Antimicrobial properties of magnesium chloride at low pH ... · The magnesium salt dramatically increased the acidity to a level that was antimicrobial in the presence of anionic

Jour

doi:1

0.16

84/m

rh.2

014.

0362

nal Identification = MRH Article Identification = 0362 Date: September 3, 2014 Time: 12:13 pm

Magnesium Research 2014; 27 (2): 57-68 ORIGINAL ARTICLE

Antimicrobial properties ofmagnesium chloride at low pH in thepresence of anionic basesPía Oyarzúa Alarcón1 , Katherine Sossa1,2 , David Contreras3 , Homero Urrutia1 ,Andreas Nocker4

1 Laboratorio de Biopelículas y Microbiología Ambiental, Centro de Biotecnología, Univer-sidad de Concepción, PO Box 160-C, Concepción, Chile; 2 Facultad de Ciencias Forestales,Universidad de Concepción, PO Box 160-C, Concepción, Chile; 3 Facultad de Ciencias Quími-cas, Centro de Biotecnología, Universidad de Concepción, PO Box 160-C, Concepción,Chile; 4 Cranfield Water Science Institute, School of Applied Sciences, Cranfield University,Cranfield, Bedfordshire, MK43 0AL, United KingdomCorrespondence: Andreas Nocker, Cranfield Water Science Institute, School of Applied Sciences, Cranfield Uni-versity, Cranfield, Bedfordshire, MK43 0AL, United Kingdom<[email protected]>

Abstract. Magnesium is an element essential for life and is found ubiqui-tously in all organisms. The different cations play important roles as enzymaticco-factors, as signaling molecules, and in stabilizing cellular components. It isnot surprising that magnesium salts in microbiological experiments are typi-
cally associated with positive effects. In this study with Listeria monocytogenesas a model organism, we focus however on the usefulness of magnesium (inform of MgCl2) as a stress enhancer. Whereas MgCl2 does not affect bacterialviability at near-neutral pHs, it was found to strongly compromise culturabil-ity and redox activity when cell suspensions were exposed to the salt at acidicpH. The principle was confirmed with a number of gram-negative and gram-
drsenntiCl2insa

positive species. The magnesium saltlevel that was antimicrobial in the prelactate, or acetate, but not TRIS. The astronger than that of NaCl, KCl, or Caor when cells were exposed to MgCl2stress was reinforced by an additional,
viability that needs further examination.disinfection, this observation might suppproperties of MgCl2 for the treatment of san acidic environment), and could contrithe Dead Sea, where Mg2+ and Cl- arehealing properties in a microbiological co
Key words: bacteria, magnesium chloride,stress

Many important findings about the effects of dif-ferent salts and cations on bacterial viability weremade in the first three decades of the 20th cen-tury which many considered to be the ‘golden age’of this type of research. Cations were reported toexert a highly characteristic effect upon bacteria:

ltwegc

To cite this article: Oyarzúa Alarcón P, Sossa K, Contreras D, Uchloride at low pH in the presence of anionic bases. Magnes Res 20

amatically increased the acidity to ace of anionic bases such as phosphate,microbial activity of MgCl2 was much. No effect was observed with MgSO4phosphate buffer with a pH≥5. Acidlt-specific effect of MgCl2 on microbial
Apart from its implications for surfaceort the commonly stated therapeutickin diseases (with healthy skin beingbute to understanding why salt fromthe most abundant cation/anion, hasntext.
sodium chloride, pH, antimicrobial, acid

ow concentrations of given salts were reportedo favor viability, whereas higher concentrationsere associated with growth inhibition [1-3]. Thisffect was visualized by an optimum curve thatenerally held true for all cations ; however, theoncentrations at which the transition between

57rrutia H, Nocker A. Antimicrobial properties of magnesium14; 27(2): 57-68 doi:10.1684/mrh.2014.0362

mailto:[email protected]

10.1684/mrh.2014.0362

10.1684/mrh.2014.0362

Jo 36

ocpbaiwatfc[e

M

B

TLl(((cdABFia1wSh

urnal Identification = MRH Article Identification = 0

P. OYARZUA ALARCON, ET AL.

beneficial and toxic occurred seemed to varygreatly among the different cations. Salts suchas NaCl or KCl, which are typically ‘favorable’(for example, in growth media or physiologicalsaline solution), were reported to be inhibitoryin sufficiently high concentrations. Whereas thelatter correlates with common scientific ‘gut feel-ing’ and the concept of osmotic stress, the sameprinciple could be implied to suggest that evenhighly toxic substances such as HgCl2, PbCl2

and other heavy metals might have a stimulat-ing effect on bacterial growth in sufficiently lowconcentrations. Although there was no clear expla-nation of this empiric observation [3], the effectsof different cations and their specific efficiencies(both in regard to stimulation and inhibition)resulted in the quantitative assignment of ‘spe-cific potency’ factors. Na+ served as a referenceand was assigned a potency factor of 1. Examplesof potencies that were reported include: K+ = 1.2,Mg2+ = 9.4, Ca2+ = 12, Mn2+ = 400, Zn2+ = 700, andCd2+ = 3000 [4].

Whereas these studies focused on the directeffect of the different ions on bacterial viability,this study addresses their impact under differ-ent pH conditions and buffer systems. The projectwas motivated by previous findings that bacterialviability was much more strongly affected by des-iccation in the presence of MgCl2 compared withother salts [5], that magnesium salts have anti-septic properties in treatments involving DeadSea salts, and that the effects of different saltson bacterial viability depended greatly upon pH.Whereas the presence of salts typically had noeffect at a neutral pH (compared to a sample with-out salt), a slightly stronger effect was observed inthe acidic range.

