From broad-spectrum biocides to quorum sensing disruptors and mussel repellents: Antifouling profile of alkyl triphenylphosphonium salts

RESEARCH ARTICLE

From Broad-Spectrum Biocides to Quorum

Sensing Disruptors and Mussel Repellents:

Antifouling Profile of Alkyl

Triphenylphosphonium Salts

Alberto J. Martín-Rodríguez^1,2*, Jose M. F. Babarro³, Fernando Lahoz⁴, Marta Sansón⁵,

Víctor S. Martín¹, Manuel Norte¹, José J. Fernández¹*

1 Institute for Bioorganic Chemistry “Antonio González”, Center for Biomedical Research of the Canary

Islands (CIBICAN), Department of Organic Chemistry, University of La Laguna, Av. Astrofísico Francisco

Sánchez 2, 38206 La Laguna, Tenerife, Spain, 2 Oceanic Platform of the Canary Islands (PLOCAN),

Carretera de Taliarte s/n, 35214 Telde, Gran Canaria, Spain, 3 Department of Biotechnology and

Aquaculture, Instituto de Investigaciones Marinas CSIC, Eduardo Cabello 6, 36208 Vigo, Spain,

4 Department of Fundamental & Experimental Physics, Faculty of Sciences, University of La Laguna, Av.

Astrofísico Francisco Sánchez s/n, 38206 La Laguna, Tenerife, Spain, 5 Department of Botany, Ecology and

Plant Physiology, Faculty of Sciences, University of La Laguna, Av. Astrofísico Francisco Sánchez s/n,

38206 La Laguna, Tenerife, Spain

OPEN ACCESS

* ajmartinr@ull.es (AJM-R); jjfercas@ull.es (JJF)

Citation: Martín-Rodríguez AJ, Babarro JMF, Lahoz

F, Sansón M, Martín VS, Norte M, et al. (2015) From

Broad-Spectrum Biocides to Quorum Sensing

Disruptors and Mussel Repellents: Antifouling Profile

of Alkyl Triphenylphosphonium Salts. PLoS ONE

10(4): e0123652. doi:10.1371/journal.pone.0123652

Abstract

‘Onium’ compounds, including ammonium and phosphonium salts, have been employed as

antiseptics and disinfectants. These cationic biocides have been incorporated into multiple

materials, principally to avoid bacterial attachment. In this work, we selected 20 alkyl-triphe-

nylphosphonium salts, differing mainly in the length and functionalization of their alkyl

chains, in fulfilment of two main objectives: 1) to provide a comprehensive evaluation of the

antifouling profile of these molecules with relevant marine fouling organisms; and 2) to shed

new light on their potential applications, beyond their classic use as broad-spectrum bio-

cides. In this regard, we demonstrate for the first time that these compounds are also able to

act as non-toxic quorum sensing disruptors in two different bacterial models (Chromobac-

terium violaceum and Vibrio harveyi) as well as repellents in the mussel Mytilus galloprovin-

cialis. In addition, their inhibitory activity on a fouling-relevant enzymatic model (tyrosinase)

is characterized. An analysis of the structure-activity relationships of these compounds for

antifouling purposes is provided, which may result useful in the design of targeted antifoul-

ing solutions with these molecules. Altogether, the findings reported herein provide a differ-

ent perspective on the biological activities of phosphonium compounds that is particularly

focused on, but, as the reader will realize, is not limited to their use as antifouling agents.

Academic Editor: Ali Al-Ahmad, University Hospital

of the Albert-Ludwigs-University Freiburg, GERMANY

Received: November 25, 2014

Accepted: February 19, 2015

Published: April 21, 2015

open access article distributed under the terms of the

Creative Commons Attribution License, which permits

unrestricted use, distribution, and reproduction in any

medium, provided the original author and source are

credited.

Data Availability Statement: All relevant data are

within the paper and its Supporting Information files.

Funding: This study was supported by the Spanish

Ministry of Economy and Competitiveness (http://

www.mineco.gob.es/), SAF2011-28883-C03-01 (JJF),

CTQ2011-28417-C02-01/BQU (VSM), AGL2010-

16464 (JMFB), MAT2013-40852-R (FL); Spanish

Ministry of Education, Culture and Sport (http://www.

mecd.gob.es/portada-mecd/), CEI 10/00018 (MN).

The funders had no role in study design, data

collection and analysis, decision to publish, or

preparation of the manuscript.

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

1 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Competing Interests: The authors have declared

that no competing interests exist.

