Production, purification, and characterization of soluble NADH-flavin Oxidoreductase (StyB) from Pseudomonas putida SN1

ORIGINAL ARTICLE

Production, purification and characterization of serraticin A,a novel cold-active antimicrobial produced by Serratiaproteamaculans 136L.A. Sanchez1, M. Hedstrom2, M.A. Delgado3 and O.D. Delgado1

1 PROIMI – CONICET, Chacabuco, Tucuman, Argentina

2 Department of Biotechnology, Center for Chemistry and Chemical Engineering, Lund University, Lund, Sweden

3 INSIBIO – CONICET, Chacabuco, Tucuman, Argentina

Introduction

The need to decrease the use of chemical additives in

food has triggered the search for natural antimicrobial

substances produced by micro-organisms from different

sources. Application of antimicrobial compounds has

been well documented as a biotechnological advantage

over different industrial sectors like food-processing,

pharmaceutical chemistry and cosmetic industries. Conse-

quently, screening and selection programs for antimicro-

bial producing micro-organisms and bioproducts have

increased in the last years, particularly focusing into

improving the process control as well as the quality and

safety of final products by inhibition of both, undesirable

pathogenic and spoilage bacteria (Zahner and Fielder

1995; Davies and Webb 1998).

Since the introduction of penicillin in the 1940s, antibi-

otics have a history of success in controlling morbidity

and mortality caused by infectious diseases. However, as a

consequence of frequent use, bacterial resistance to

known classes of antibiotics has become a severe global

problem in recent years and presents a continuous chal-

lenge (Pohlmann et al. 2005). Nowadays, it cannot be

ignored that the antibiotic age is under adverse circum-

stances. The high levels of antibiotic resistance among

important pathogens along with an irregular supply of

Keywords

antimicrobials, bacteriocins, psychrophiles,

psychrotolerant, Serratia proteamaculans.

Correspondence

Osvaldo Delgado. PROIMI, Av. Belgrano y Pje.

Caseros (4000) Tucuman. Argentine.

E-mail: [email protected]

2009 ⁄ 2139: received 13 December 2009,

revised 10 March 2010 and accepted 10

March 2010.

doi:10.1111/j.1365-2672.2010.04720.x

Abstract

Aim: This study focuses on the production, purification and characterization of

serraticin A, a novel cold-active antimicrobial produced by Serratia proteamac-

ulans 136.

Methods and Results: A Ser. proteamaculans strain producing a novel cold-

active antimicrobial was isolated from Isla de los Estados, Argentina. Antimi-

crobial production was optimized in a BIOFLO 101 bioreactor under batch

culture mode, with temperature, pH and dissolved oxygen controlled condi-

tions. A purification protocol was developed including activated charcoal

adsorption, solid-phase C18 extraction (SPE) and semi-preparative HPLC. The

molecular weight was determined by LC ⁄ QTOF ⁄ MS ⁄ MS mass analysis.

Conclusions: Serratia proteamaculans 136 produces a cold-active low molecular

bacteriocin-like compound named serraticin A. In this work, it has been labo-

ratory-scale produced, purified and partially characterized. Cross-immunity test

revealed that serraticin A is very different from other well-known microcins

assayed, with a wide inhibitory spectrum, showing an interesting biotechnology

potential to be applied as a control agent against pathogenic bacteria.

Significance and Impact of the Study: The present study is the first report of a

cold-active compound with antimicrobial activity from Ser. proteamaculans.

The work also highlights that cold environments could be a suitable source of

micro-organisms with ability to produce cold-active biomolecules of biotechno-

logical interest.

Journal of Applied Microbiology ISSN 1364-5072

ª 2010 The Authors

Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 1

novel classes of antibiotics, and the reduced number of

pharmaceutical companies involved in the discovery and

development of such compounds reveals an emergency

state into this scope (Wenzel 2004). The rate for drug dis-

covery ⁄ resistance has decreased in the latest 20 years with

evidence for novel resistance levels among pathogens such

as vancomycin-resistant enterococci, methicillin-resistant

and vancomycin-resistant Staphylococcus aureus and

multidrug-resistant Pseudomonas aeruginosa (Song 2008).

This current increase in drug resistance among several

common pathogenic bacteria highlights the necessity to

discover novel active compounds (Zahner and Fielder

1995; Davies and Webb 1998). Most new classes of anti-

biotics have arisen by testing natural sources (Gootz

1990), classically including several microbiological tools

like the isolation of novel micro-organisms; or otherwise

by means of the modification of well-known antibiotic-

producing micro-organisms as well as by metabolic path-

way diversion by engineering fermentation process (Souza

et al. 1982). The possibility to achieve more precise adap-

tations for a given industrial process, like low-temperature

antibiotic activity may contribute a significant biotechno-

logical improvement; with this aim, extremophilic micro-

organisms have been relatively unexplored group with

respect to mesophilic ones. Evidences from their study are

a thermophilic bacteriocin producing Bacillus (Kabuki

et al. 2007); a heat-labile b-lactamase purified from the

psychrophile Psychrobacter immobilis (Feller et al. 1997),

several archaea able to produce archaeocins which inhibit

closely related species (Aravalli et al. 1998; Prangishvili

et al. 2000) and recently described psychrophiles with

ability to produce antimicrobials at low temperatures

(O’Brien et al. 2004; Sanchez et al. 2009). Hence, the

extremophiles seem to exhibit a potential whose study is

growing day by day. This article introduces to the produc-

tion parameters, purification, biochemical characterization

and some important properties of serraticin A.

Materials and methods

Bacterial strains

Bacterial strains used in this work were obtained from

clinical isolations and collections (Table 1); Ser. protea-

maculans 136, natural antimicrobial producer, was

isolated in our laboratory (Sanchez et al. 2009).

Cross-streaking test

Cross-immunity or cross-inhibition against known micro-

cin producers was determined (Table 2) involving cross-

testing by the deferred (delayed) antagonism method

(Tagg et al. 1976).