Listeria monocytogenes served as a model organ-ism because of our laboratory’s interest in thisfood-borne pathogen, although Escherichia coli,Salmonella typhimurium, Enterococcus faecium,and Staphylococcus aureus were also used. Plat-ing onto nutrient agar was further supplementedby cultivation-independent measurement of redoxactivity. Over the course of the study, we devel-oped the hypothesis that the impact on bacterialviability is, in part, due to the enhancement ofacid stress. Whereas it is known that an aqueoussolution of MgCl2 is slightly acidic as a result ofhydrolysis [6], the effect is greatly potentiated inthe presence of naturally occurring, strong inor-ganic and organic bases. The idea was addressedin more detail because of the lack of understanding

Ootf

S

T�td1ovr

58

2 Date: September 3, 2014 Time: 12:13 pm

f how salts and different acid-base combinationsan affect microbial viability. The research hasotential implications for enhancing antimicro-ial action and for medical treatments. Low pHnd the presence of inorganic and organic basess a situation typically encountered on the skinhere MgCl2 has been attributed healing andntiseptic properties. Whereas detailed informa-ion about the effect of the salt on skin barrierunction and its anti-inflammatory and biochemi-al properties in eukaryotic cells has accumulated7-9], there is little knowledge with regard to itsffect on microbial viability.

aterials and methods

acterial strains and growth conditions

he bacterial strains used for this study includedisteria monocytogenes (ATCC 19115), Staphy-

ococcus aureus (ATCC 2913), Escherichia coliK-12), Salmonella enterica serovar Typhimuriumisolated from human feces), Enterococcus faeciumisolated from vaginal excretions); where not indi-ated otherwise, strains were from the Facultade Ciencias Biológicas, University of Concepción.ll bacteria were grown on tryptic soy agar (TSA;ecton, Dickinson and Company, Le Pont de Claix,rance) at 30 ◦C. Liquid cultures were obtained by

noculating 15 mL of tryptic soy broth (TSB) into50 mL Falcon tube, shaken at a 45◦ angle for

8 hours, at 120 rpm and at 30◦C. Cell densityas measured in a spectrophotometer (TU-1810plit Beam UV-VIS, Electronic Co Ltd, Shang-ai, China) at 600 nm (OD600) and adjusted to an
D600 = 1.0 by addition of fresh medium. Aliquotsf 1 mL were transferred into 1.5 mL microcen-rifuge tubes, and centrifuged (5,000 rpm, 5 min),ollowed by careful removal of the supernatant.
ample preparation and pH exposure

he bacterial cell pellet was resuspended in 500L of phosphate buffer (100 mM) followed by addi-

ion of 500 �L of either water or solutions ofifferent salts (final salt concentrations of 10, 50,50, 400, or 1,000 mM; final buffer concentrationf 50 mM). Phosphate buffer was adjusted to pHalues between 2 and 11. Alternatively, cells wereesuspended in TRIS, acetate or lactate solutions

Jour 62

pcettms(

F

CBruiwsfbnpwr(fleSiwU

Ca

C

nal Identification = MRH Article Identification = 03

(final concentrations of 50 mM each) adjusted to apH of 4 by the addition of NaOH or HCl. Cells wereexposed to the different solutions without salt orsupplemented with different salts for 20 min each.The salts used in this study were: MgCl2 (cat. nr.BM-0970; Winkler Ltda., Santiago, Chile), NaCl(cat. nr. SO-1455; Winkler Ltda., Santiago, Chile),KCl (cat. nr. 1.04936.1000; Merck), MgSO4 (cat.nr. MA-0980; Winkler Ltda.), and CaCl2 (cat. nr.CA-0520; Winkler Ltda). After salt exposure, cellswere harvested by centrifugation (5,000 g, 5 min)and resuspended in phosphate buffer (50 mM;pH7).

Cultivation on plates

Aliquots of undiluted cell suspensions weretransferred into the top row of sterile, 96-wellNunclonTM plates (Nunc, Roskilde, Denmark).Dilutions were made by stepwise mixing of 10 �Lof cell suspension with 90 �L TSB pre-aliquotedin the lower rows. All dilutions and transferswere made using multichannel pipettes to allowfor a rapid sample processing. Volumes of 3 �Lof the undiluted cell suspension and the differ-ent dilutions were spotted onto TSA Petri dishes,with the highest cell concentration in the top rowof the grid. After brief drying, plates were incu-bated at 30◦C for approximately 20 h. Images ofgrowth patterns were taken with a digital camera(Scion Corporation, Japan), and visualized usingthe Gel-Proanalyzer program (Media Cybernetics,USA).

