Accepted Manuscript Title: Bacteriophage phi11 lysin: physicochemical characterization and comparison with phage phi80␣ lysin Author: Lyubov Y. Filatova David M. Donovan Juli Foster-Frey Vladimir G. Pugachev Natalia F. Dmitrieva Tatiana A. Chubar Natalia L. Klyachko Alexander V. Kabanov PII: DOI: Reference:

S0141-0229(15)00056-3 http://dx.doi.org/doi:10.1016/j.enzmictec.2015.03.005 EMT 8740

To appear in:

Enzyme and Microbial Technology

Received date: Revised date: Accepted date:

10-12-2014 26-3-2015 30-3-2015

Please cite this article as: Filatova LY, Donovan DM, Foster-Frey J, Pugachev VG, Dmitrieva NF, Chubar TA, Klyachko NL, Kabanov AV, Bacteriophage phi11 lysin: physicochemical characterization and comparison with phage phi80rmalpha lysin, Enzyme and Microbial Technology (2015), http://dx.doi.org/10.1016/j.enzmictec.2015.03.005 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Highlights

2

A physicochemical investigation of lysins of phages phi11 and phi80α was carried out.

3

Me++ ions and cysteine residues maintain the active conformation of phage phi11 lysin.

4

Lysin of phage phi11 is much more stable than lysin of phage phi80α.

5

Schemes of inactivation of lysins of phages phi11 and phi80α are suggested.

cr

ip t

1

Ac ce p

te

d

M

an

us

6

Page 1 of 36

6

Bacteriophage phi11 lysin: physicochemical characterization and comparison with

7

phage phi80α lysin

8

Lyubov Y. Filatovaa*, David M. Donovanb, Juli Foster-Freyb, Vladimir G. Pugachevc,

10

Natalia F. Dmitrievad, Tatiana A. Chubara, Natalia L. Klyachkoa,e, Alexander V.

11

Kabanova,f

cr

ip t

9

12 13

a

14

University, Moscow, Russia.

15

b

16

NEA, ARS, USDA, Beltsville, MD, USA

17

c

18

d

19

e

20

f

21

Eshelman School of Pharmacy, University of North Carolina at Chapel Hill, USA.

22

*Corresponding author: telephone number: +74959393476; fax number: +74959395417; e-

23

mail address: [email protected] (Lyubov Y. Filatova)

us

an

M

Animal Biosciences and Biotechnology Laboratory, Beltsville Agricultural Research Center,

FBSE SRC of Virology & Bioengineering «Vector», Novosibirsk, Russia.

d

I.M. Sechenov First Moscow State Medical University, Moscow, Russia.

te

Pirogov Russian National Research Medical University (RNRMU), Moscow, Russia

Division of Molecular Pharmaceutics, Center for Nanotechnology in Drug Delivery, UNC

Ac ce p

24

Department of Chemical Enzymology, Faculty of Chemistry, M.V. Lomonosov Moscow State

Page 2 of 36

Abstract

25

Phage lytic enzymes are promising antimicrobial agents. Lysins of phages phi11

26

(LysPhi11) and phi80α (LysPhi80α) can lyse (destroy) cells of antibiotic-resistant strains of

27

Staphylococcus aureus. Stability of enzymes is one of the parameters making their practical

28

use possible. The objectives of the study were to investigate the stability of lysins of phages

29

phi11 and phi80α in storage and functioning conditions, to identify optimum storage

30

conditions and causes of inactivation. Stability of the recombinant LysPhi11 and LysPhi80α

31

was studied using turbidimetry. CD-spectroscopy, dynamic light scattering, and

32

electrophoresis were used to identify causes of inactivation. At 37°C, pH 7.5 and

33

concentration of NaCl not higher than 150 mM, LysPhi11 molecules contain a high

34

percentage of random coils (43%). However, in spite of this the enzyme has high activity

35

(0.4-0.8 OD600nms-1mg-1). In storage conditions (4°C and 22°C, pH 6.0-9.0, 10-500 mM NaCl)

36

LysPhi11 is inactivated by a monomolecular mechanism. The optimum storage conditions for

37

LysPhi11 (4°C, pH 6.0-7.5, 10 mM NaCl) were selected under which the time of the enzyme

38

half-inactivation is 120-160 days. LysPhi80α stability is insufficient: at 37°C the enzyme

39

loses half of its activity almost immediately; at 4°C and 22°C the time of half-inactivation of

40

LysPhi80α varies in the range from several hours to 3 days. Despite the common properties in

41

the manifestation of antistaphylococcal activity the kinetic behavior of the enzymes is

42

different. LysPhi11 is a more promising candidate to be used as an antimicrobial agent.

44

cr

us

an

M

d

te

Ac ce p

43

ip t

24

Highlights

A physicochemical investigation of lysins of phages phi11 and phi80α was carried out. Me++ ions and cysteine residues maintain the active conformation of phage phi11

45 46

lysin.

47

Lysin of phage phi11 is much more stable than lysin of phage phi80α.

48

Schemes of inactivation of lysins of phages phi11 and phi80α are suggested.

Page 3 of 36

Keywords: bacteriolytic enzymes, phages, staphylococcal infections, activity, stability,

49 50

structure Abbreviations:1

52

1. Introduction

53

Gram-positive Staphylococcus aureus causes a large number of dangerous (often fatal)

54

diseases of the human body, such as pneumonia, osteomyelitis, endocarditis, toxic shock, and

55

sepsis [1]. Furthermore, S. aureus rapidly develops resistance to a broad spectrum of

56

antibiotics used in clinical medicine, including reserve antibiotics. Mortality from infections

57

caused by methicillin-resistant (MRSA) and vancomycin-resistant (VRSA) strains of

58

staphylococci reached 20 thousand people in 2008 in the United States [2]. In addition to

59

controlling of the use of over-used antibiotics there is also a need to develop alternative

60

methods of antibacterial treatment. One reasonable alternative is the use of phage lytic

61

enzymes (lysins, lysozymes) which are employed by the phages during the initial and final

62

stages of their life cycle [3-7].

te

d

M

an

us

cr

ip t

51

Currently, a large number of lysins for Gram-positive bacteria are reported in

64

recombinant form and a strategy to expose the purified protein to the target pathogen and

65

obtain lysis (lysis from without) is well known. There are several advantages of phage lysins,

66

such as their near-species selectivity, their fundamentally new mechanism of acting externally

67

(avoiding intracellular resistance mechanisms); activity against antibiotic-resistant strains of

68

microorganisms, and the low probability of the emergence of resistant pathogens due to

69

coevolution of host and phage [3-5].