Effect of Temperature, pH and media composition on

antimicrobial production

To determine the optimal medium for antimicrobial

production, different media compositions were evaluated:

LB, M9, M63 (Miller 1972) and R modified medium

(Reasoner and Geldreich 1985). Solid and soft media

contained 1Æ5 and 0Æ8% agar respectively and, when appro-

priate, l-amino acids (25 lg ml)1) were added.

Cell-free supernatants were evaluated for antimicrobial

production after 72 h incubation at 8�C. Optimal temper-

ature and pH values for antimicrobial production were

determined by incubating liquid cultures at both, differ-

ent temperatures and pH values.

Antimicrobial activity was determined by the critical

dilution method (Mayr-Harting et al. 1972), using Salmo-

nella enterica ser. Newport as indicator strain. Activity is

expressed either as antibiotic arbitrary units per millilitre

Table 1 Bacterial strains used in this study. Known microcin (Mcc)

producers are also listed

Strains Reference ⁄ Source

Shigella flexneri ATCC 29508

Staphylococcus aureus ATCC 29213;

ATCC 25923

Micrococcus luteus ATCC 10240

Bacillus subtilis 168 BGSC 1A1

Staphylococcus epidermidis ATCC 10390

Salmonella enterica

serovar Newport

Delgado et al. (2006)

Salmonella enterica

serovar typhimurium

Delgado et al. (2006)

Listeria monocytogenes ATCC 13932

Shigella sonnei Delgado et al. (2006)

Escherichia coli AB1133 CGSC

E. coli ATCC 35218;

ATCC 25922

E. coli O157:H7 (stx1,stx2);

O26:H11; O15:NM

Sanchez et al. (2009)

Citrobacter freundii ATCC 14135

Enterobacter aerogenes ATCC 10699

Enterobacter cloacae Sanchez et al. (2009)

Pseudomonas aeruginosa ATCC 27853

Salmonella enteritidis Delgado et al. (2006)

Proteus mirabilis Sanchez et al. (2009)

Proteus vulgaricus Sanchez et al. (2009)

Klebsiella pneumoniae

(Mcc E492)

De Lorenzo and

Pugsley (1985)

E. coli (Mcc J25) Salomon and Farias (1999)

E. coli (Mcc H47) Lavina et al. (1990)

E. coli (Mcc C7 ⁄ C51) Garcıa-Bustos et al. (1984);

Kurepina et al. (1993)

E. coli (Mcc V) Waters and Crosa (1991)

E. coli (Mcc B17) Baquero et al. (1978)

Serratia proteamaculans

136 (serraticin A)

This work

Serraticin A, a novel cold-active antimicrobial L.A. Sanchez et al.

2 Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology

ª 2010 The Authors

(AU ml)1) or as micrograms per millilitre (lg ml)1)

(Sanchez et al. 2009).

Shake-flasks and batch cultivations

Shake-flask cultivations were conducted in 1-L Erlen-

meyer flasks, containing 300 ml of LB broth and inocu-

lated with 10% v ⁄ v of an overnight culture of

Ser. proteamaculans 136. Flasks were incubated at 8�C in

an orbital shaker at 200 rev min)1. A 7Æ5-l stirred tank

bioreactor (BioFlo 110 Modular Benchtop Fermen-

tor ⁄ Bioreactor; New Brunswick Scientific, Edison, NJ,

USA), with a working volume of 5 l (LB broth) and

10% inoculum was used. Agitation consisted of two

6-bladed Rushton-type impellers (52 mm), operating at

200 rev min)1. The culture pH was controlled at 7Æ0(initial pH was 6Æ8), while temperature was maintained

at 8�C. Air was injected at 0Æ5 v.v.m. flow rate. Growth

kinetic of both, shake-flasks and bioreactor was fol-

lowed by OD600nm measurements and antimicrobial

production was assayed along cultivation against

Salm. enterica ser. Newport as indicator strain. The cells

were removed by centrifugation at 8000 g at 4�C. Cell-

free supernatants were maintained at 4�C for further

antimicrobial purification.

Antimicrobial purification

Cell-free supernatant was mixed with granular activated

charcoal (6 g l)1), stirred and filtered through chromato-

graphy paper. Charcoal-bonded pigments from culture

medium were removed by resuspending in 300 ml of

30% (v ⁄ v) acetone and filtered as mentioned earlier. The

target material was suspended in 100 ml of 80% (v ⁄ v)

acetone and filtered. Acetone was evaporated from eluate

and original volume was corrected by water addition. The

antimicrobial was further adsorbed on a C18 Sep-Pak

SPE cartridge (Varian, Inc., Palo Alto, CA, USA) and

eluted by using 50% (v ⁄ v) acetonitrile. Further purifica-

tion steps were carried out by RP-HPLC by using a semi-

preparative C18 l Bondapak HPLC column (125A pore

size, 10l granulometry). A 0–100% water-acetonitrile

gradient (containing 0Æ1% TFA) at a flow rate of 6 ml

min)1 was established during 50 min, and optical density

was monitored by a 2998 Waters PDA detector (Waters

Co., Milford, MA, USA). Active fractions were pooled

and concentrated. Purity of the compound was re-evalu-

ated by analytical HPLC on the C18 column using similar

conditions. Protein concentration was assayed after each

purification step by using a Bicinchoninic acid protein kit

(Sigma Co., St Louis, MO, USA).

Minimal inhibitory concentration

Minimal inhibitory concentration (MIC) of the antimi-

crobial against different micro-organisms was determined

by the two-fold serial dilution assay (Yaron et al. 2003).

MIC was considered as the lowest concentration that

showed no turbidity increments (OD600nm) after 48 h of

incubation.

Antimicrobial mode of action on sensitive cell viability

Escherichia coli AB1133 and Salm. enterica ser. Newport

were exponentially grown in LB broth and to both

cultures, antimicrobial was added at MIC. The inhibition

of growth because of antimicrobial addition was followed

by OD600nm measurements at appropriate intervals of time.

To evaluate indicator strains viability, cells were counted

by the plate count method on LB and LB + antimicrobial

agar plates.

