. Obviously, much more investigation is needed.
The results from enzyme activity are not accurate
enough to say what kind of inhibition DNA makes on
lysozyme. The long binding cleft that is known for the
hydrolyase activity of lysozyme quite possibly could
double for DNA binding as well. Lysozyme shows pairs
of arginine residues that come very close together,
potential binding sites for the phosphate of DNA.
It is tempting to speculate that lysozyme may have a
functional role in the regulation of DNA and RNA in
the cytoplasm and in the extracellular environment.
The recent report by Lee-Huang et al. (13) showed that
lysozyme and RNAase could be found in the human
chorionic gonadotropinis, and that lysozyme and
RNAase could protect ACH2 lymphocytes and U1
monocytes from HIV infection as monitored by p24
protein ELAIS assay.
Lee-Huang et al. (13) have called lysozyme a “killer
protein” against the AIDS virus. The present work
would extend this concept to suggest the possibility
that lysozyme may have a broader protective spectrum
than just the ability to hydrolyze the cell walls of Gram
negative bacteria. Specifically, lysozyme may be a defense mechanism, perhaps a very ancient mechanism,
against virus attack that could be active extracellularly or in the cytoplasm, and could possibly operate on
the processes of transcription and of replication. It has
been shown that lysozyme comes from what is probably
a very ancient gene (16). If such be the case, the anti-
369
Vol. 266, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
bacterial function and the anti viral function have been
evolving together for a long time. It is also probably no
coincidence that there is great similarity in the appearance from X-ray crystallography of lysozyme and the
histones.
CONCLUSIONS
It has been shown that lysozyme will interact with
DNA and RNA in a process that prevents the nucleic
acid from moving under electrophoresis, a process that
seems to be relative independent of the nature of the
nucleic acid. It has also been shown that DNA will
inhibit the enzymatic function of lysozyme and that
this process is strongly dependent on the nature of the
DNA. Moreover. The results of co-precipitation have
shown that the conformation of the DNA is important
in determining how much DNA binds to lysozyme.
The results are consistent with lysozyme binding to
the double helix conformation of DNA, and that the
binding of lysozyme molecules can render a long length
of double strand helix unavailable for electrophoretic
mobility. At the same time the binding of DNA to
lysozyme at a ratio of about 20 base pairs per molecule
of lysozyme is required to inhibit the enzyme activity.
ACKNOWLEDGMENTS
Support for this work has been provided by Grants NSC 87-2622110-001 and NSC 87-2311-B110-008 from the National Science
Council, Republic of China.
REFERENCES
1. Fleming, A., and Allison, V. D. (1922) Proc. R. Soc. London B39,
306 –317.
2. Jollès, P. (Ed.) (1996) in Biochemistry and Biology, Birkhauser
Verlag, Basel.
3. van Berkel, P. H., Geeris, M. E., van Veen, H. A., Mericskay, M.,
de Boer, H. A., and Nuijens, J. H. (1997) Biochem. J. 15, 145–
151.
4. Werbin, H., and Cheng, C. (1985) Carcinogenesis 6, 1689 –1691.
5. Stoliarova, G. S. (1977) Mol. Biol. (Mosk.) 11, 389 –393.
6. Silberstein, S., and Inouye, M. (1974) Biochim. Biophys. Acta
366, 149 –158.
7. Blake, C. C. F., Fenn, R. H., North, A. C. T., Phillips, D. C., and
Polijak, R. J. (1962) Nature (London) 196, 1173–1175.
8. Stanford, R. H., Marsh, R. E., and Corey, R. B. (1962) Nature
(London) 96, 1176 –1178.
9. Dickerson, R. E., Reddy, J. M., Pinkerton, M., and Steinrauf,
L. K., (1962) Nature (London) 196, 1178.
10. Steinrauf, L. K. (1957) Acta Cryst. 12, 77–79.
11. Steinrauf, L. K. (1998) Acta Cryst. D54, 767–779.
12. Walsh, M. A., Schneider, T.A., Sieker, L. C., Dauter, Z., Lamzin,
V. S., and Wilson, K. S. (1998) Acta Cryst. D54, 522–546.
13. Lee-Huang, S., Huang, P. L., Sun, Y., Kung, H.-f., Blithe, D. L.,
and Chen, H. C. (1999) Proc. Natl. Acad. Sci. USA 96, 2678 –
2681.
14. Sanbrook, J., Fritsch, E. F., and Maniatis, T. (Eds.) (1989) Molecular Cloning, Laboratory Manual, Vol. I, 2nd ed. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, NY.
15. Shugar, D. (1952) Biochim. Biophys. Acta. 8, 302–305.
16. Prager, E. M. (1996) in Lysozymes: Model Enzymes in Biochemistry and Biology (Jollès, P., Ed.). Birkhauser Verlag, Basel.
370