Redox activity

For the measurement of redox activity, WST-8 (GenScript, Piscataway, USA) and menadione(2-methyl-1,4-naphthoquinone; ACROS Organics,Geel, Belgium) were dissolved in water and DMSOrespectively, to obtain stocks of 10 mM and 8 mMthat were stored at -20 ◦C. WST-8, menadione, andwater were mixed in ratios of 9:1:10 to obtain adetection reagent that was pre-aliquoted in 20 �Lvolumes in black, flat-bottom, 96-well microtiterplates (cat. nr. 5530100; Orange Scientific; Braine-l’Alleud, Belgium). The reaction was started byaddition of 180 �L of 10-fold-diluted cell sus-pensions (prepared previously for the cultivationanalysis) using a multichannel pipettor. Dilutedcell suspensions were used as controls; withoutdilution the signals from untreated control sam-

ascu(U

R

C

Tor

Date: September 3, 2014 Time: 12:13 pm

Antimicrobial properties of MgCl2

les were obtained too quickly. After addition ofells and mixing by pipetting up and down sev-ral times, plates were immediately transferredo a TECAN F200-Pro plate reader (TECAN, Aus-ria). Signals were measured at 450 nm every twoin for a total of three hours. Before every mea-

urement, the plate was shaken for five secondslinear shaking, amplitude of 3).

luorescence microscopy

ells were stained using the LIVE/DEAD®

acLightTM bacterial viability kit (L13152; Invit-ogen, Carlsbad, California). Following the man-facturer’s instructions, SYTO9 and propidium

odide were each dissolved in 2.5 mL of sterileater and subsequently blended to obtain a 2×

taining solution. Cell suspensions were stainedor 15 min in the dark followed by filtration onlack, 0.22 �m, Isopore polycarbonate filters (cat.r. GTBP02500, Millipore, USA). Filters werelaced on a slide using the mounting oil providedith the kit. Images were acquired from a fluo-

escence Olympus BX51 microscope using a 100×UPlanFI, Olympus, USA) objective, FITC and PIuorescence filter sets (ex485/20, em535/25 andx540/20, em635/350, respectively), and a CoolNAP-Pro Digital Kit camera (Media Cybernet-

cs Inc., USA). The software used for visualizationas Image-Pro Plus 5.1 (Media Cybernetics Inc.,SA).

hemical speciation and statisticalnalysis

alculation of the chemical speciation of salts
t different pH values was performed with theoftware “CHEAQS pro V. 2004.1” [10]. Statisti-al ANOVA and Tukey analyses were performedsing statistical software GraphPad Prism 5.0
GraphPad Software Inc., San Diego, California,SA).

esults

omparison of different salts

o compare the effects of different common saltsn bacterial viability, Listeria monocytogenes wasesuspended in phosphate buffer at pH7 and pH3

59

Journal Identification = MRH Article Identification = 036


no

salt KClNaCl Mg-

Cl2

Ca-

Cl2

no

salt KClNaCl Mg-

Cl2

Ca-

Cl2

buffer pH3buffer pH7

Se

ria

l d

ilu

tio

n

C

IrbMspoewsomppitimeaspmroafptihfisptties revealed significant differences between themean increases in activity signals obtained for

7.0 5.66.96.8 3.04.7 2.62.53.02.9Buffer-

salt pH

Figure 1. Effect of different salts on culturabilityof L. monocytogenes at neutral or acidic pH. Cellswere suspended in phosphate buffers of pH7 andpH3 (initial pH without salt) before adding saltsto a final concentration of 400 mM (or an equiv-alent volume of water). Samples were incubatedfor 20 min, followed by resuspension, serial dilu-tion and spotting aliquots on TSA. Measured pHvalues of mixed buffer-salt solutions are indicated.Pictures of representative plates are shown.

without salt, or supplemented with 400 mM ofNaCl, KCl, MgCl2, or CaCl2 (figure 1). As theaddition of salt was found to change the pH ofthe phosphate buffer, pH values of the buffer-saltmixtures (that were used to resuspend cells) areshown, in addition to initial buffer pH values with-out salt. None of the salts affected culturability
at pH7 (initial buffer pH). Neither was survivalcompromised in the sample buffered at pH3 inthe absence of salt, suggesting that exposure topH3 for 20 min did not affect the culturabilityof L. monocytogenes. The presence of the salts atacidic pH on the other hand, resulted in a generaldecrease in growth, with reductions of approxi-mately 2 log units (NaCl and KCl), 4 log units(CaCl2), and 5-6 log units for MgCl2. A substantialdrop in pH when mixing buffer and salt solu-tions was only observed for the divalent cations,offering a potential explanation for the effectsof MgCl2 and CaCl2, but not for the monovalentions.
de

E

TcpsMbwM≥

60


omparison of effect at different pHs

n a next step L. monocytogenes aliquots wereesuspended in phosphate buffer (initial pHetween 2 and 11) in the presence or absence ofgCl2 (figure 2). In the absence of salt, the only

ample where viability was compromised was atH2, whereas culturability was comparable for allther pHs (figure 2A). In the presence of salt, noffect on culturability was seen for the samplesith an initial buffer pH of 5 or higher. Of the

amples with an initial buffer pH of 4 or lowern the other hand, MgCl2 was associated with aarked reduction of survival. The effect can, in

art, be explained by the salt-induced decrease inH (values of buffer-salt mixtures are indicatedn figure 2). On the other hand, culturability inhe pH3 sample (without salt) was higher thann the pH4 sample with salt (pH of buffer-salt

ixture = 3.3), suggesting a pH-independent saltffect of unknown nature. The same observationpplies when comparing culturability of the pH2ample (without salt) with the one of the pH3 sam-le with salt (pH of buffer-salt mixture = 2.5). Aore detailed insight was obtained by measuring

edox activity (figure 2B): whereas the presencef salt did not affect activities in the pH range 6nd 9, a positive effect (more activity) was seenor pH11, and a negative effect (less activity) forH values ≤5. The positive effect of the salt inhe basic pH range is probably due to the acid-fying effect of MgCl2 mitigating the stress atigh pH. Absolute differences in results betweengures 2A and B are due to differences in theensitivity of the two diagnostic methods withlate reader assays being inherently less sensitivehan culture. Statistical analysis of redox activi-

ifferent treatments. Significance (p) was less orqual than 0.05.