Antimicrobial effects on sensitive cell morphology

To establish an antimicrobial effect on sensitive cells

morphology, flask containing 10 ml of LB medium was

Table 2 Cross-streaking test

Microcin

Microcin producer sensitivity ⁄ immunity

Klebsiella

pneumoniae

(Mcc E492)

Escherichia

coli (Mcc C7)

E. coli

(Mcc H47)

E. coli

(Mcc V)

E. coli

(Mcc J25)

E. coli

(Mcc B17)

Serratia protemaculans

(serraticin A)

Salmonella

enterica ser.

Newport

MccE492 R S S S R S R R

MccC7 ⁄ C51 S R S S S S S S

MccH47 S S R S S S S R

MccV S S S R S S S S

MccJ25 R S S S R S S S

MccB17 S S S S S R S S

serraticin A R S S S S S R S

Mcc, Microcin; R, Resistant; S, Sensitive.

L.A. Sanchez et al. Serraticin A, a novel cold-active antimicrobial

ª 2010 The Authors

Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 3

inoculated with approximately 108 cells of either E. coli

AB1133, ATCC 35218, DH5a or Salm. enterica ser.

Newport sensitive strains and incubated 2 h at 37�C, after

that, purified antimicrobial was added at MIC and

incubation was continued. After 12 h of culturing, cell

morphology was evaluated regarding to the control with

no antimicrobial addition. Sensitive strain cell morphology

was examined on a Nikon Eclipse 80i (Nikon GmbH,

Dusseldorf, Germany) microscope at · 1000 magnification.

Haemolytic, bio-emulsifier and bio-surfactant activities

Antimicrobial haemolytic activity was evaluated on blood

agar-plates (Banat 1993); LB soft-agar medium was sup-

plemented with 2% (v ⁄ v) of a fresh erythrocyte suspen-

sion (Moran et al. 2002). Ten microlitres of two-fold

serial dilutions of the antimicrobial stock solution

(500 lg ml)1) was prepared in physiological solution (pH

7Æ2) and deposited over LB-blood agar plates and incu-

bated at 37�C during 3 days. Staph. aureus ATCC 29213

was used as positive haemolytic control (Cooper et al.

1964).

Antimicrobial emulsifier property was also evaluated

according to Moran et al. (2000): 2 ml of a two-fold serial

dilution was prepared as described earlier, mixed with

3 ml of kerosene in a test tube and vortexed for 2 min.

After 24 h, the emulsified kerosene proportion was evalu-

ated. In the same way, the evaluation for bio-surfactant

activity was carried out by the drop-collapse test (Jain

et al. 1991) by using mineral, olive and corn oil.

Fluorescence spectrometry

To gain insight into the antimicrobial nature, fluorescence

of antimicrobial was measured. A purified solution

(5 lg ml)1) was prepared in sterile water, and fluores-

cence was measured in a PerkinElmer LS-55 fluorescence

spectrometer (PerkinElmer, Waltham, MA, USA), using

emission ⁄ excitation wavelengths associated with aromatic

aminoacids: tryptophan (kem = 348 nm; kex = 280 nm);

phenylalanine (kem = 282 nm; kex = 257 nm); and tyro-

sine (kem = 303 nm; kex = 274 nm).

Mass determination

Molecular mass of the compound was determined on a

Hybrid Quadrupole-TOF LC ⁄ MS ⁄ MS Mass Spectrometer

using a QSTAR� hybrid pulsar-i instrument (Applied

Biosystems, Foster City, CA, USA). Prior to, and in

sequence with MS, the sample was separated on RP

(C18)-HPLC Kromasil column (Eka, Sweden). Twenty

microlitres of sample (approximately 0Æ05 mg ml)1) was

injected in a linear gradient of water ⁄ acetonitrile at a

starting point of 80% water. The gradient was allowed to

proceed for 30 min to a final amount of 10% water. The

electrospray ionization (ESI) TurboIonSpray� source was

set to positive ion mode with a source voltage of +5500 V.

The quadrupole system was adjusted to scan between m ⁄ z50–3000 in TOF-MS mode, whereas for product ion mode

(i.e. MS ⁄ MS) a range of m ⁄ z 50–2000 was chosen. The

m ⁄ z value of individual precursor ions was automatically

selected in the information-dependent acquisition software

feature for fragmentation and was collided under argon

pressure using rolling collision energies ranging from 12

to 60 eV (i.e. collision induced dissociation).

Results

Cross-streaking test

Serratia proteamaculans 136 isolation was previously

described (Sanchez et al. 2009), and broad inhibition

spectra against both, Gram (+) and Gram ()) were

observed, with E. coli O26:H11 and Enterobacter aerogenes

the only exceptions, in which growth was not inhibited.

Cross-streaking test showed that the antimicrobial isolated

inhibited the growth of several up to date microcin-

producers (Table 2) with the only exception of MccE492

producer, Klebsiella pneumoniae.

Effect of media composition, pH and temperature on

serraticin A production

A significant dependence on media composition for anti-

microbial production was observed when Ser. proteamacu-

lans 136 was grown in several media with a maximum of

800 AU ml)1 in LB broth, although approximately

400 AU ml)1 was detected in other tested media.

Production of the antimicrobial compound was

observed only at 4–20�C range, while nonproduction was

observed at 30–37�C in spite of growth. Culture pH did

not affect significantly the antimicrobial production; only

at pH 5Æ0 or lower, a significant decrease in antimicrobial

production was observed.

Culture condition effects on Serratia proteamaculans

growth and antimicrobial production

Growth kinetics and antimicrobial production were dif-

ferent in shake-flasks and bioreactor cultivations (Fig. 1).

In shake-flask scale, culture growth was slower (lmax =

0Æ11), and the activity was detected in supernatants after

84 h of cultivation with a maximum of 600 AU ml)1 after

130 h of cultivation; on the other hand, a lmax = 0Æ18

and 800 AU ml)1 were reached at bioreactor scale after

45 h of cultivation; whereas the antimicrobial was

Serraticin A, a novel cold-active antimicrobial L.A. Sanchez et al.