Ac ce p

63

1

LysPhi11 – phage phi11 endolysin; LysPhi80α – phage phi80α endolysin; MRSA – methicillin-resistant Staphylococcus aureus; VRSA – vancomycin-resistant Staphylococcus aureus; OD600nm – optical density at 600 nm; PPB – 20 mM potassium phosphate buffer with pH 7.5; PPB-NaCl –20 mM potassium phosphate buffer, pH 7.5, 150 mM NaCl; DLS – dynamic light scattering, CD – circular dichroism; kin – first-order inactivation constant; ksec-in – second-order inactivation constant; n – inactivation order; effective hydrodynamic radius, s – second.

τ

1/2

– half-inactivation time; Rh –

Page 4 of 36

Lysins of phages phi11 (LysPhi11, GenBank accession number YP_500516.1) and

71

phi80α (LysPhi80α, ABF71642.1) are monosubunit proteins with a molecular weight of ~54

72

kDa, which molecules contain two catalytic domains (CHAP, Amidase 2) and one C-terminal

73

cell wall binding domain (SH3_5). LysPhi11 and LysPhi80α hydrolyze chemical bonds in the

74

peptidoglycan through the activities of both types D-Ala-Gly-endopeptidase and N-

75

acetylmuramyl-L-Ala amidase [6-8]. A common property of these enzymes is that LysPhi11

76

and LysPhi80α reveal the same activity in the live cells lysis of more than 30 strains of

77

Staphylococcus aureus. In addition, these enzymes exhibit a similar efficacy in the lysis of

78

staphylococcal biofilms [8]. Apart from the microbiological experiments described above [6-

79

8] an enzymological approach was applied to evaluate the applicability of lysins of phages

80

phi11 and phi80α.

M

an

us

cr

ip t

70

As far as both theory and practice are concerned, it is important that the enzymes not

82

only reveal the lytic activity, but can also maintain it for long periods of time (or manifest

83

stability). Lysins of phages phi11 and phi80α, likewise most biocatalysts, require specific

84

storage and transportation conditions and can be subject to various inactivating effects when

85

functioning. Currently neither optimum storage conditions of these enzymes have been

86

identified nor their stability has been estimated.

te

Ac ce p

87

d

81

The objective of this study was to carry out a physicochemical investigation of lysins

88

of phages phi11 and phi80α. It contained the study of the lysins activity and stability in

89

physiological conditions; the study of storage stability of the enzymes and identifying of the

90

optimum storage conditions; the determination of the enzymes inactivation causes and the

91

comparison of inactivation parameters.

92

2. Materials and methods

93

2.1. Materials

Page 5 of 36

Recombinant lysins (C-terminally 6xHis tagged proteins) were prepared according to

95

the procedures reported previously [8]. LysPhi11 (19.0 mg/mL) and LysPhi80α (18.0 mg/mL)

96

were stored at -20°C in an elution buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM

97

imidazole, 30% glycerol, рН 8.0).

ip t

94

The culture of Staphylococcus aureus bacteria (strain 209 B-580) was grown in L-

99

broth up to the exponential phase (4.0x108 cells/mL), was titrated to determine the cells

100

concentration, and was autoclaved during 30 minutes at a pressure of 0.6-0.7 atm. The cooled

101

suspension was precipitated by centrifugation at 9.0x103 rpm during 20 minutes. The

102

precipitate was washed twice with physiological saline and dissolved in 1/20 of the original

103

volume of the saline. The resulting suspension was examined for the presence of live bacteria

104

by plating the samples in an agar nutrient medium.

M

an

us

cr

98

Due to the need of eliminating the use of infectious material in the chemical

106

laboratory, a special culture of Staphylococcus aureus cells was to be prepared: the bacteria

107

should lose the ability to grow and reproduce, but the cells (cell wall peptidoglycan) should

108

not be disrupted (Fig. S1 A). Apart from that the methods of heat treatment of the cells used

109

should not cause a change in the enzyme activity (Fig. S1 B). The method of autoclaving of

110

Staphylococcus aureus cells was selected subject to the requirements of these two conditions.

111

The use of the autoclaved Staphylococcus aureus cells ensures reproducibility of the test

112

results of the enzymes preparations activity that allows to objectively evaluate both

113

techniques to prepare them and the kinetic parameters of inactivation.

te

Ac ce p

114

d

105

To prepare the buffer solutions the following was used: Tris, potassium

115

dihydrophosphate and potassium citrate, sodium chloride and sodium hydroxide, and

116

hydrochloric acid produced by Sigma. All measurements using instrumental methods were

117

performed in triplicate.

118

2.2. Measurement of LysPhi11 activity

Page 6 of 36

Enzyme activity was estimated in a turbidity reduction assay from time-dependent

120

turbidity changes in a suspension of Staphylococcus aureus cells. The cell suspension of S.

121

aureus with OD600nm, which equals to 0.6 absorbance units, was prepared using PPB. 10-20

122

µL of the enzyme solution (0.2-0.8 mg/mL) was added to 0.5 mL of the cell suspension

123

(content of LysPhi11 solution was within limits of 2-4% of the final reaction volume).

124

Measurement of the optical density change over time was determined on the

125

spectrophotometer Jenway 6405 UV/Vis at 37°C. The initial reaction rates typical for the

126

catalytic activity of LysPhi11 were determined based on the initial linear regions of the

127

kinetic curves of the optical density (OD600nm) versus time (s) (Fig. S1 B).

cr

us

an

129

2.3. Investigation of the influence of salt concentration in solution media on the LysPhi11 activity at 37°C

M

128

ip t

119

The cell suspension of S. aureus with OD600nm = 0.6 absorbance units was prepared

131

using PPB with NaCl concentration of 10-1000 mM. 10-20 µL (0.4 mg/mL) of the enzyme

132

was added to 0.5 mL of the cell suspension, followed by measurement of turbidity at 37°C.

te

d

130

2.4. Study of LysPhi11 inactivation at 37°C

134

The thermal stability of LysPhi11 was studied at 37°C in two different buffers: PPB-

Ac ce p

133

135

NaCl or PPB. The enzyme (0.1-0.4 mg/mL) was pre-incubated at 37°C during selected time

136

intervals. After pre-incubation, aliquots of 10-20 µL were taken and activity measurement

137

under standard conditions was performed (2.2).