ffect of salt concentration

o address the dependence of the effect on the saltoncentration, L. monocytogenes was exposed tohosphate buffer at pH3 (initial pH without salt)upplemented with increasing concentrations ofgCl2 or NaCl (figure 3). Both salts affected via-

ility, but at different concentrations. Comparedith a control sample without salt, the presence ofgCl2 and NaCl reduced growth in concentrations150 mM and ≥400 mM, respectively. This phe-

Journal Identification = MRH Article Identification = 0362 Date: September 3, 2014 Time: 12:13 pm


Ser

ial d

ilutio

n

Buffer-saltpH

Initial pH of buffers (without salt)

WS

T-8

sig

nal

incr

ease

in 3

h

+- +- +- +- +- +- +- +- +- +-

pH11pH10pH9pH8pH7pH6pH5pH4pH3pH2

+- +- +- +- +- +- +- +- +- +-

pH11pH10pH9pH8pH7pH6pH5pH4pH3pH2

A

B 0.8

0.6

0.4

0.2

0

MgCl2

MgCl2

2.0 1.7 3.0 2.5 4.0 3.3 5.0 3.7 6.0 4.7 7.0 5.6 8.0 6.5 9.0 7.0 10.0 7.0 11.0 7.0

Figure 2. Effect of pH in the absence (-) and presence (+) of 400 mM MgCl2 on culturability (A) andredox activity (B) of L. monocytogenes. Cells were exposed to different pHs for 20 min (measured pHvalues of mixed buffer-salt solutions are indicated) followed by resuspension in neutral buffer. A) Serialdilutions of cells spotted on TSA. Pictures of representative plates are shown. B) Effect of salts and pHon redox activity. Values show the increase of WST-8 signals within 3 h. Error bars represent standarddeviations from three independent experiments.

61



nosalt 1505010 400 1000

[MgCl2 ] (in mM) [NaCl] (in mM)

Ser

ial d

ilutio

n

1505010 1000400

buffer pH7 buffer pH3

no salt

+ NaCl

+ MgCl2

no salt

+ NaCl

+ MgCl2

Fsiswhms

oataa

3.0 2.62.93.0 3.02.5 2.72.93.03.02.3Buffer-salt pH

Figure 3. Effect of increasing concentrations ofMgCl2 and NaCl on L. monocytogenes at lowpH. Cells were exposed for 20 min to differentconcentrations of MgCl2 and NaCl at pH3 (ini-tial pH), followed by assessment of culturability(after resuspension of cells in neutral buffer, serialdilution, and spotting of aliquots on TSA) in com-parison with a control without salt. pH values ofmixed buffer-salt solutions are indicated. Picturesof representative plates are shown.

nomenon might have been caused synergisticallyby increasing osmolarity and the aforementionedpH.

Membrane integrity

In addition to assessing culturability and redoxactivity, the impact of the two salts was studied
using fluorescence microscopy by staining treatedcells with SYTO9 and propidium iodide (as partof the LIVE/DEAD® BacLightTM kit). A greencolor indicates an intact membrane, a red colorindicates membrane damage. Whereas all cellsappeared green at pH7 (initial buffer pH), inde-pendent of the presence of salt, a few red cellswere observed at pH3 without salt and in the sam-ple containing NaCl (figure 4). In the sample withan initial pH of 3 supplemented with MgCl2, morethan half of the cells stained red.
Testing other bacterial species

To test whether our observations held true fora wider range of bacterial species, pure cultures

paefoaitos

C

Tp

62

igure 4. Microscopic images of BacLightTM-tained L. monocytogenes exposed to pH7 or pH3n the presence of MgCl2 or NaCl in compari-on with controls without salt. Salt concentrationsere 400 mM each. Green color indicates that cellsave intact cell membranes, red color indicatesembrane damage. A 40 × magnification was cho-

en. Representative pictures are shown.

f E. coli, S. typhimurium, E. faecium, and S.ureus were exposed to pH7 and acidic pHs inhe absence of salt or in the presence of MgCl2

nd NaCl (figure 5). Due to the different toler-nces of the different species towards acid stress,
H3 was chosen for the two gram-negative speciesnd pH2 for the two gram-positive species. No saltffect was observed for the initial buffer pH of 7or any of the bacteria. At acidic pHs the presencef MgCl2 almost completely abolished growth ofll of the species, whereas the effect of NaCl var-ed between species. Whereas it had no effect onhe survival of gram-negative species, the impactn gram-positive species was comparable to thateen with MgCl2.
omparison of different anionic bases

o assess whether the effect is limited to theresence of phosphate, L. monocytogenes was



pH7 pH2

l dilu

tion

pH7 pH2pH7 pH3pH7

E. coli S. typhimurium E. faecium S. aureus

pH3

nosalt NaClMgCl2

nosalt NaClMgCl2

nosalt NaClMgCl2

nosalt NaClMgCl2

nosalt NaClMgCl2

nosalt NaClMgCl2

nosalt NaClMgCl2

nosalt NaClMgCl2

pH w/osalt

.9

riumMer-

d

eMs

D

Ra

Ser

ia

7.0 2.55.6 2.93.0Buffer-salt pH 6.8 7.0 2.55.6 23.06.8

Figure 5. Effect of exposure of E. coli, S. typhimuat pH7, pH 3, or pH2. Cells were exposed to 400salt served as a control. pH values of mixed buffcells were resuspended in neutral buffer, seriallyrepresentative plates are shown.