Introduction

Marine biofouling is a deleterious process that imposes a plethora of costly problems to human

activities in the ocean, particularly to the shipping industry [1]. It is estimated that the costs as-

sociated to this phenomenon already exceed US $200 billion every year [2]. Biofouling is a

complex phenomenon that involves a wide array of organisms, from microbes to invertebrates.

It is often depicted as a successional process with four main stages, illustrated with an hypo-

thetical material surface that is submerged in the sea (Fig 1): 1) adsorption, from the first sec-

onds after immersion, of organic particles onto the submerged surface, with the development

of a so-called ‘conditioning film’ that constitutes the molecular fouling and promotes the 2) ar-

rival of primary colonizers, initially (first 24 h) pioneer motile bacteria and, within the first

days, an array of microorganisms, with bacteria and benthic diatoms as the main representa-

tives, that form complex multispecies biofilms (microfouling) and tend to promote the 3) set-

tlement of macroalgal zoospores (e.g. ulvophycean) and 4) invertebrate larvae (e.g. mussel

pediveligers, barnacle cyprids) that end up forming a complex macroscopic fouling communi-

ty. Consequently, to characterize the antifouling profile of any given substance accurately, it is

necessary to conduct bioassays with key organisms that are representative of the different

stages of this phenomenon [3].

Over the second half of the past century, the problem had been relatively under control

through the use of organotin compounds, in particular bis-(tris-n-butyltin)oxide (TBTO).

However, increasing evidence of the detrimental environmental impact of these chemicals

arose from the 1980s [4–6]. That increasing concern led to a stepwise prohibition of organotin

compounds in antifouling paint formulations that was established by the International Mari-

time Organization and fully entered into force in September 2008.

Current antifouling research focus on the design of less harmful solutions that include:

the tethering of biocides to the coating matrix, which remain fixed or are released to the

Fig 1. Schematic representation of the biofouling process.

doi:10.1371/journal.pone.0123652.g001

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

2 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

surroundings in a controlled fashion (self-polishing antifouling systems) [1,7], low-surface-en-

ergy fouling-release coatings [8], hybrid antifouling-fouling release coatings [9], engineered

microtopographies [10,11], enzyme-based coatings [12,13] or natural product antifoulants

[14,15]. As biofouling begins with the formation of a biofilm, which in turn conditions the sub-

sequent settlement of macroscopic organisms (Fig 1) [1,16], the use of inhibitors of bacterial

cell-to-cell communication or quorum sensing (QS) has emerged as a non-toxic mechanism

for the control of the early stages of the biofouling process [17].

Cationic compounds, such as quaternary ammonium and phosphonium salts, have been

used since the 1930s as disinfectants and antiseptics and they have stood out due to their

broad-spectrum antimicrobial activity and relatively low toxicity [18]. In comparison, phos-

phonium salts display better antimicrobial properties than their ammonium counterparts,

either as single molecules in solution [19] or in their polymeric forms [20]. Phosphonium cat-

ions also exhibit higher thermal stabilities and ionic conductivities [21,22]. Their use has not

been restricted to industrial environments and even in the biomedical field, phosphonium salts

have been proved to display better antitumoral activities and lower toxicities than ammonium

salts [23].

Over the last years, the abovementioned properties had led to the inclusion of phosphonium

moieties in polymers for biomedical applications [22,24], water treatment [25], food packaging

[26] or antifouling purposes [27,28]. They have been included in rubbers [29] or clays [30,31]

to confer antimicrobial properties on these materials. As alkyl-triphenylphosphonium cations

readily trespass the lipid bilayers, they have been employed as carriers to deliver drugs and

therapeutics inside the mitochondria [32,33] or genes inside cells [34]. Other fields of applica-

tion of phosphonium cations include the supramolecular chemistry [35,36], and the design of

smart materials due to their self-assembly properties [37].

Despite such an intensive research, there was still a missing piece: a comprehensive assess-

ment of their profile as antifouling agents with relevant fouling species that goes beyond the

classical view of these compounds as broad-spectrum antimicrobials. To that end, 20 alkyl-tri-

phenylphosphonium salts (Fig 2), differing in the length and functionalization of their alkyl

chains, were selected for their study in a multidisciplinary approach that included their biologi-

cal evaluation towards a wide panel of marine fouling organisms, an innovative use as bacterial

QS disruptors and mussel repellents, and an enzymatic and fluorescence-based characteriza-

tion as tyrosinase inhibitors, an enzyme that plays an essential role in mussel byssal production.