138 139

2.5. Investigation of the influence of pH, temperature and buffer on LysPhi11

stability under storage conditions

140

For the selection of the optimum storage conditions the following parameters were

141

varied: temperature (4°C or 22°C), pH (6.0-9.0), enzyme concentration (0.2-0.8 mg/mL), and

142

low-molecular weight electrolyte (NaCl, 10-500 mM). To maintain the constant pH of the

143

lysin solutions a universal buffer was used, containing 20 mM of each compound: Tris,

Page 7 of 36

144

potassium phosphate and citrate, respectively. Solutions of LysPhi11 were pre-incubated at

145

4°C and 22°C during chosen time intervals and then aliquoted for measurement of the activity

146

under standard conditions (2.2). 2.6. Electrophoresis

148

Electrophoresis of LysPhi11 samples was performed according to the procedures

149

described in [9]. SDS-PAGE electrophoregrams were analyzed using Biospectrum AC Chemi

150

HR-410.

152

cr

2.7. Investigation of the aggregation kinetics of LysPhi11 by dynamic light

us

151

ip t

147

scattering

The effective hydrodynamic radius of LysPhi11 was determined using Zetasizer Nano

154

ZS (He-Ne laser, 5 mW, 633 nm) at temperatures of 22°C and 37°C and enzyme

155

concentrations of 0.2-0.8 mg/mL.

M

an

153

2.8. Circular dichroism spectroscopy of LysPhi11

157

Circular dichroism spectra for LysPhi11 (0.4 mg/mL, PPB or PPB-NaCl) were

158

recorded at 37°C. The optical path length of the quartz cuvette was 1 mm, the measurements

159

were carried out in the far UV region (195-260 nm) on the device Jasco J815 with a scanning

160

speed of 2 nm/sec. Before recording the CD spectra enzyme solutions were filter sterilized via

161

0.22 μM Millex filter (Millipore). The predicted secondary structure elements were

162

determined via analysis of the spectra using the program CDNN [10].

164

te

Ac ce p

163

d

156

2.9. Testing reactivation of LysPhi11 with Ca++ and Mg++cations The cell suspension of S. aureus with OD600nm = 0.6 absorbance units was prepared in

165

deionized water followed by the addition of calcium and magnesium cations up to the

166

concentration of 1-10 mM. 10 µL of the inactivated enzyme solution in deionized water with

167

concentration of 0.4 mg/mL was added to 0.5 mL of the cell suspension, followed by

168

measuring of LysPhi11 activity at 37°C.

Page 8 of 36

169

2.10. Study of the inactivation of LysPhi80α

170

Study of the inactivation of LysPhi80α was held following the procedures similar to

171

those for LysPhi11 (2.2-2.7). 3. Results

173

3.1. Investigation of the influence of NaCl on LysPhi11 activity and stability at 37°C

cr

174

ip t

172

It is important to investigate the activity and stability of lytic enzymes under the

176

functioning conditions from the viewpoint of practical use. NaCl is the major component of

177

both blood plasma and tissue fluids of humans and animals. The dependence of LysPhi11

178

activity on NaCl concentration was examined under the medium conditions close to

179

physiological (37°C, pH 7.5). The activity of phage phi11 endolysin decreases with increasing

180

of NaCl concentration up to nearly zero at a salt content of 750 mM (Fig. 1). At physiological

181

concentration of NaCl (about 150 mM) LysPhi11 loses ~ 40% of its activity in relation to the

182

“salt-free” level.

te

d

M

an

us

175

The loss of the enzyme activity may be associated with structural changes caused by

184

the increase in salt concentration. LysPhi11 structure was described by CD-spectroscopy (the

185

spectra examples are shown in Fig. S2, their deconvolution results are given in Table 1). To

186

determine the causes of the secondary structure enzyme changes CD spectra were obtained at

187

two temperatures, 37°C and 22°C (one of the enzyme storage temperatures), and in two

188

buffers (PPB and PPB-NaCl). The shape of the CD spectra obtained for phage phi11

189

endolysin is typical for β-structural proteins with a minimum at 210-215 nm and a positive

190

value of molar ellipticity at 195 nm [11].

Ac ce p

183

191

Content of random coils in LysPhi11 molecule at 22°C is by ~20% lower than at 37°C

192

(Table 1). The effect of NaCl on the secondary structure of the enzyme at both temperatures

193

was evaluated by comparing the results of the enzyme CD spectra deconvolution in a medium

Page 9 of 36

194

without NaCl and in a medium with NaCl. The content of different elements of the secondary

195

structure of the enzyme molecule in the medium with NaCl is identical to that in the solution

196

without NaCl (Table 1). It is essential to emphasize that the enzyme LysPhi11 in unfolded conformation has a

198

non-zero value of the activity at 37°C (Fig. 1). LysPhi11 enzyme activity is comparable in

199

value with staphylolytic enzyme activity with an ordered structure in similar conditions

200

(37°C, pH 7.5). The activity of LysK, which molecule contains only 23% of random coils at

201

37°C, equals to 1.6 OD600nms-1mg-1 [12], which is comparable with the activity of LysPhi11

202

(0.8 OD600nms-1mg-1, Fig. 1).

us

cr

ip t

197

As shown above, the structural changes in LysPhi11 molecules at 37°C lead to the

204

unfolding of enzyme molecules. The unfolding results in additional interactions between the

205

enzyme molecules, which causes aggregation. At 22°C (a storage temperature) LysPhi11

206

particle size is 5±1 nm, the enzyme heating to 37°C causes an increase in the particle size up

207

to 100-150 nm in PPB and 300-500 nm in a medium containing NaCl. Aggregation of the

208

enzyme molecules is the reason for the decrease of LysPhi11 activity with increasing of salt

209

concentration.