4 Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology

ª 2010 The Authors

detected at early stationary phase of growth (30 h of

culturing), no detectable antimicrobial activity was

observed at exponential growth phase, even in 50-fold-

concentrated samples. Final pH for both cultivation scales

was around 9Æ8 when it was left un-controlled.

Antimicrobial production and purification

Once the parameters for antimicrobial production were

optimized, 5-l batch culivations were carried out as

described earlier. After 50 h of culturing both, air and

stirrer were increased to allow foam formation. Two

litres of foam was collected, with two-fold antimicrobial

activity concentration (1600 AU ml)1), and centrifuged

to separate cells. A purification scheme was developed

based on the observation that the antimicrobial com-

pound could be efficiently adsorbed on activated

charcoal particles and from there, eluted with 80% ace-

tone without activity loss. After acetone wash, and aceto-

nitrile elution, SPE (C18) achieved a significant specific

activity increment (Table 3) reaching the most important

separation from contaminants. A final purification step

was carried out by using preparative RP-HPLC detecting

two main peaks in chromatograms with similar retention

times. Peaks were separately collected and antimicrobial

activity was only detected in the one with major reten-

tion-time (31Æ5 min). Purity of the antimicrobial fraction

was verified using analytical RP-HPLC monitored by a

photo-diode array detector and also by mean of

LC ⁄ MS ⁄ MS showing the homogeneity of the compound

with a retention time of 16Æ2 min and a kmax at

298Æ9 nm.

Minimal inhibitory concentration

Sensitivity of several bacterial genera against antimicrobial

was assayed and MIC for each one was determined

(Table 4). Serraticin A was active against E. coli, Salmo-

nella, Shigella, Listeria, Staphylococcus and Bacillus subtilis

168 strains displaying a broad inhibition spectrum; how-

ever, Gram ()) bacteria appeared to be inhibited at lower

MIC than Gram (+) assayed bacteria.

While medium composition did not modify antimicro-

bial sensitivity on Salmonella strains, sensitivity of

Staph. aureus and E. coli AB1133 strains was most

significantly affected, reaching an eight-fold increase.

However, in Listeria monocytogenes, it led to a moderate

enhancement (5Æ4-fold) when LB medium was used. On

the other hand, Ser. proteamaculans 136 producer strain

could grow in the presence of 10 lg ml)1 of antibac-

terial, indicating its immunity to serraticin A exogenous

addition.

Effect of serraticin A over viability and morphology of

sensitive cells

Serraticin A effect on cell viability was evaluated on E. coli

AB1133 and Salm. enterica ser. Newport sensitive strains.

After 2 h of antimicrobial addition, biomass (OD600nm)

continued to increase at a slower rate when compared to

the control without antibiotic addition. During the treat-

ment, cells could not develop and optical density kept

constant regarding to controls (Fig. 2). At the final time,

the control had 1Æ4 · 1011 viable cells, while in treated

cultures viable cells were 3 · 107 for Salm. enterica

ser. Newport and 1 · 107 for E. coli, being these in the

0

0·1

1

1000

800

600

400

Ant

ibio

tic a

ctiv

ity A

U m

l–1

200

0

Abs

orba

nce

(600

nm

)

24 48 72 96Time (h)

120 160

Figure 1 Growth curve and antimicrobial production by Serratia pro-

teamaculans 136 on shaker and bioreactor. Fully lines indicating

growth and broken lines indicating antimicrobial production; ,

= growth and antimicrobial production on shaker. ,

= show the growth and antimicrobial production on bioreactor.

Table 3 Serraticin A purification parameters

Purification step

Volume

(ml)

Activity

(AU ml)1)

Protein

(mg ml)1)

Total activity

(AU)

Total protein

(mg)

Specific activity

(AU mg)1)

Purification

factor

Yield

(%)

Soluble extract* 2000 1600 8.5 32 00 000 17 000 188 – 100

Activated charcoal 1500 1600 3 24 00 000 4500 533 3 75

Solid phase extraction 450 3200 0.3 14 40 000 135 10 667 57 45

Preparative RP-HPLC 40 6400 0.1 256 000 4 64 000 340 8

*Corresponding at collected foam after cultivation.

L.A. Sanchez et al. Serraticin A, a novel cold-active antimicrobial

ª 2010 The Authors

Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 5

inoculum order. Both sensitive strains showed a similar

effect over viability.

To avoid over-value because of antimicrobial spontane-

ously resistant-mutant colonies, the counting was done by

replica on LB and LB + serraticin A. The values kept in

the same values, therefore resistant-mutants could not be

isolated at the used antibiotic concentrations. Growth rate

reduction of Salm. enterica ser. Newport and E. coli

AB1133 was 98 and 99% respectively.

Serraticin A effect on E. coli AB1133, ATCC 35218 and

Salm. enterica ser. Newport cell morphology showed a fil-

amentous growth when the cultures were treated with the

antimicrobial at MIC. Higher concentrations of the anti-

microbial (0Æ50 and 2Æ5 lg ml)1) had a similar effect.

E. coli strains filamentation was dependent on growth

phase in which serraticin A was added. Microscope exam-

ination showed a gradual elongation of cells during the

first few hours of incubation. This continued until long

unseptate filaments could be observed after 12 h incuba-

tion. Biomass increment in treated cultures can hence

predominantly be designated to the increase in cell length.

A similar effect on growth-phase dependance has also

been detected when sensitive Salmonella strains were used.

A similar behaviour was observed when E. coli DH5a(recA1) was evaluated to determine if the SOS system was

involved in the filamentous phenotype (induced by ser-

raticin A addition), suggesting that filamentation does not

appear to be correlated with the SOS system response.

General properties

To establish if the antimicrobial activity could be because

of amino acid analogues, isoleucine and methionine were

added to M9 plates and reversibility of the antibiotic

effects of serraticin A was not observed. This fact con-

firms that the antimicrobial effects are not because of

amino acid toxic-analogues like valine or methionine.