Overall, the findings reported herein provide a detailed description, based on the chemical

structure, of the profile of these molecules as active ingredients in antifouling coatings not nec-

essarily as biocides but also as non-toxic repellents and disruptors of key processes for the bio-

logical colonization of immersed substrata.

Materials and Methods

Chemicals

A total of 20 triphenylphosphonium salts were used in this study (Fig 2). Compounds 1–3, 11

and 15 were obtained from the chemical library of the Institute for Bioorganic Chemistry

“Antonio González”, University of La Laguna. Compounds 4–7 were purchased from Sigma-

Aldrich, and compounds 9, 10, 14 and 16 were acquired from Alfa-Aesar. Compounds 8, 13,

and 17–20 were synthesized according to [38]. Briefly, a mixture of triphenylphosphine and

the corresponding alkyl bromide was refluxed for 48 h. The resulting salt was filtered, washed

in ether (x3) and dried. General methodological information and spectroscopic data relative to

the synthetic products are available in S1 File.

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

3 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 2. Structures and cLogP values of the compounds used in this study.

doi:10.1371/journal.pone.0123652.g002

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

4 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Octanol/water partition coefficients were calculated with MarvinSketch 6.0.0 (ChemAxon

Ltd., Budapest, Hungary). Unless otherwise stated, product samples were dissolved in dimethyl-

sulfoxide (DMSO) at a concentration of 40 mM and conserved at -20ºC. TBTO (Sigma-Aldrich)

was used as antifouling standard.

General considerations for bioassays

All the biological assays were run by triplicate. Unless otherwise stated, the cut-off concentra-

tion was set at 100 μM and the maximum proportion of solvent was 0.25% (v/v). A product

causing no effect at this concentration was considered not active. For the active products, serial

dilutions were performed to calculate the half-maximal inhibitory concentrations (IC₅₀) or the

minimal inhibitory concentrations (MIC). Unless otherwise stated, the test concentrations

were 100, 50, 25, 10, 5, 2.5, 1, 0.5 and 0.1 μM. Dose-response curves were adjusted with Graph-

Pad Prism 5 software using a four-parameter non-linear regression model.

Bioassays with marine bacteria

Six strains of marine bacteria were purchased from the Spanish Type Culture Collection

(CECT): Cobetia marina CECT 4278, Pseudoalteromonas atlantica CECT 579, Shewanella

algae CECT 5021, Vibrio alginolyticus CECT 521, Vibrio anguillarum CECT 4347 and Cellulo-

phaga lytica CECT 5014. Growth inhibition was assessed in Marine Broth (Conda) according

to CLSI guidelines for broth microdilution susceptibility testing [39]. The incubation time was

24 h and the incubation temperature was 30ºC. Flat-bottom microtiter plates were employed

(Nunc 167008). Bacterial growth inhibition was determined by measuring the optical density

at 625 nm (OD₆₂₅).

Bioassays with marine-derived fungi

Three strains of marine-derived fungi were kindly provided by Dr. Á. Trigos (University of Ve-

racruz, Mexico). The strains were isolated from reef organisms collected from the Veracruz

Reef System and identified as Aspergillus sp., Fusarium sp. and Alternaria sp. [40].

Fungal strains were cultured in Potato-Dextrose Agar (PDA) supplemented with filtered

seawater (FSW) (50% v/v) at 26ºC for 5 days prior to experiments. Bioassays for fungal

growth inhibition were conducted in 96-well plates in RPMI-1640 medium (+ L-Glutamine,—

NaHCO₃, Biochrom) supplemented with 2% glucose according to the EUCAST protocol [41].

Briefly, fungal inocula were prepared by addition of 5 ml of sterile saline solution to the cul-

tures. Gently swabbing released conidia. Inocula were adjusted to 2–5 x 10⁶cfu ml^-1by count-

ing in a haemocytometer and diluted 1:10 in RPMI-1640 medium before dispensing into the

wells (100 μl) in the presence of the appropriate dilution of the test products in RPMI (100 μl).

The plates were incubated for 5 days at 26ºC. The calculation of the half-maximal inhibitory

concentrations (IC₅₀) was performed using OD₄₀₅as endpoint [42].