M

d

te

Ac ce p

210

an

203

Changing the salt composition of the environment (introduction of sodium and

211

chloride ions) affects not only the activity, but also the mechanism of LysPhi11 inactivation at

212

37°C. In PPB-NaCl, the dependence of LysPhi11 residual activity versus time is a function of

213

the enzyme concentration (Fig. S3 A). Increase of LysPhi11 concentration from 0.1 to 0.4

214

mg/mL leads to the enzyme stability decrease (Table 2), which is to say that intermolecular

215

interactions is likely the main reason of the enzyme inactivation. The obtained dependences of

216

the residual activity versus time were straightened in the coordinates of the second order

217

equation (Fig. S3 B):

218

1 = 1 + k sec in [E] 0 t A / A0

(Eq. 1)

Page 10 of 36

219

where A/A0 is the residual activity of the enzyme, [E] 0 is the initial concentration of

220

the enzyme, ksec-in is the constant of inactivation of the second order, t is the time of the

221

enzyme to achieve the residual activity A/A0. The second order kinetics of the enzyme inactivation includes the bimolecular

223

interactions between the enzyme molecules (autolysis or aggregation). Using electrophoresis

224

under denaturing conditions showed that the molecular weights of the active and inactivated

225

enzymes are virtually identical (Fig. 2). It can be concluded that in PPB-NaCl inactivation of

226

LysPhi11is caused by non-covalent aggregation.

228

cr

us

In PPB (medium without NaCl) the rate of enzyme inactivation does not depend on its concentration (Fig. S4 A, Table 2) and is described by the first order equation (Fig. S4 B):

an

227

ip t

222

229

ln( A / A0 ) = k in t

230

where A/A0 is the residual activity of the enzyme, kin is the constant of the first order

231

inactivation, t is the time of the enzyme to achieve the residual activity A/A0. First-order

232

kinetic corresponds to enzyme unfolding.

te

d

M

(Eq. 2)

As stated above, at the initial time moment (37°C) aggregates of LysPhi11 are formed

234

with the hydrodynamic radius of 100-150 nm in PPB and 300-500 nm in PPB-NaCl. Further

235

incubation at 37°C causes an increase of the particle size of LysPhi11, wherein in PPB-NaCl

236

the process is more intense (Fig. 3). After one-hour incubation at 37°C the particle radius of

237

LysPhi11 in PPB does not exceed 400 nm, in PPB-NaCl, the effective hydrodynamic radii of

238

particles are over one micron. The temperature initially induces very fast denaturation

239

processes and the formation of initial aggregates (particles of radius less than 100 nm), which

240

further form aggregates of higher order [13]. Aggregation depends on the concentration of

241

enzyme, at higher concentrations it is more intense (Fig. 4) presumably due to increasing of

242

the frequency of collisions between molecules.

Ac ce p

233

Page 11 of 36

243 244

In the simple form, the scheme of enzyme inactivation in PPB can be depicted as follows: N→D↔A

246

In PPB-NaCl, the enzyme is inactivated by the following scheme:

247

2N→D2↔A

248

N – native, D – denatured, A – aggregated forms of LysPhi11.

249

Irreversible process of inactivation was confirmed by measurement of activity of

cr

LysPhi11 cooled solutions.

us

250

ip t

245

3.2. Stability of phage phi11 lysin under storage conditions

252

Thermal stability of enzyme is one of the most important parameters that can limit its

253

practical application. There are many ways to represent stability data, but the kinetic of

254

thermal inactivation is the most objective data that can accurately predict the stability of the

255

enzyme through any period of time. An important point in the study of thermal stability is to

256

consider a wide range of conditions e. g. pH, composition of the media, and the presence of

257

substrates and reaction products. To optimize storage conditions the influence of the

258

following factors on LysPhi11 stability was investigated: the effect of pH (6.0-9.0),

259

temperature (4°C and 22°C), ionic strength (10-500 mM NaCl) and concentration of the

260

biocatalyst (0.2-0.8 mg/mL). The range of enzyme concentrations available to study depends

261

on whether the enzymatic reaction rate can be measured. The concentration range of NaCl and

262

the pH was adjusted dependent on the possibility of non-zero enzyme activity (Fig. 1, [14]).

263

Most pharmaceutical drugs based on proteins are stored at temperatures of 1-25°C, so the

264

study of enzyme stability was held using two temperatures close to the boundary values of

265

this interval.

Ac ce p

te

d

M

an

251

266

Under storage conditions the dependence of the residual activity of LysPhi11 versus

267

time is irrespective of the enzyme concentration and can be described by a first order equation

Page 12 of 36

268

(Eq. 2). Irreversible process of inactivation was confirmed by measurement of activity of

269

LysPhi11 cooled solutions, which corresponds to different degrees of inactivation. The dependences of effective values of the constants of first order inactivation of

271

LysPhi11 on pH and salt concentration at 4°C and 22°C are shown on Fig. 5 A, B. The values

272

of the inactivation constants at 4°C are much lower than at 22°C, i.e. at lower temperatures

273

the rate of unfolding of the polypeptide chain is likely to decrease, reducing the inactivating

274

interactions.

cr

ip t

270

In the pH range of 6.0-8.0, a rise of pH does not affect the first order inactivation

276

constants (Fig. 5 A, B), but a dramatic increase was observed between pH 8.0-9.0. Unlike pH,

277

ionic strength of the solution is not a parameter of the medium that affects the inactivation

278

constant of LysPhi11 at 4°C and 22°C (Fig. 5 A, B). The concentration range of NaCl

279

required for inactivation (i.e. degradation of electrostatic interactions) is 500-1500 mM [15].

280

The salt concentrations selected in this study are insufficient to exhibit similar effects.

d

M

an

us

275

From the perspective of novel applications, the enzyme stability could be most clearly

282

estimated using the time of half-inactivation by determining the dependence of the LysPhi11

283

residual activity versus time. As Fig. 6 A shows, at 22°C, the half-inactivation time of the

284

enzyme does not depend on the NaCl concentration and decreases with increasing pH from 6-

285

8 days (5x105-7x105 s) to 1-3 days (9x104-3x105 s).