Erythrocytes lysis, because of serraticin A haemolytic

activity could not be observed, even when LB plates

supplemented with fresh blood were streaked with Ser.

_proteamaculans 136. Haemolysis halos were not observed

around colonies although the producer strain retained

antimicrobial activity. Drops of a purified solution of

serraticin A did not collapse and appeared like firm drops

on tested oils, whereas the addition of surfactant (like

SDS) caused the drops to spread out. In the same way, the

treatment with kerosene did not demonstrate perdurable

emulsions formation; therefore serraticin A bio-emulsifier

and bio-surfactant activities were not detected.

Concerning to its physical properties, serraticin A

showed to be resistant and active at extreme pH values

(2–12). On the other hand, 100% activity was retained

after the incubation at 55�C, while a 20% of activity was

detected after 1 h of incubation in boiling water.

Serraticin A was soluble in water, however, when

higher concentrations were used, aggregate formation was

observed. Total solubility was obtained in 50% v ⁄ v aceto-

nitrile even at 800 lg ml)1.

When serraticin A samples were subjected to fluores-

cence spectroscopy; emission at different wavelengths

revealed only the presence of tryptophan at kem =

348 nm, when the molecule was excited at kex = 280 nm.

43·5

32·5

2

1

0·5

0 2 4 6Time (h)

8 20 24

Abs

orba

nce

(600

nm

)

Figure 2 Growth viability curves of sensitive Escherichia coli and Sal-

monella enterica ser. Newport. Fully lines indicating growth control;

broken lines indicating antimicrobial treatment; , = growth

control for E. coli AB1133 and S. enterica ser. Newport respectively.

, = antimicrobial treatment curve for E. coli AB1133 and

S. enterica ser. Newport respectively. The arrow indicates the purified

antimicrobial addition at 0.10 lg ml)1.

Table 4 MICs for several sensitive strains

Strain

MIC (lg ml)1) in test

medium

LB M9-G*

Escherichia coli AB1133 0.04 0.32

E. coli ATCC 35218 0.08 0.32

E. coli O157:H7 (stx1,stx2) 0.04 ND

E. coli O26:H11 0.08 ND

Listeria monocytogenes 0.12 0.64

Shigella sonnei 0.08 0.32

Shigella flexneri 0.06 0.16

Shigella enterica ser. Newport 0.03 0.03

Shigella enterica ser. Typhimurium 0.01 0.01

Staphylococcus aureus ATCC 29213 0.08 0.64

Pseudomonas aeruginosa ATCC 27853 0.08 0.08

Bacillus subtilis 168 0.08 ND

ND, Not determined.

*M9-G, M9-Glucose.

Serraticin A, a novel cold-active antimicrobial L.A. Sanchez et al.

6 Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology

ª 2010 The Authors

In the same way, when samples were excited at kex = 274

and 257 nm; fluorescence was not observed at kem = 303

and 282 nm, suggesting tryptophan residues presence.

Mass spectrometry

The mass of the compound was evaluated by LC ⁄ ESI ⁄ Q-

TOF ⁄ MS ⁄ (MS), and spectra from both MS and MS ⁄ MS

(fragmentation pattern) were obtained. Purified serraticin

A subjected to MS revealed essentially three peaks with

identified mass-over-charges of 480Æ2, 959Æ4 and 1455Æ6,

the first one being the [M + H]+ ion form, hence with a

mass of the compound of 479Æ2 g mol)1. Other ions pres-

ent in the spectrum correspond to a dimer formation

[2M + H]+ = 959Æ4 m ⁄ z, and the mass of a multimeric

ion indicates that an additional ammonium molecule is

attached to the structure [3M + NH4]+ = 1455Æ6 m ⁄ z.

When the ions [2M + H]+ and [3M + NH4]+ were

subjected to MS ⁄ MS analysis, they showed the same

fragmentation pattern observed for the monomer. Also

the form [M + Na]+ = 502Æ2 m ⁄ z is conclusively identi-

fied with MS ⁄ MS.

Discussion

In a previous study, we have isolated a group of psychro-

philic and psychrotrophic micro-organisms from Isla de

los Estados-Ushuaia, Argentina which have the capability

to inhibit a wide range of pathogenic strains (Sanchez

et al. 2009). In this study, serraticin A, a cold-active anti-

microbial produced by Ser. proteamaculans 136 has been

produced, purified and partially characterized.

A cross-immunity test revealed that serraticin A is very

different from other well-known microcins assayed, show-

ing cross-immunity only with MccE492; however, they

differ in other important characteristics such as their

molecular weight, inhibition spectra, producer bacterium

and antimicrobial growth phase production. These

cross-sensitivity ⁄ inhibition, suggested us the novelty of

the antimicrobial discovered, named serraticin A. The

cross-immunity observed for both strains, could indicate

an affiliation between them or similarity in the cellular

structure where they are active. Several antibiotic-produc-

ing strains are sensitive to their own antibiotic in the

onset of antibiotic production (Martin and Demain

1980). From the cross-streak test, we have found that

Ser. proteamaculans 136 displayed a constitutive immu-

nity, as it was resistant to exogenous addition of high

concentrations of its own antibiotic. This fact is in accor-

dance with the statements for microcins, where genetic

studies show gene clusters coding immunity factors,

structural aminoacids and signal peptides (Duquesne

et al. 2007).

The isolated bacterium produced the antimicrobial

compound at low temperatures (4 and 8�C); while no

production was detected at higher temperatures (30, 37,

45�C), in spite of the fact that bacterium was able to

grow. This versatility demonstrates a compromise

between cold environments and antimicrobial production;

in recent years, more extensive studies have been carried

out about cold adaptation in both, prokaryote and

eukaryote organisms (Jones et al. 1987; Schindler et al.

1999; Yinghua et al. 2008; Schmid et al. 2009; Gocheva

et al. 2009).

Serratia proteamaculans 136 produced a heat-stable,

cold active, low molecular weight antimicrobial with a

broad inhibitory spectrum; being this, the first communi-

cation of this specie regarding antimicrobials production.