Bioassays with diatoms

Five strains of benthic diatoms, Nitszchia sp. BEA 0497, Navicula cf. salinicola BEA 0055,

Phaeodactylum tricornutum, Cylindrotheca sp. and Amphora sp. were used to study the effect

of the compounds on microalgal growth. The Nitzschia and Navicula strains were purchased

from the Spanish Bank of Algae (Marine Biotechnology Center, University of Las Palmas de

Gran Canaria). The other three strains were kindly provided by Dr. G. Courtois (University of

Las Palmas de Gran Canaria). Diatoms were cultured at 19 1ºC in Erlenmeyer flaks (250 ml)

containing 150 ml of Guillard’s F/2 medium, and subjected to a photoperiod of 18:6. Tests

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

5 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

were run in 48-well plates. Inocula were prepared by adjusting diatom concentration to

2–4 × 10⁶cells ml^-1using a Neubauer chamber. Test products were dissolved in F/2 medium

(500 μl) to which diatom inocula (500 μl) were added. Thus, the final assay volume was 1 ml

and the starting cell densities were ca. 1–2 x 10⁶diatoms ml^-1. Plates were incubated under the

abovementioned conditions for 5 days and then chlorophyll-a (Chla) was quantified.

In order to extract Chla, the content of each well was transferred to a microcentrifuge tube

and centrifuged at 10,000 rpm for 10 min. The supernatants were discarded and 200 μl of

DMSO were added to the pellets. The tubes were incubated at 65ºC for 2 h in total darkness

and vortexed every 30 min. Then, the content of each tube was transferred to a 96-well plate

and the amount of Chla was determined spectrophotometrically [43]. Path length correction

factor for the DMSO extracts was determined [44].

Bioassays with macroalgal spores

To evaluate the effect of the phosphonium salts on the germination of macroalgal spores, Gayr-

alia oxysperma (Kützing) K.L.Vinogradova was selected as model organism. G. oxysperma

(Ulotrichales, Chlorophyta) is a cosmopolitan member of the Ulvophyceae [45]. Unlike Ulva

species, which produce biflagellate male and female gametes, as well as quadriflagellate spores,

G. oxysperma only produces biflagellate spores. This is an advantage as bioassays can be con-

ducted straightforwardly without the need to distinguish between gamete-producing and zoo-

spore-producing plants. G. oxysperma specimens were collected from the upper eulittoral at El

Médano, Tenerife, Canary Islands (UTM 28R 348359 3102405). Voucher specimens are depos-

ited as TFC Phyc 14912 (Herbarium University of La Laguna). Fresh fertile fragments were se-

lected and placed in Petri dishes. Spores were then released in Von Stosch Solution (VSS) by

the osmotic method [46,47]. Bioassays were conducted in flat-bottom 96-well plates as de-

scribed by Chambers and co-workers [48], with slight modifications. Each well was filled with

50 μl of the appropriate dilution of the products in VSS to which 50 μl of spore inoculum (ca.

2 x 10⁵spores ml^-1) were added. Plates were incubated at 19 1ºC for 6 days. After the incuba-

tion time, the bottom of each well was examined for the presence of germinated spores with an

inverted microscope. A spore was considered as germinated when the germ tube was visible.

The MIC was recorded as the lowest concentration inhibiting spore germination.

Artemia salina tests

Artemia salina cysts (INVE Aquaculture, Ghent, Belgium) were hatched in brackish water

(30 ‰ salinity) at 28ºC with aeration and under constant light. Newly hatched instar I nauplii

were harvested for bioassays. Tests were conducted in 96-well plates (15 5 nauplii per well,

test volume = 200 μl). The number of dead and alive individuals was recorded after 24 h of in-

cubation at 28ºC and 24-hour photoperiod.

Quorum sensing bioassays with Chromobacterium violaceum

The reporter strain C. violaceum CVO26 (CECT 5999) was used to screen the ability of the

products to interfere with violacein production, a QS-regulated phenotype. C. violaceum

CVO26 is a mini-Tn5 mutant that depends on an exogenous source of autoinducer (AI) (N-

hexanoyl homoserine lactone, HHL) for violacein production. C. violaceum CVO26 was cul-

tured in LB broth (Sigma-Aldrich) supplemented with 25 μg ml^-1kanamycin. Inocula were

prepared by dilution (1:100) of an overnight culture of the reporter strain [49]. One hundred

microliters of inoculum with HHL (Sigma-Aldrich, 6 μM) or without the AI were added to

96-well plates containing 100 μl of the appropriate dilutions of the test products in LB. Thus,

two sets of plates were prepared. One set of plates (without HHL) was used to evaluate the

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

6 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

effect of the products on bacterial growth, whereas the other set of plates (with HHL) was used

to evaluate their effect on violacein production. Both sets of plates were incubated at 30ºC with

agitation (150 rpm) for 18 h.