Ac ce p

286

te

281

At 4°C, the half-inactivation time of the enzyme increases with decreasing pH and

287

NaCl concentration (Fig. 6 B). The maximum time of LysPhi11 half-inactivation at 4°C (120

288

– 160 days, 1.0x107-1.4x107 s) was observed at pH 6.0-7.5 and 10 mM NaCl. These

289

conditions can be considered as optimal for the storage of the enzyme (of those examined).

290

The effective hydrodynamic radius of the molecules with enzyme activity of 100%

291

equals to 4-5nm (Fig. S5 A). After 1 day at 22°C the particle size of LysPhi11 remains 4-5

292

nm (the enzyme activity decreases to 50-80%); after 12 days at 22°C LysPhi11 particles

Page 13 of 36

enlarge to the effective hydrodynamic radius of 300-400 nm (remaining enzyme activity is

294

10-30%) and after 24 days of incubation particle size becomes 400-500 nm (with 5-15%

295

remaining activity) (Fig. S5 B, C). After 30 days of incubation at 4°C the particle size of

296

LysPhi11 equals to 4-5 nm (50-90% of enzyme activity remains); after 120 days of incubation

297

both particles with effective hydrodynamic radius of 5 nm and the formation of particles of

298

radius of 100-400 nm (with 10-60% remaining enzyme activity) are observed; after 270 days

299

under these conditions the particle radius becomes more than 1 micron (2-20% enzyme

300

activity) (data not shown). When LysPhi11 is inactivated under the storage conditions after

301

reaching zero residual enzyme activity, precipitation is observed.

us

cr

ip t

293

As the particle size of LysPhi11 remains the same after storing during 1 day at 22°C

303

and during 30 days at 4°C, with a measurable loss of activity, it is likely that the limiting step

304

is the unfolding (denaturation) of enzyme molecules, with the next step being nucleation (the

305

formation of aggregates sized up to 100 nm), which are then combined into aggregates of

306

higher order. We thus propose that the irreversible inactivation of the enzyme in the storage

307

conditions could be described by the following classical scheme:

309

M

d

te

N→D↔A

Ac ce p

308

an

302

By means of electrophoresis in denaturing conditions it was indicated that the

310

molecular weight of LysPhi11 does not change during inactivation (Fig. S6, to identify

311

formation of –S-S- bonds between LysPhi11 molecules denaturing sample buffer without 2-

312

mercaptoethanol was used). Therefore, the formation of aggregates is provided by non-

313

covalent interactions, such as electrostatic, hydrophobic, and hydrogen bonds. The inactivated

314

enzyme was partially reactivated by adding Ca++ and Mg++ cations to the reaction media to a

315

concentration >0.6 mM (Fig. 7).

316

3.3. A comparison of the kinetic properties of LysPhi11 and LysPhi80α

Page 14 of 36

The kinetic behavior of lysins of phages phi11 and phi80α at 37°C (PPB-NaCl) differ

318

both in the velocity and mechanism of inactivation. The process of LysPhi80α inactivation at

319

37°C with a sufficiently high accuracy can be described by a first order equation (Eq. 2),

320

which corresponds to a monomolecular mechanism with the inactivation constant of 0.01 s-1

321

(Fig. 8). Lysin of phage phi80α loses its half activity for less than 1 minute (60 s) which is

322

significantly lower than half-inactivation time for LysPhi11 (Table 2) inactivated under these

323

conditions according to a bimolecular mechanism.

cr

ip t

317

LysPhi80α in both storage conditions (4°C and 22°C, 10-500 mM NaCl, enzyme

325

concentration 0.2-0.8 mg/mL) and functioning conditions (37°C, PPB-NaCl) is much less

326

stable than the LysPhi11. The constants of the first order irreversible inactivation for

327

LysPhi80α are much higher than the inactivation constants of LysPhi11 (Fig. 9). The half-

328

inactivation time for LysPhi80α does not exceed 3 days (3x105 s) at 4°C and 1/4 day (2x104 s)

329

at 22°C, which is 1-2 orders of magnitude lower than the same τ1/2 for LysPhi11 (Fig. 6).

d

M

an

us

324

In addition to the enormous difference instability, another difference in the kinetic

331

behavior of the enzymes is that NaCl stabilizes LysPhi80α under storage conditions (Fig. 9).

332

The increased NaCl concentration disrupts the electrostatic interactions and enhances the

333

hydrophobic ones. Due to the increased hydrophobic interactions the compaction of the

334

protein globule might lend itself to the appearance of additional stability to unfolding,

335

resulting in a decrease of the constants of monomolecular process.

Ac ce p

336

te

330

In solutions of partially or completely inactivated LysPhi80α, as well as LysPhi11, the

337

formation of aggregates of denatured molecules with the effective hydrodynamic radius from

338

several hundred nanometers to 1-2 microns is observed (data not shown). The difference is

339

likely due to the aggregation rate, which is often caused by the difference in denaturation rate

340

of enzymes.

Page 15 of 36

Higher-ordered aggregates of molecules of inactivated LysPhi80α are likely to be

342

formed by non-covalent interactions; it is evident from the absence of bands with high

343

molecular weights on SDS-PAGE electrophoregram (Fig. 10 A, lanes 3-6). For inactivated

344

LysPhi80α the presence of dimers with a molecular weight of about 110 kDa was revealed

345

(Fig 10 A, lanes 4 and 6), that are likely to be formed through –S-S- bonds, and active

346

enzyme doesn’t demonstrate such dimers (Fig. 10 A, lane 2).

cr

ip t

341

Based on these data, we propose a scheme of LysPhi80α inactivation, shown in Fig.

348

10 B. The irreversible denaturation, i.e. the transition N → D (storage conditions), described

349

by the first order kinetic equation (Figs. 5 and 9), is the limiting stage of inactivation of lysins

350

of phages phi11 and phi80α. Molecules of LysPhi11 and LysPhi80α in the unfolded state are

351

aggregated due to the appearance of additional areas of (hydrophobic) interaction between

352

them (Fig. S5).

M

an

us

347

4. Discussion

354

The study of the activity and stability of LysPhi11 at 37°C clearly shows that changing

355

the salt composition of the solution can be a way to control the properties of the biocatalyst.

356

We found that the activity of LysPhi11 decreases with increasing of NaCl concentration.