When different media were evaluated for serraticin A pro-

duction, our results indicated that in LB the maximum

production was obtained. This result differ significantly

from other well-known microcin producers, in which

maximum activity was obtained in a minimal or poor

media, and the role of ecological effectors in microbial

complex ecosystems has been postulated (Asensio 1976).

However, these data are in accordance with E. coli

producing MccC7 ⁄ C5 (Fomenko et al. 1996), the only

exception for which optimal antimicrobial production

was detected in rich media.

Serraticin A production was detected as soon as cul-

tures reached stationary phase of growth, this behaviour

was observed for other microcin synthesis, like MccB17

(Hernandez-Chico et al. 1986), MccC7 (Novoa et al.

1986) and MccJ25 (Salomon and Farias 1999) with the

exception of MccE492 which is produced in an active-

form during exponential phase of growth (Orellana and

Lagos 1996). In turn, several publications about this

dependence on growth-stage for synthesis are available for

other products of Serratia sp. producers (Braun and

Schmitz 1980; Givshov et al. 1988; Iusupova et al. 2002).

Hence, the relationship between serraticin A production

and the growth phase was strongly affected by dissolved

oxygen present in culture medium and indirectly the

antimicrobial production. The use of dissolved-oxygen

controlled fermentations to decrease the time of fermen-

tation from 82 to 32 h and also a significant increase in

serraticin A concentration up to 800 AU ml)1. This indi-

cates that antimicrobial synthesis could be associated with

an oxidative metabolic pathway. Similar results for bacte-

riocin production were previously reported for pediocin

SJ-1 (Schved et al. 1993), plantaricin C (Gonzales et al.

1994) and pediocin SA-1 (Anastasiadou et al. 2007).

An efficient method for antimicrobial recovery, based

on foam formation at stationary growth phase and

purification steps was elaborated. Literature describes the

peptide hydrophobicity as an important parameter which

L.A. Sanchez et al. Serraticin A, a novel cold-active antimicrobial

ª 2010 The Authors

Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 7

has influence on foam stability as well as foaming prop-

erty studies of peptidic hydrolysates from several sources

(Popineau et al. 2002; Larre et al. 2006). The hydrophobic

nature of serraticin A clearly explains foam concentration,

activated charcoal and C-18 matrixes adsorption during

the purification steps.

The serraticin A antimicrobial activity because of

amino acid analogues was discarded, as the addition of

isoleucine, methionine and amino acid mixtures to a

minimal culture media did not antagonize the antibiotic

action. Evidences of natural isolates producing and

excreting amino acids which are toxic for E. coli K-12

family strains were previously reported (De Felice et al.

1979; Asensio 1976; Salomon and Farias 1999).

When sensitive cells treated with serraticin A were

microscopically observed, they seemed to be unseptated;

forming long filaments. This effect is commonly associ-

ated with a blocking of DNA replication and SOS system

response; an interruption in chromosome segregation or

inhibition on septation process (Martin and Demain

1980); similar data were reported for MccJ25 (Salomon

and Farias 1999).

Nowadays, it is well known that hydrophobic interac-

tions involving peptides and neutral lipid membranes of

erythrocytes play a significant role in the haemolytic

activity, while hydrophobic and charge interactions

between peptides and negatively charged lipid bacterial

membranes play an important role in antimicrobial acti-

vity (Matsuzaki 1999; Shai 2002). For this fact, haemolysis

provoked by serraticin A was evaluated showing no

haemolytic activity against fresh red blood cells at the

concentrations evaluated.

The purified fraction allows us to determine the mass

of the compound as 479Æ2 g mol)1. Noncovalent com-

plexes such as protonated dimeric ions [2M + H]+ and

multimeric ions [3M + NH4]+ was reported as a common

behaviour in LC ⁄ MS (Chen et al. 2007). Extensive

ESI-MS experiments designed to determine the underly-

ing principles in the formation of proton-bound dimer

ions and sodium-bridged dimer ions from different sub-

stances were reported (Hamilton and Chen 1988; Berthod

et al. 2005). Dimer and multimeric ion formation (959Æ4m ⁄ z, 1455Æ6 m ⁄ z) was observed; however, MS ⁄ MS spectra

come to a point in a single charged ion with 480Æ2 m ⁄ z.

Work is under way to characterize and to get sequence

and final structure of the antimicrobial compound, which

would help to clarify the mode of action of this antibiotic

compound over sensitive cells.

References

Anastasiadou, S., Papagianni, M., Filiousisa, G., Ambrosiadisa,

I. and Koidisa, P. (2007) Pediocin SA-1, an antimicrobial

peptide from Pediococcus acidilactici NRRL B5627:

Production conditions, purification and characterization.

Bioresour Technol 99, 5384–5390.

Aravalli, R.N., She, Q. and Garrett, R.A. (1998) Archaea and

the new age of microorganisms. Trends Ecol Evol 13,

190–194.

Asensio, C.. (1976) Molecular ecology. In Reflections on Bio-

chemistry ed. Kornberg, A., Horecker, B.L., Comudella, L.

and Oro, J.. pp. 235–240. Oxford: Pergamon Press.

Banat, I.M. (1993) The isolation of a thermophilic biosurfac-

tant producing Bacillus sp. Biotechnol Lett 15, 591–594.

Baquero, F., Bouanchaud, D., Martinez-Perez, M.C. and

Fernandez, C. (1978) Microcin Plasmids: a group of

extrachromosomal elements coding for Low-

Molecular-Weight antibiotics in Escherichia coli.

J Bacteriol 135, 342–347.

Berthod, A., Berthod, L. and Armstrong, D.W. (2005) Selectiv-

ity of a native b-cyclodextrin column in the separation of

catechins. J Liq Chromatogr Relat Technol 28, 1669.

Braun, V. and Schmitz, G. (1980) Excretion of a protease by

Serratia marcescens. Arch Microbiol 124, 55–61.