Growth inhibition was quantified in the first batch of plates by re-suspending bacterial pel-

lets and measuring the OD₆₂₅. From the second batch of plates violacein was extracted and

quantified [50]. Briefly, the plates were dried overnight (60ºC) and violacein was re-solubilized

by the addition of 200 μl of DMSO. The plates were shaken for 3 h and then the OD₅₉₀was

determined.

Quorum sensing bioassays with Vibrio harveyi

V. harveyi BB120 (wild type strain), BB170 (luxN::Tn5kan), BB886 (luxPQ::Tn5kan) and

BB721 (luxO::Tn5lacZ) were acquired from ATCC. To screen the ability of the compounds to

interfere with QS, the test products were serially diluted in 100 μl of Autoinducer Bioassay

(AB) medium [51] using white, clear-bottom 96-well microtiter plates (Costar 3610) as assay

platform. Aerobic bacterial cultures were incubated overnight (30ºC) and diluted 1:50. One-

hundred μl of the diluted cultures were dispensed inside each well in the microtiter plate. That

gave a starting cell density of 1–2 x 10⁷cfu ml^-1. The plates were covered with a sterile sealing

film. Luminescence and OD₆₀₀were monitored every 15 min over 18 h with a multimode plate

reader (Perkin-Elmer EnSpire) in order to correlate the effects of the products on both the

growth and bioluminescence kinetics.

Tyrosinase inhibition assays

Mushroom tyrosinase (EC 1.14.18.1) was purchased from Sigma-Aldrich. Tyrosinase inhibi-

tion assays were conducted as described in [52] with slight modifications. First, in order to

characterize the Michaelis-Menten parameters of the enzyme kinetics, 100 μl of a mushroom

tyrosinase solution (25 U) in sodium phosphate buffer (50 mM, pH 6.5) were pipetted in each

well of a flat-bottom microtiter plate (Nunc 167008). Two-fold dilutions of L-Dopa, from

4.8 to 0.075 mM (final concentrations) were mixed (100 μl) with the enzyme solution. The en-

zymatic reaction was followed by measuring the absorbance at 475 nm every 30 s over 15 min,

with shaking (150 rpm) between measurements. Temperature was kept constant at 30ºC. The

values of K_Mand v_maxunder these conditions were obtained from Lineweaver-Burk plots.

For tyrosinase inhibition assays, each well of the microtiter plate was filled with 100 μl of a

mushroom tyrosinase solution and the appropriate amount of the test substance dissolved in

DMSO. The compounds were pre-incubated with the enzyme at 30ºC for 10 min. Subsequent-

ly, the enzymatic reaction was triggered by addition of 100 μl of L-Dopa (1.2 mM test concen-

tration). Formation of dopachrome was followed by measuring the absorbance at 475 nm as

described above. Kojic acid (Sigma-Aldrich) was used as positive control.

Fluorescence spectroscopy analysis

Fluorescence measurements of tyrosinase solutions were performed using an Edinburgh Instru-

ments LifeSpec II fluorescence spectrometer, exciting the complexes at 280 nm with an Edin-

burgh Instruments EPLED-280 subnanosecond pulsed diode source (typically pulse width at

half maximum around 860 ps, repetition rate 10 MHz) and using Edinburg Instruments F900 ac-

quisition software. A multichannel plate photomultiplier was used as the detector using single

photon counting technique. Lifetime estimation was made using instrument response function

(IRF) reconvolution analysis with FAST software by Edinburgh Instruments, providing a tempo-

ral resolution around 0.2 ns. A blank medium without enzyme was used to confirm that the col-

lected fluorescence comes from the enzyme molecules and any influence of the fluorescence of

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

7 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

the quencher and buffer components on the enzyme fluorescence is negligible. Enzyme solutions

(2 mg ml^-1) were prepared in phosphate buffer (50 mM, pH 6.5) in the presence of different con-

centrations of compound 16 (0–30 μM). A denatured enzyme sample was prepared by mixing

the enzyme solution with an extremely high quencher concentration (1 mM), which provides a

surfactant effect.

Mussel foot retraction assay

The mussel Mytilus galloprovincialis was used as target organism. Individuals of 4.0–4.5 cm

shell length were collected at intertidal rocky shore of the outer Ría de Vigo (NW Spain). Ani-

mals were kept in running seawater aquarium tanks with open-flow design for few days and

fed daily by pulses with a mixture of the microalgae (Tahitian Isochrysis aff. galbana) and sedi-

ment from the seafloor below the mussel rafts.