357

Previously the cases of endolysins losing their activity with increasing of salt concentration

358

were identified. The deletion construct expressing the staphylolytic LysK CHAP catalytic

359

domain loses 30% of activity at 150 mM NaCl and is almost completely inactivated at 500

360

mM electrolyte [16]. The activity of peptidoglycan hydrolase HydH5 is reduced twice at 200-

361

500 mM NaCl [17]. Endolysin PlyGRCS exhibits maximum activity at 125-500 mM NaCl

362

[18]. Endolysin LysK exhibits maximum activity at 200-400 mM NaCl [19].

Ac ce p

te

d

353

363

Using the analysis of the enzyme CD spectra under various conditions it was

364

established that high levels of random coils (more than 40%) at 37°C is the result of thermal

365

denaturation of the enzyme molecules. The unfolding results in additional areas of interaction

Page 16 of 36

between LysPhi11 molecules, which causes aggregation. Presence of NaCl enhances

367

intermolecular hydrophobic interactions, thereby intensifying the process of aggregation and

368

enlarging of particles. Aggregation is the reason for the decrease of the enzyme activity with

369

increasing of salt concentration, since there are steric constraints to access the substrate.

ip t

366

Modifying the enzyme activity as a function of NaCl concentration can be associated

371

with the state of the cell wall of S. aureus. Inside Staphylococcus aureus bacterial cells the

372

salt concentration is very high and can reach 6% w/v [21]. The salt content, equal to 6% w/v,

373

corresponds to ~1000 mM NaCl, i.e. under the measurement conditions of the enzyme

374

activity the salt concentration in the external medium is lower than inside the bacterial cells.

375

This results in a salt concentration gradient and cells swelling. The degree of swelling of the

376

cells, which increases with decreasing the salt concentration in the external medium, affects

377

the degree of stretching of the cell walls, which determines the peptidoglycan conformation

378

and accessibility of chemical bonds for the hydrolysis and hence the activity of the enzyme.

d

M

an

us

cr

370

Inactivation of LysPhi11 (37°C) according to first order mechanism occurs quickly

380

enough: the enzyme loses half of its activity after 10 minutes (6x102 s) due to denaturation.

381

Inactivation by aggregation of the enzyme in PPB-NaCl is slower than in PPB and can be

382

suppressed by diluting: LysPhi11 at a concentration of 0.4 mg/mL loses half of its activity in

383

10 minutes (6x102 s) and at a concentration of 0.1 mg/mL loses half of its activity in 150

384

minutes (9x103 s). It is worth noting the importance of having a high-ordered secondary

385

structure of lytic enzymes. Lytic enzyme LysK is inactivated at 37°C and pH 7.5 by the

386

aggregation mechanism and loses half of its activity after 360-900 minutes (2x104-5x104 s)

387

[12]. The secondary structure of LysK contains 24% of random coils, which is twofold less

388

than the more rapidly inactivating LysPhi11.

Ac ce p

te

379

389

Under storage conditions (4°C and 22°C, 10-500 mM NaCl, enzyme concentration

390

0.2-0.8 mg/mL, pH 6.0-9.0) dramatic increase of the constants of first order inactivation of

Page 17 of 36

LysPhi11 was observed between pH 8.0-9.0. This could mean that increasing pH promotes

392

the transition of the enzyme from one form to another. This transition may be caused by

393

deprotonation of any group of the enzyme, resulting in a drop of lysin stability. The following

394

groups have pK values above 8.0 in the protein molecule: the thiol group of a cysteine residue

395

(8.5-8.8), the amino group of a lysine residue (10.0-10.2), the guanidinium group of arginine

396

residue (>12.0), and the hydroxyl group of tyrosine residue (9.6-10.0) [21]. The thiol group of

397

a cysteine residue has the closest pK value to the observed interval of pH where the decrease

398

of the first order inactivation constants was observed. It was established previously that this

399

residue is essential for catalytic activity of lytic enzymes. Amino acid substitutions of cysteine

400

residues cause loss of activity of lysin PlyC (Cys333) [22], the catalytic CHAP domain of

401

staphylolytic enzymes LysK (Cys54) and LysGH15 (Cys54) [23, 24]. In this case, the

402

importance of a cysteine residue for the functioning of phage phi11 lysin was indirectly

403

confirmed by kinetic studies. Catalytic CHAP domain of LysPhi11 molecule contains only

404

one cysteine residue (Cys32), which is likely to be important for activity.

te

d

M

an

us

cr

ip t

391

X-ray analysis revealed that CHAP domains of staphylolytic enzymes LysK and

406

LysGH15 contain Ca-binding sites, in which Ca++ ions are coordinated by residues Asp (45,

407

47, 56), Tyr (49), and His (51) [23, 24]. Amino acid substitutions of Asp residues lead to

408

complete loss of CHAP domains activity, which helps confirm the importance of Ca++ ions

409

for catalysis. In the process of LysPhi11 inactivation a violation of the spatial configuration of

410

Me++ binding site occurs as a result of unfolding of the protein. The loss of activity of the

411

enzyme is likely to be caused by either dissociation of Me++ or limiting the possibility of

412

binding to the substrate. Because of the proximity of the Ca++ and Mg++ ionic radii the

413

recovery process of the enzyme activity is probably nonselective.

Ac ce p

405

Page 18 of 36

414

The maximum time of LysPhi11 half-inactivation (120 – 160 days, 1.0x107-1.4x107 s)

415

was observed at 4°C, pH 6.0-7.5 and 10 mM NaCl. It can be concluded that LysPhi11 enzyme

416

is very convenient for the long-term storage. Lysins of phages phi11 and phi80α are only homologous in the first 181 amino acids.

418

Molecules of LyPhi11 and LysPhi80α contain the same number of charged amino acid

419

residues, 97 and 95, respectively. The content of hydrophobic amino acid residues in the

420

molecules of the enzymes is different: there are 191 in LysPhi11, and 176 in LysPhi80α. The

421

higher content of hydrophobic amino acids provides a higher degree of compaction of the

422

protein globule in an aqueous medium and imparts resistance to unfolding. The main reason

423

for the difference in the kinetic behavior of the enzymes is the difference in their amino acid

424

composition. The similarity in the manifestation of staphylolytic activity may be due to the

425

presence of areas with a high degree of homology in the molecules of the enzymes (from the

426

1st to the 181st amino acid). It is in this area of the amino acid sequences of both lysins where

427

the CHAP domain is located containing groups responsible for catalysis [8]. We can conclude

428

that non-homologous domains such as Amidase and SH3_b play the key role in maintaining

429

the structure of LysPhi11 and LysPhi80α molecules which is the reason for differences in the

430

parameters of enzymes inactivation.