resuspended in solutions of other acid-base sys-tems (figure 6). All solutions were adjusted topH4 prior to adding salt and to resuspendingcells. No salt effect was observed when cells weresuspended in TRIS-HCl buffer, on the level of culti-vation, or on the level of redox activity. An effect ofMgCl2 on viability was again seen for both tests inthe presence of acetic acid/acetate, and especiallyof lactic acid/lactate. The presence of salt resultedin both cases in a substantial drop in pH.

Effect of the salt anion

To study the effect of the salt anion, experi-ments at neutral and acidic pH were performedin the presence of MgSO4. Samples withoutsalt and supplemented with MgCl2 and NaClserved as controls (figure 7). The NaCl concen-tration (800 mM) chosen was twice as highas the MgCl2 concentration (400 mM) in orderto obtain the same chloride concentration. Asexpected, no effect of either salt was observedat neutral pH on viability (cultivation and redoxactivity) of L. monocytogenes. At acidic pH bothchloride salts had an impact on the two via-bility parameters (with MgCl2 showing more

oteutetpltrhitpc

7.0 1.35.6 2.02.06.8 7.0 1.35.6 2.02.06.8

m, E. faecium and S. aureus to MgCl2 and NaClsalt for 20 min each. A cell suspension without

salt solutions are indicated. Following exposure,iluted, and aliquots spotted on TSA. Pictures of

ffect than NaCl), whereas the sample containinggSO4 appeared comparable with the one without

alt.

iscussion

esults presented in this study indicate thatt low pH and in the presence of inorganic or
rganic bases, extracellular MgCl2 at concentra-ions above 150 mM can result in an antimicrobialffect. As the salt does not affect bacterial viabilitynder ‘normal’ conditions (i.e. near-neutral pH),he action qualifies as synergistic with acidity. Theffect of MgCl2 was substantially stronger thanhat exerted by other chloride salts and might, inart, be explained by enhanced acidification. Theatter can be attributed to the special properties ofhe Mg2+ cation which has, among the biologicallyelevant cations, the smallest ionic radius and theighest charge density [11]. This in turn results
n a strong interaction with the water moleculeshat surround the cation in two shells [12]. Theolarity induced by the high charge density of theentral Mg2+ cation renders water molecules more

63



0.25

0.20

0.15

0.10

0.05

0.00

Free Mg+2

Strongly acidic

2+

2+

-

Weakly acidic

Outer hydrationshell

Aqueous MgCl2 inpresence ofinorganic ororganic bases:

Aqueous MgCl2-:

Inner hydrationshell

• PO43-

• Organic bases: lactate, glycolate, maleate, citrate, etc.

Free Mg+2

[MgCI]+

[MgHPO4]

[MgHPO4]

[MgSO4]

MgSO4MgCI2

A

B

pH pH

Con

cent

ratio

n (m

ol L

-1)

Con

cent

ratio

n (m

ol L

-1)

1 2 3 4 5 6 7 1 2 3 4 5 6 7

0.25

0.20

0.15

0.10

0.05

0.00

Figure 6. Simplified models to explain the roles of the cation and anion. A) Schematic diagram of howinorganic and organic bases increase the acidity of dissolved MgCl2. Mg2+ cations are surrounded by 18water molecules forming two hydration shells. Whereas aqueous MgCl2 is a weak acid with polarizedwater molecules releasing protons, the presence of inorganic and organic bases enhances the release ofprotons resulting in increased acidification. B) Speciation analysis of MgCl2 and MgSO4 in the pH rangebetween pH1 to pH7. The diagram shows the molar concentrations of free Mg2+ and the correspondingionic couple.

64



MgCl2 +- +- +

TRIS-HCl Acetate Lactate

Ser

ial d

ilutio

n

examples of molecules that can increase acidity inthe presence of MgCl2, whereas no effect is seenwith TRIS (figure 7). The reason for the lattercan be seen in the fact that at a pH substan-tially lower than its pKa (8.3), the protonated andthus positively charged TRIS is not a base. Inthe cases of acetic acid (pKa = 4.79) and lacticacid (pKa = 3.86) on the other hand, the acid-baseratio at the experimental pH of 4 produces suf-ficient base molecules to result in acidification inthe presence of MgCl2. Acidification, in turn, shiftsthe acid-base equilibrium towards acidic. Only theprotonated, uncharged forms of these moleculesexert an antimicrobial effect, a mechanism sharedamong weak-acid preservatives [13]. The differ-ence in the pKa values of the two acids is probablyreason why the antimicrobial effect of the salt-acidmixture was substantially stronger in the pres-ence of lactate than with acetate. The acidificationof MgCl2 in the presence of phosphate buffer onthe other hand, might be better explained by thelow solubility of magnesium phosphate salts. Theprotons left behind after precipitation result inacidification. The acidification caused by differentconcentrations of MgCl2 in the presence of phos-phate buffer is shown in table 1.