Chen, G., Zhang, L.K. and Pramanik, B.N. (2007) LC ⁄ MS:

theory, instrumentation and applications to small

molecules. In HPLC for Pharmaceutical Scientists ed.

Kazakevich, Y. and Lobrutto, R., pp. 281–346. New York:

John Wiley & Sons.

Cooper, L.Z., Madoff, M.A. and Weinstein, L. (1964) Hemoly-

sis of rabbit erythrocytes by purified Staphylococcal alpha–

toxin I. Kinetics of the lytic reaction. J Bacteriol 87,

127–135.

Davies, J. and Webb, V. (1998) Antibiotic resistance in bacte-

ria. In Emerging Infections ed. Krause, R.M. pp. 239–273.

New York: Academic Press.

De Felice, M., Levinthal, M., Iaccarino, M. and Guardiola, J.

(1979) Growth inhibition as a consequence of antagonism

between related amino acid: affect of valine in Escherichia

coli K-12. Microbiol Mol Biol Rev 43, 42–58.

De Lorenzo, V. and Pugsley, A.P. (1985) Microcin E492 a

Low-Molecular-Weight peptide antibiotic which causes

depolarization of the Escherichia coli cytoplasmic mem-

brane. Antimicrob Agents Chemother 27, 666–669.

Delgado, M., Mouslim, C. and Groisman, E.A. (2006) The

PmrA ⁄ PmrB and RcsC ⁄ YojN ⁄ RcsB systems control expres-

sion of the Salmonella O-antigen chain length determinant.

Mol Microbiol 60, 39–50.

Duquesne, S., Destoumieux-Garzon, D., Peduzzi, J. and

Rebuffat, S. (2007) Mirocins, gene-encoded antibacterial

peptides from enterobacteria. Nat Prod Rep 24, 708–734.

Feller, G., Zekhnini, Z., Lamotte-Brasseur, J. and Gerday, C.

(1997) Enzymes from cold-adapted microorganisms. The

class C ß-lactamase from the antarctic psychrophile

Psychrobacter immobilis A5. Eur J Biochem 244, 186–191.

Fomenko, D.E., Basyuk, E.I., Bezrukov, V.M., Volodin, A.A.,

Metlitskaya, A.Z. and Khmel, I.A. (1996) Identification

and expression of plasmid genes involved in synthesis of

Serraticin A, a novel cold-active antimicrobial L.A. Sanchez et al.

8 Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology

ª 2010 The Authors

microcin C51. Russ J Genet 32, 1149–1155. (Erratum

33:116, 1997).

Garcıa-Bustos, J.F., Pezzi, N. and Asensio, C. (1984) Microcin

7: purification and properties. Biochem Biophys Res

Commun 119, 779–785.

Givshov, M., Olsen, L. and Molin, S. (1988) Cloning and

expression in Escherichia coli of the gene for extracellular

phospholipase Al from Serratia liquefaciens. J Bacteriol 170,

5855–5862.

Gocheva, Y.G., Tosi, S., Krumova, E.T., Slokoska, L.S., Miteva,

J.G., Vassilev, S.V. and Angelova, M.B. (2009) Tempera-

ture downshift induces antioxidant response in fungi

isolated from Antarctica. Extremophiles 2, 273–281.

Gonzales, B., Arca, P., Mayo, B. and Suarez, J.E. (1994) Detec-

tion, purification and partial characterization of plantaricin

C, a bacteriocin produced by a Lactobacillus plantarum

strain of dairy origin. Appl Environ Microbiol 60, 2158–

2163.

Gootz, T.D. (1990) Discovery and development of new antimi-

crobial agents. Clin Microbiol Rev 1, 13–31.

Hamilton, J.A. and Chen, L. (1988) Crystal structures of inclu-

sion complexes of b-cyclodextrin with (S)-(+)- and (R)-

())-febpren. J Am Chem Soc 110, 4379.

Hernandez-Chico, C., San Millan, J.L., Kolter, R. and Moreno,

F. (1986) Growth phase and ompR regulation of transcrip-

tion of microcin B17 genes. J Bacteriol 167, 1058–1065.

Iusupova, D.V., Petukhova, E.V., Sokolova, R.B. and

Gabdrakhmanova, L.A. (2002) The accumulation of

proteins with chitinase activity in the culture media of the

parent and mutant Serratia marcescens strain grown in

the presence of mitomycin C. Mikrobiologiia 71, 635–638.

Jain, D.K., Collins-Thompson, D.L., Lee, H. and Trevors, J.T.

(1991) A drop-collapsing test for screening surfactant-pro-

ducing microorganisms. J Microbiol Methods 13, 271–279.

Jones, P.G., Van Bogelen, R.A. and Neidhardt, F.C. (1987)

Induction of proteins in response to low temperature

in Escheichia coli. J Bacteriol 169, 2092–2095.

Kabuki, T., Uenishi, H., Watanabe, M., Seto, Y. and Nakajima,

H. (2007) Characterization of a bacteriocin, Thermophilin

1277, produced by Streptococcus thermophilus SBT1277.

J Appl Microbiol 102, 971–980.

Kurepina, N.E., Basyuk, E.I., Metliskaya, A.Z., Zaitsev, D.A.

and Khmel, I.A. (1993) Cloning and mapping of genetic

determinants for microcin C51 production and immunity.

Mol Genet Genomics 241, 700–706.

Larre, C., Mulder, W., Sanchez-Vioque, R., Lazko, J., Berot, S.,

Gueguen, J. and Popineau, Y. (2006) Characterization and

foaming properties of hydrolysates derived from rapeseed

isolate. Colloids Surf B Biointerfaces 49, 40–52.

Lavina, M., Gaggero, C. and Moreno, F. (1990) Microcin H47,

a chromosome-encoded microcin antibiotic of Escherichia

coli. J Bacteriol 172, 6585–6588.

Martin, J.F. and Demain, A.L. (1980) Control of antibiotic

biosynthesis. Microbiol Mol Biol Rev 44, 230–251.