432

cr

us

an

M

d

te

Ac ce p

431

ip t

417

5. Conclusion

At 37°C lysins of phages phi11 and phi80α are inactivated following various

433

mechanisms. Under storage conditions LysPhi11 and LysPhi80α are inactivated by a

434

monomolecular mechanism, but the constants of inactivation of the enzymes differ

435

considerably. For phage phi11 lysin there are optimum storage conditions under which the

436

time of half-inactivation of LysPhi11 is 120-160 days. To ensure a long-term storage of

437

LysPhi80α it needs to be stabilized (e.g., by forming enzyme-polymer complexes and by

438

adding a low molecular weight additives, etc.). Differences in the parameters of inactivation

Page 19 of 36

of the enzymes are the result of the difference in their amino acid composition. The common

440

properties in the antimicrobial activity are due to the fact that in both LysPhi11 and

441

LysPhi80α molecules there is an area with a high degree of homology (from the 1st to 181st

442

amino acid), where the catalytic CHAP domain is located.

ip t

439

6. Acknowledgements

444

The study was supported by Ministry of Science and Education of Russian Federation

445

(Project 11.G34.31.0004) and Russian Foundation for Basic Research (grant number 15-04-

446

07995).

us

cr

443

7. References

448

[1]. Lowy FD. Staphylococcus aureus infections. The New England Journal of

452 453 454 455 456

M

aureus infection. Clinical Infectious Diseases 2008;46:350 - 59.

d

451

[2]. Gordon RJ, Lowy FD. Pathogenesis of methicillin-resistant Staphylococcus

[3]. Schmelcher M, Donovan DM, Loessner MJ. Bacteriophage endolysins as novel

te

450

Medicine 1998;339:520 - 32.

antimicrobials. Future Microbiol 2012;10:1147 - 71. [4]. Nelson DC, Schmelcher M, Rodriguez-Rubio L, Klumpp J, Pritchard DG, Dong S

Ac ce p

449

an

447

et al. Endolysins as antimicrobials. Advances in Virus Research 2012;83:299 - 365. [5]. Szweda P, Schielmann M, Kotlowski R, Gorczyca G, Zalewska M, Milewski S.

457

Peptidoglycan hydrolases-potential weapons against Staphylococcus aureus. Applied

458

Microbiology and Biotechnology 2012;96:1157 – 74.

459

[6]. Donovan DM, Lardeo M, Foster-Frey J. Lysis of staphylococcal mastitis

460

pathogens by bacteriophage phi11 endolysin. FEMS Microbiology Letters 2006;265:133 – 39.

461

[7]. Navarre WW, Ton-That H, Faull KF, Schneewind O. Multiple enzymatic

462

activities of the murein hydrolase from staphylococcal phage phi11. Identification of a D-

463

alanyl-glycine endopeptidase activity. Journal of Biological Chemistry 1999;274:15847 – 56.

Page 20 of 36

[8]. Schmelcher M, Shen Y, Nelson DC, Eugster MR, Eichenseher F, Hanke DC et al.

465

Evolutionarily distinct bacteriophage endolysins featuring conserved peptidoglycan cleavage

466

sites protect mice from MRSA infection. Journal of Antimicrobial Chemotherapy 2015;(in

467

press).

471 472 473

cr

470

bacteriophage T4. Nature 1970;227:680 - 85.

[10]. Bohm G, Muhr R, Jaenicke R. Quantitative analysis of protein far UV circular dichroism spectra by neural networks. Protein engineering 1992;5:191 - 95.

us

469

[9]. Laemmli UK. Cleavage of structural proteins during the assembly of the head of

[11]. Greenfield NG. Using circular dichroism spectra to estimate protein secondary structure. Nat Protoc 2006;1:2876 – 90.

an

468

ip t

464

[12]. Filatova LY, Becker SC, Donovan DM, Gladilin AK, Klyachko NL. LysK, the

475

enzyme lysing Staphylococcus aureus cells: specific kinetic features and approaches towards

476

stabilization. Biochimie 2010;92:507 - 13.

479

d

te

478

[13]. Kurganov BI. Kinetic of thermal aggregation of enzymes. Biochemistry (Moscow) 1998;63:430 - 32.

[14]. Filatova LY, Donovan DM, Becker SC, Priyma AD, Kabanov AV, Klyachko

Ac ce p

477

M

474

480

NL. An investigation of the structure and function of antistaphylococcal endolysins using

481

kinetic methods. Moscow University Chemistry Bulletin 2014;69:107 – 111.

482

[15]. Zhang J. Protein-Protein Interactions in Salt Solutions. In Cai W, Hong H,

483

editors. Protein-Protein Interactions - Computational and Experimental Tools, Croatia:

484

InTech; 2012, p. 359 - 376.

485

[16]. Fenton M, Ross RP, McAuliffe O, O’Mahony J, Coffey A. Characterization of

486

the staphylococcal bacteriophage lysin CHAPK. Journal of Applied Microbiology

487

2011;111:1025 – 35.

Page 21 of 36

488

[17]. Rodríguez L, Martínez B, Zhou Y, Rodríguez A, Donovan D, García P. Lytic

489

activity of the virion-associated peptidoglycan hydrolase HydH5 of Staphylococcus aureus

490

bacteriophage vB_SauS-phiIPLA88. BMC Microbiology 2011;11:138. [18]. Linden SB, Zhang H, Heselpoth RD, Shen Y, Schmelcher M, Eichenseher F et

492

al. Biochemical and biophysical characterization of PlyGRCS, a bacteriophage endolysin

493

active against methicillin-resistant Staphylococcus aureus. Appl Microbiol Biotechnol

494

2015;99:741-52.