Apart from the role of the Mg2+ cation, thecomparison between MgCl2 and MgSO4 (figure 8)raises the question how much of the effect can beattributed to the anion. Both salts were comparedin the presence of phosphate buffer; when added

Table 1. Effect of different concentrations ofMgCl2 on pH when added to phosphate buffer ofdefined initial pH. Numbers show final pH valuesafter addition of aqueous salt solution. The finalconcentration of phosphate buffer after MgCl2

4.04.0 4.0 4.03.2 2.9Buffer-salt pH

Figure 7. Effect of different buffers/bases at pH4on culturability of L. monocytogenes in absenceor presence of MgCl2 (400 mM). pH values ofmixed buffer-salt solutions are indicated. Cellswere exposed to different buffer-salt mixtures for20 min followed by resuspension in neutral buffer,serial dilution and spotting of aliquots on TSA.Pictures of representative plates are shown.

likely to donate a proton than in the absence of
salt. Depending on the point of view, Mg2+ ionstherefore qualify as a Lewis acid (accepting elec-trons, based on the Lewis acid-base theory), orhydrated Mg2+ can be seen as an acid (donatingprotons to a base, in agreement with the Brønsted-Lowry acid-base theory). Although there might beother possible explanations, the strong polarity ofMg2+ is likely to cause a slightly acidic pH whendissolving MgCl2 in water.
Whereas the polarity of Mg2+ explains why anaqueous MgCl2 solution is slightly acidic, theacidification is greatly enhanced in the presenceof strong bases resulting in stronger hydrolysis(as schematically summarized in figure 6A). Theacetate or lactate used in this study serve as

addition was 50 mM.

MgCl2concentration(in mM)

initial pH 10 50 150 400 1000pH 11 9.78 8.00 7.31 6.98 6.63pH 10 9.10 8.03 7.37 6.67 6.66pH 9 8.60 7.95 7.35 7.01 6.43pH 8 7.97 7.62 7.14 6.48 6.14pH 7 6.94 6.54 6.09 5.62 5.05pH 6 5.96 5.52 5.11 4.66 4.07pH 5 4.93 4.48 4.11 3.73 3.20pH 4 3.98 3.82 3.59 3.31 2.90pH 3 3.00 2.89 2.60 2.50 2.26pH 2 2.01 1.99 1.87 1.68 1.32

65

Journal Identification = MRH Article Identification = 036


buffer pH7 buffer pH3

nosalt

Mg-Cl2

Mg-SO2

NaCl nosalt

Mg-Cl2

Mg-SO2

NaCl

Ser

ial d

ilutio

n

asbpamicphbtd[maanatii1[itae[tsoirppfrom dead keratinocytes accumulating on the sur-

7.0 6.86.05.6 3.0 3.12.5 2.9Buffer-salt pH

Figure 8. Comparison of effects of MgCl2,MgSO4, and NaCl at neutral and acidic pH on cul-turability of L. monocytogenes. Cells were exposedfor 20 min to 400 mM MgCl2 and MgSO4 or800 mM NaCl at pH7 and pH3 (initial buffer pHwithout salt). An aliquot without salt served asa control. pH values of mixed buffer-salt solu-tions are indicated. Following exposure, cells wereresuspended in neutral buffer, serially diluted,and aliquots spotted on TSA. Pictures of represen-tative plates are shown.

to neutral phosphate buffer, acidification resulted.When added to phosphate buffer at pH3 on theother hand, acidification was only observed withMgCl2. Differences in speciation of the two saltsat low pH might be a potential explanation of thisphenomenon. A computer-based analysis revealeda slightly higher concentration of free Mg2+ withMgCl2 compared to MgSO4 (figure 6B), suggestingthat free Mg2+ might play a role. A full speciationanalysis for MgCl2 when dissolved in phosphatebuffer can be found in table 2 (Appendix). Furtherresearch will, however, be necessary to investigatethe differences between these salts in relation totheir different biological effects.

f(a(iwiaMocu[Aa

66


Although the conditions applied in this studyre ‘artificial’, they demonstrate that differentalts have very different effects on bacterial via-ility and that the effects are pH-dependent. AtH conditions that are critical but sub-lethal, thedditional presence of the salt can render cellsore susceptible, exceeding the tolerable stress

ntensity. Acidic pH is one of nature’s most effi-ient strategies to control microbial growth. LowH is, for example, an essential requirement forealthy skin where the pH has been reported toe, on average, around 4.7 as a result of acidshat are either secreted by the human body or pro-uced by bacteria that are part of normal skin flora14]. Disturbance of this protective mantle is com-

on in skin disorders such as atopic dermatitisnd eczema [15]. An increase in skin pH can bessociated with a general increase in skin colo-ization, a higher abundance of pathogens [16],nd modulated virulence of pathogens [17] andheir adhesion [18]. In the context of this study,t is noteworthy that MgCl2 is the dominant saltn the Dead Sea, with a Mg2+ concentration of.89 M and Cl- representing 99% of all anions19]. The Dead Sea has been credited with heal-ng properties for skin diseases since historicimes. Scientific studies on its effect on microbesre extremely rare, although antimicrobial prop-rties have been described for Dead Sea mud20]. Interestingly, we could show in this studyhat Staphylococcus aureus (which is a commonkin pathogen) was susceptible to the presencef MgCl2 at an acidic pH. Microbiological stud-es in relation to skin seem appropriate for futureesearch. Anionic bases can be expected to beresent on skin in the form of skin excretionroducts, bacterial metabolites and cellular debris

ace. In contrast to the harsh acidic pH conditionstypically pH 3) chosen in this study to look for anntimicrobial effect within a short exposure time20 min), less severe (and thus more physiolog-cally relevant) pH conditions might be effectivehen applying longer exposure times. It is tempt-

ng to speculate that it might be beneficial to applylpha hydroxy acids (AHAs) in combination withgCl2 for skin treatment. AHAs comprise a group

f organic carboxylic compounds (lactic acid, gly-olic acid, malic acid, citric acid, etc.) commonlysed in cosmetics and dermatological applications