Matsuzaki, K. (1999) Why and how are peptide–lipid interac-

tions utilized for self-defense? Magainins and tachyplesins

as archetypes Biochem Biophys Acta – Biomembranes 1462,

1–10.

Mayr-Harting, A., Hedges, A. and Berkeley, R.C.W. (1972)

Methods for studying bacteriocins. In Methods Microbiol

7ª-7 ed. Norris, J.R. and Ribbons, D.W. pp. 315–422.

London ⁄ New York: Academic Press.

Miller, J.H. (1972) Cold Spring Harbor Laboratory. In Experi-

ments in Molecular Genetics. pp. 352–355. New York: Cold

Spring Harbor Laboratory Press.

Moran, A.C., Olivera, N., Commendatore, J., Esteves, J.L. and

Sineriz, F. (2000) Enhancement of hydrocarbon waste bio-

degradation by addition of a biosurfactant from Bacillus

subtilis O9. Biodegradation 11, 65–71.

Moran, A.C., Martinez, A.M. and Sineriz, F. (2002) Quantifica-

tion of surfactin in culture supernatants by hemolytic

activity. Biotechnol Lett 24, 177–180.

Novoa, M.A., Dıaz-Guerra, L., San Millan, J.L. and Moreno, F.

(1986) Cloning and mapping of the genetic determinants

for microcin C7 production and immunity. J Bacteriol 168,

1384–1391.

O’Brien, A., Sharp, R., Russell, N. and Roller, S. (2004)

Antarctic bacteria inhibit growth of food-borne microor-

ganisms at low temperatures. FEMS Microbiol Ecol 48,

157–167.

Orellana, C. and Lagos, R. (1996) The activity of microcin

E492 from Klebsiella pneumoniae is regulated by a micro-

cin-antagonist. FEMS Microbiol Lett 136, 297–303.

Pohlmann, J., Lampe, T., Shimada, M., Nell, P.G., Pernerstor-

fer, J., Svenstrup, N., Brunner, N., Schiffer, G. et al. (2005)

Pyrrolidindione derivatives as antibacterial agents with a

novel mode of action. Bioorg Med Chem Lett 15,

1189–1192.

Popineau, Y., Huchet, B., Larre, C. and Berot, S. (2002)

Foaming and emulsifying properties of fractions of gluten

peptides obtained by limited enzymatic hydrolysis and

ultrafiltration. J Cereal Sci 35, 327–335.

Prangishvili, D., Holz, I., Stieger, E., Nickell, S., Kristjansson,

J.K. and Zillig, W. (2000) Sulfolobicins, specific

proteinaceous toxins produced by strains of the extre-

mophilic archaeal genus Sulfolobus. J Bacteriol 182,

2985–2988.

Reasoner, D.J. and Geldreich, E.E. (1985) A new medium for

the enumeration and subculture of bacteria from potable

water. Appl Environ Microbiol 49, 1–7.

Salomon, R. and Farias, N. (1999) Microcin 25 a novel antimi-

crobial peptide produced by Escherichia coli. J Bacteriol

174, 7428–7435.

Sanchez, L.A., Gomez, F.F. and Delgado, O.D. (2009)

Cold-adapted microorganisms as a source of new antimi-

crobials. Extremophiles 13, 111–120.

Schindler, T., Graumann, P.L., Perl, D., Ma, S., Schmid, F.X.

and Marahiel, A. (1999) The family of cold shock proteins

L.A. Sanchez et al. Serraticin A, a novel cold-active antimicrobial

ª 2010 The Authors

Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology 9

of Bacillus subtilis stability and dynamics in vitro and in

vivo. J Biol Chem 274, 3407–3413.

Schmid, B., Klumpp, J., Raimann, E., Loessner, M.J., Stephan,

R. and Tasara, T. (2009) Role of cold shock proteins

(Csp) for growth of Listeria monocytogenes under cold and

osmotic stress conditions. Appl Environ Microbiol 75,

1621–1627.

Schved, F., Lalazar, A., Henis, Y. and Juven, B.J. (1993)

Purification, partial characterization and plasmid-

linkage of pediocin SJ-1, a bacteriocin produced by

Pediococcus acidilactici. J Appl Bacteriol 74,

67–77.

Shai, Y. (2002) From innate immunity to de-novo

designed antimicrobial peptides. Curr Pharm Des 8,

715–725.

Song, J.H. (2008) What’s new on the antimicrobial horizon?.

Int J Antimicrob Agents 32, S207–S213.

Souza, N.J., Ganguli, B.N. and Reden, J. (1982) Strategies in

the discovery of drugs from natural sources. Annu Rep

Med Chem 17, 301–310.

Tagg, J.R., Dajani, A.S. and Wannamaker, L.W. (1976) Bacte-

riocins of gram-positive bacteria. Microbiol Mol Biol Rev

40, 722–756.

Waters, V.L. and Crosa, J.H. (1991) Colicin V virulence plas-

mids. Microbiol Mol Biol Rev 55, 437–450.

Wenzel, R.P. (2004) The antibiotic pipeline challenges, costs,

and values. N Engl J Med 351, 523–526.

Yaron, S., Rydlo, T., Shachar, D. and Mor, A. (2003) Activity

of dermaseptin K4-S4 against foodborne pathogens.

Peptides 24, 1815–1821.

Yinghua, Z., Yuting, L. and Guicheng, H. (2008) Expression

of cold-shock-protein genes from Lactococcus lactis and

analysis of the cryoprotection function. Acta Microbiol Sin

48, 1203–1207.

Zahner, H. and Fielder, H.P. (1995) The need for new antibi-

otics: possible ways forward. In Fifty Years of Antimicro-

bials: Past Perspectives and Future Trends ed. Hunter, P.A.,

Darby, G.K. and Russell, N.J. pp. 67–83. SGM Symp 53

Cambridge: Cambridge University Press.

Serraticin A, a novel cold-active antimicrobial L.A. Sanchez et al.

10 Journal compilation ª 2010 The Society for Applied Microbiology, Journal of Applied Microbiology

ª 2010 The Authors