496

cr

[19]. Becker SC, Foster-Frey J, Donovan DM. The phage K lytic enzyme LysK and

us

495

ip t

491

lysostaphin act synergistically to kill MRSA. FEMS Microbiol Lett 2008;287:185 – 91. [20]. Shlegel G. Microbiology. 6th ed. Moscow: Mir; 1987.

498

[21]. Kantor Ch, Shimmel P. Biophysical chemistry. Moscow: Mir; 1984.

499

[22]. Nelson D, Schuch R, Chahales P, Zhu S, Fischetti VA. PlyC: a multimeric

M

bacteriophage lysin. PNAS 2006;103:10765 – 10770.

d

500

an

497

[23]. Sanz-Gaitero M, Keary R, Garcia-Doval C, Coffey A, Raaij MG. Crystal

502

structure of the lytic CHAPK domain of the endolysinLysK from Staphylococcus aureus

503

bacteriophage K. Virology Journal 2014;11:133.

Ac ce p

504

te

501

[24]. Gu J, Feng Y, Feng X, Sun C, Lei L, Ding W et al. Structural and Biochemical

505

Characterization Reveals LysGH15 as an Unprecedented ‘‘EF-Hand-Like’’ Calcium-Binding

506

Phage Lysin. PLOS Pathogens 2014;10:e100410.

507 508 509 510 511 512

Legends to figures

Fig. 1. Influence of NaCl on LysPhi11 activity. Experimental conditions: 37°C, pH 7.5, bacteria OD600nm = 0.6 absorbance units, LysPhi11 8 μg/mL. Fig. 2. SDS-PAGE analysis of LysPhi11. Lane M, protein molecular weight markers; lane 2, LysPhi11 before inactivation; lane 3, LysPhi11 after inactivation at 37°C.

Page 22 of 36

514 515 516

Fig. 3. Aggregation of LysPhi11 in different media. Experimental conditions: 37°C, LysPhi11 0.4 mg/mL, PPB (white squares), PPB-NaCl (black squares). Fig. 4. Influence of incubation time on the particle size of LysPhi11. The enzyme was incubated at 37°C for 60 minutes; PPB-NaCl (grey columns); PPB (white columns).

ip t

513

Fig. 5. Inactivation of LysPhi11 under storage conditions (kin values).

518

Experimental conditions: LysPhi11 0.2-0.8 mg/mL, universal buffer (Tris-phosphate-citrate),

519

22°C (A), 4°C (B).

cr

517

Fig. 6. Inactivation of LysPhi11 under storage conditions (τ1/2 values).

521

Experimental conditions: LysPhi11 0.2-0.8 mg/mL, universal buffer (Tris-phosphate-citrate),

522

22°C (A), 4°C (B).

an

us

520

Fig. 7. Reactivation of LysPhi11 by Ca++ and Mg++. Experimental conditions: 37°C,

524

deionized water, bacteria OD600nm = 0.6 absorbance units, LysPhi11 8 μg/mL. Ca++ (white

525

columns), Mg++ (grey columns).

528

d

te

527

Fig. 8. Dependence of LysPhi80α residual activity on time. Experimental conditions: 37°C, PPB-NaCl, LysPhi80α 0.4 mg/mL. Fig. 9. Inactivation of LysPhi80α under storage conditions. Experimental

Ac ce p

526

M

523

529

conditions: LysPhi80α 0.2-0.8 mg/mL, universal buffer (Tris-phosphate-citrate), pH 7.5. 22°C

530

(black circles), 4°C (white circles).

531

Fig. 10. Analysis of LysPhi80α inactivation by SDS-PAGE. A. SDS-PAGE

532

electrophoregram of LysPhi80α before and after inactivation. Lanes 1, 2, LysPhi80α before

533

inactivation; lanes 3, 4, LysPhi80α after inactivation at 22°C; lanes 5, 6 LysPhi80α after

534

inactivation at 4°C; lanes M, molecular weight protein markers. Lanes 1, 3, 5 correspond to

535

the enzyme samples mixed with denaturing sample buffer without 2-mercaptoethanol, lanes 2,

536

4, 6 correspond to the enzyme samples mixed with denaturing sample buffer with 2-

Page 23 of 36

537

mercaptoethanol. B. Schematic representation of LysPhi80α inactivation process. N – native

538

enzyme, D – denatured enzyme, A – aggregated enzyme, D-S-S-D – dimer.

Ac ce p

te

d

M

an

us

cr

ip t

539

Page 24 of 36

539

Table 1.

540

Content of secondary structure elements (%) in LysPhi11 molecule as deconvoluted from CD spectra. 22°C

Secondary

37°С

No NaCl

150 mM NaCl

No NaCl

150 mM NaCl

Helix

14.4±1.3

13.4±1.0

12.1±0.5

12.3±1.1

Antiparallel

19.9±4.9

18.8±1.0

13.7±1.8

13.0±4.6

Parallel

17.3±1.7

16.9±1.5

12.9±1.0

β-turns

21.5±0.6

23.5±1.6

18.3±0.5

18.6±0.5

Random coils

26.8±4.2

27.4±1.2

43.0±1.3

43.1±4.0

cr

us

13.0±1.6

Ac ce p

te

d

M

542

an

541

ip t

structure element

Page 25 of 36

542

Table 2.

543

Kinetic parameters of LysPhi11 inactivation at 37°C.

Buffer

τ

n

Inactivation constant

PPB-NaCl

2

ksec-in = (6.2±0.5)x10-3 s-1M

10-150

PPB

1

kin = (7.8±0.2)x10-4 s-1

10

cr

min

us

544

1/2,

ip t

Parameters of inactivation

Ac ce p

te

d

M

an

545

Page 26 of 36

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 27 of 36

Ac ce p

te

d

M

an

us

cr

ip t

Figure(s)

Page 28 of 36

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 29 of 36

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 30 of 36

Ac ce p

te

d

M

an

us

cr

ip t

Figure(s)

Page 31 of 36

Ac ce p

te

d

M

an

us

cr

ip t

Figure(s)

Page 32 of 36

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 33 of 36

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 34 of 36

Ac ce p

te

d

M

an

us

cr

ip t

Figure(s)

Page 35 of 36

Ac

ce

pt

ed

M

an

us

cr

i

Figure(s)

Page 36 of 36