21]. Our findings might also shed new light onlCl3 which is a common ingredient of deodor-nts, its effect being attributed to the blocking



Table 2. Molar concentrations (M) of chemical magnesium and phosphate species at different pHvalues. The speciation analysis is based on 400 mM of MgCl2 dissolved in 50 mM phosphate buffer.

pH of solutionChemical species 7 6 5 4 3 2

1632481113

4

aapoebheograluso

free Mg2+ 0.19 0.19PO4

3- 1.48E-09 1.46E-10HPO4

2- 7.03E-05 7.01E-05H2PO4

- 3.77E-05 3.80E-04H3PO4 3.07E-10 3.12E-08MgOH+ 4.32E-06 4.29E-07MgPO4

- 8.18E-07 8.08E-08MgHPO4 1.00E-03 9.99E-04MgH2PO4+ 1.03E-06 1.04E-05MgH3PO4

2+ 1.21E-08 1.22E-07MgH4PO4

3+ 3.69E-11 3.75E-09MgCl+ 0.16 0.16Mg(OH)2 (s) 0 0MgH(PO4)(H2O)3 (s) 4.88E-02 4.84E-02Mg3(PO4)2 (s) 0 0

of sweat glands [22] and a direct toxicity of Al3+

[23]. Extrapolation of the findings of this studymight suggest that, in common with MgCl2, someof the antimicrobial action might also be explainedby the decrease in pH (Al3+ is a stronger Lewisacid than Mg2+) [24]. Although a stronger antimi-crobial effect can be obtained with such metals,an obvious advantage of MgCl2 over more toxicsalts consists in the harmless nature of the saltas regards to human health and environmentalimpact.

Despite the focus on the enhancement of pHstress (which we currently see as the dominantfactor for the explanation of the results obtained),other potential mechanisms of action of MgCl2

cannot be excluded. Figure 2 suggests that acid-
ification might not be the only factor responsiblefor the impact on bacterial viability. Comparingsamples with and without MgCl2, viability wasaffected more strongly at a final pH of 3.3 in thepresence of salt than at pH 3.0 in the absenceof salt. Similarly, survival was lower at pH 2.5with salt than at pH 2 without salt. Additionally,results suggest distinct pH-independent effectsof different salts as demonstrated in the directcomparison of MgCl2 and NaCl (figure 3), wherethe two salts exert different effects on bacterialviability at comparable pH values. These specificsalt effects which seem to add to the pH effectwill need confirmation and further investigation.Factors such as osmotic pressure, transport mech-anisms, and interaction of the ions with proteins
ctnh

tovpchdabga

0.19 0.21 0.21 0.22.48E-11 1.49E-12 1.39E-14 8.50E-17.89E-05 6.06E-05 7.03E-06 4.35E-07.62E-03 2.75E-02 3.94E-02 2.46E-02.87E-06 1.86E-04 3.27E-03 2.06E-02.50E-08 5.84E-09 5.05E-10 5.01E-11.37E-09 9.83E-10 1.00E-11 6.16E-14.00E-03 1.00E-03 1.25E-04 7.76E-06.00E-04 8.40E-04 1.28E-03 7.99E-04.16E-06 8.84E-06 1.59E-06 6.15E-08.45E-07 2.23E-05 4.92E-05 1.92E-050.16 0.18 0.18 0.18

0 0 0 0.46E-02 1.63E-02 0 0

0 0 0 0

nd lipids have to be considered. Apart fromffecting membrane permeability and membraneotential, MgCl2 has been shown (for eukary-tic cells) to interact with a large number ofxchangers and channels found in cellular mem-ranes [7]. As regards to osmotic stress, Listeriaas been reported to be extremely resistant. Liut al. (2005), when examining the salt tolerancef different virulent and avirulent L. monocyto-enes strains, found that all strains tested wereesistant to saturated NaCl (corresponding topproximately 6.1 M) for at least 20 h and possiblyonger, as tested by enumeration of colony-formingnits [25]. This finding is in line with a latertudy showing that no decrease in viability wasbtained when exposing Listeria to a highly con-
entrated NaCl solution (4.8 M) at neutral pH forhree hours [5]. Osmotic stress should thereforeot contribute greatly to the observations reportedere.In summary, this study demonstrates the effect
hat the presence of a ‘harmless’ salt can haven microbial viability. Whereas MgCl2 does notisibly affect cells under ‘normal’ conditions, itsresence can have a severe impact under criti-al (but yet sublethal) conditions. Although weypothesize that the effect is largely due to arop in pH, other factors might be involved in thisntimicrobial activity. Future studies will greatlyenefit from the incorporation of a skin model,iven its potential implications for dermatologicalpplications.

67

Jo 36

1

1

1

1

1

1

1

2

2

urnal Identification = MRH Article Identification = 0


Disclosure

Financial support: this work was, in part, sup-ported by the Chilean Council for Science andTechnology (Project FONDECYT 1101009). Con-flict of interest: none.

References

1. Hotchkiss M. Study on salt action. VI. The stimu-lating and inhibitive effect of certain cations uponbacterial growth. J Bacteriol 1923; 8: 141-62.

2. Winslow CE, Dolloff AF. Relative importance ofadditive and antagonistic effects of cations upon bac-terial viability. J Bacteriol 1928; 15: 67-92.

3. Winslow CE, Haywood ET. The specific potency ofcertain cations with reference to their effect on bac-terial viability. J Bacteriol 1931; 22: 49-69.

4. Winslow CE. The influence of cations upon bacterialviability. Quart Rev Biol 1934; 9: 259-74.

5. Nocker A, Sossa Fernández P, Montjin R, Schuren F.Effect of air drying on bacterial viability: A multipa-rameter viability assessment. J Microbiol Methods2012; 90: 86-95.

6. Shriver & Atkins Inorganic Chemistry. 2009. 5thedn, ed. by P. Atkins P, T. Overton, J. Rourke, M.Weller, and F. Armstrong. 2009. Oxford UniversityPress.

7. Durlach J, Guiet-Bara A, Pagès N, Bac P, Bara M.Magnesium chloride or magnesium sulphate: a gen-uine question. Magnes Res 2005; 18: 187-92.

8. Proksch E, Nissen HP, Bremgartner M, UrquhartC. Bathing in a magnesium-rich Dead Sea salt solu-
tion improves skin barrier function, enhances skinhydration, and reduces inflammation in atopic dryskin. Int J Dermatol 2005; 44: 151-7.
9. Schempp CM, Dittmar HC, Hummler D, et al. Mag-nesium ions inhibit the antigen-presenting functionof human epidermal Langerhans cells in vivo and invitro. Involvement of ATPase and cytokines. J InvestDermatol 2000; 115: 680-6.

10. Verweij W. CHemical Equilibria in AQuatic Systems(CHEAQS) Pro program. The Netherlands: 2008.http://home.tiscali.nl/cheaqs/

11. Moncrief MB, Maquire ME. Magnesium transport inprokaryotes. J Biol Inorg Chem 1999; 4: 523-7.

12. Markham GD, Glusker JP, Bock CW. The arrange-ment of first- and second-sphere water moleculesin divalent magnesium complexes: Results from

2

2

2

2

68


molecular orbital and density functional theory andfrom structural crystallography. J Phys Chem B2002; 106: 5118-34.

3. Lambert RJ, Stratford M. Weak-acid preservatives:modelling microbial inhibition and response. J ApplMicrobiol 1999; 86: 157-64.

4. Lambers H, Piessens S, Bloem A, Pronk H, FinkelP. Natural skin surface pH is on average below 5,which is beneficial for its resident flora. Int J CosmetSci 2006; 28: 359-70.

5. Remitz A, Reitamo S. The clinical manifestations ofatopic dermatitis. In: Reitamo S, Luger TA, Stein-hoff M, eds. Textbook of atopic dermatitis. InformaHealthcare, 2008, p. 1-12.

6. Kurabayashi H, Tamura K, Machida I, Kubota K.Inhibiting bacteria and skin pH in hemiplagia:effects of washing hands with acidic mineral water.Am J Phys Rehabil 2002; 81: 40-6.

7. Rippke F, Schreiner V, Doering T, Maibach HI. Stra-tum corneum pH in atopic dermatitis: impact on skinbarrier function and colonization with Staphylococ-cus aureus. Am J Clin Dermatol 2004; 5: 217-23.

8. Mempel M, Schmidt T, Weidinger S, et al. Role ofStaphylococcus aureus surface-associated proteinsin the attachment to cultured HaCaT keratinocytesin a new adhesion assay. J Invest Dermatol1998; 111: 452-6.

9. Oren A. Life and survival in a magnesium chloridebrine: The biology of the Dead Sea. Proceedings ofSPIE - The International Society for Optical Engi-neering 1998; 3441: 44-54.

0. Zeev M, Henis Y, Alon Y, Orlov E, Sørensen KB,Oren A. Antimicrobial properties of Dead Sea blackmineral mud. Int J Dermatol 2006; 45: 504-11.

1. Kornhauser A, Coelho SG, Hearing VJ. Applica-tions of hydroxyl acids: classification, mechanisms,
and photoactivity. Clin Cosmet Investig Dermatol2010; 3: 135-42.
2. Benohanian A. Antiperspirants and deodorants.Clinics Dermatol 2001; 19: 398-405.

3. Hölzle E, Neubert U. Antimicrobial effects of anantiperspirant formulation containing aqueous alu-minum chloride hexahydrate. Arch Dermatol Res1982; 272: 321-9.

4. Greenwood NN, Earnshaw A, eds. Chemistry of theElements, 2nd edn. Pergamon Press, 1997.

5. Liu D, Lawrence ML, Ainsworth AJ, AustinFW. Comparative assessment of acid, alkali andsalt tolerance in Listeria monocytogenes viru-lent and avirulent strains. FEMS Microbiol Lett2005; 243: 373-8.

http://home.tiscali.nl/cheaqs/

Documents

Antimicrobial properties of magnesium chloride at low pH ... · The magnesium salt dramatically increased the acidity to a level that was antimicrobial in the presence of anionic