Mussel foot retracting assay was conducted following the method reported by Hayashi and

Miki [53]. Briefly, subsamples of individuals from the maintenance tanks were transferred to

containers filled with 1-μm FSW. After 30 min, the posterior adductor muscle (PAM) was cut

to open both animal´s valves as shown in S1A Fig, and byssus filaments were removed from

their insertion with the soft tissues. Animals were disposed individually on flexible innocuous

plasticine basis as a holder still for 15 min in FSW. The latter disposition avoided animal´s ma-

nipulation before and after each test solution was dripped directly on the foot of immobilized

animals. FSW was removed and animals exposed their internal cavities in dry conditions for

the dripping actions but maintaining a humidity layer over soft tissues.

Initially two different controls (negative and positive) were considered. As negative control,

2.5 μl of DMSO in 0.2-μm FSW was used. Only animals that did not react were considered fur-

ther. After the negative control test, the animals were disposed back in 1-μm FSW for 15 min.

With regard to the positive control, 1000 ppm of CuSO₄was used. For the latter case, only ani-

mals that did react to such positive control were used. After this second control, animals were

disposed back in 1-μm FSW, this time for 25 min in order to remove completely any rest of cop-

per. All animals considered for the experiment have accomplished the two controls satisfactorily.

Different concentrations (6.25, 12.5, 25, 50, 100 and 200 μM) of two test compounds, 3 and

16 were used. Stocks of 40 mM for the two compounds under investigation were prepared in

DMSO. Posterior dilution actions to get the distinct concentrations desired were prepared with

0.2-μm FSW. Concentrations were tested in ascending order from 6.25 to 200 μM. Animals

were kept in 0.2-μm FSW for 15 min at each time interval between sampling with the aim to

wash the internal soft tissues. After each removal of seawater, animals were ready to use again.

Eight series of ten animals each were considered for retracting muscle foot analysis and each

test compound. Each potential foot-reacting substance was individually dripped on the foot of

the mussels (S1B Fig) and the animal’s reactions were noted. Activity was reported for each test

compound concentration as the percentage of mussels showing reaction at several concentra-

tions of the test compound according to the formula: (No. of reacted mussels / No. of total

mussels) x 100. Final data are given as means SD.

Results and Discussion

In this study, the antifouling profile of a collection of 20 alkyl triphenylphosphonium salts has

been evaluated in detail. These compounds have a common triphenylphosphonium moiety

and differ in the length of and chemical functionalities present in the alkyl chain (Fig 2).

To that end, the compounds were assayed in a panel composed by target microscopic and mac-

roscopic fouling species, as well as an acute toxicity model (A. salina), QS models (C. viola-

ceum, V. harveyi), and an enzymatic model (tyrosinase). In the following sections, organized

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

8 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

according to the kind of activity displayed, insights about their modes of action and structure-

activity relationships are discussed.

Biocidal activity

The biocidal profile of the compounds evaluated is summarized in Table 1. The following lines

discuss the results showed therein. With respect to their bioactivity, a borderline can be estab-

lished below C₇alkyl chains. These short-chain compounds were in general not active inhibit-

ing bacterial or fungal growth with a few exceptions with P. atlantica and C. lytica. These

products were also unable to inhibit the germination of G. oxysperma spores (only compounds

1 and 5 exhibited activity at 100 μM) and displayed little or null toxicity in A. salina nauplii,

with the only exception of 5 that exhibited an unusually low LC₅₀of 17.2 μM. Apparently, the

secondary amine bound to the phosphorus atom is the responsible for this activity, since it is

the only difference between 5 and 7 (LC₅₀> 100 μM). Conversely, diatoms were more sensitive

to the action of the products, and IC₅₀values in the range 8.2–81.5 μM were determined for

this subset of compounds.

A clear shift in terms of activity is observed for products with alkyl chains of C₇or above. In

the antibacterial tests, the lowest IC₅₀values were generally achieved by compounds 12–14. In-

deed, the activity reached a maximum for alkyl chains between C₁₀and C₁₆, with a rapid loss of

activity for longer chains. The activity for this group of compounds was notably better than

that displayed by TBTO. However, the presence of a carboxyl group at the end of the hydrocar-

bon chain (compound 19) led to a total loss of activity, not only with bacteria but in all the bio-

assays conducted. Activity towards fungi followed an analogous pattern to that described for

bacteria. In this case, the lowest IC₅₀values were recorded for compounds 13–15. Interestingly,

the presence of an additional triphenylphosphonium moiety (compound 20) lowered the IC₅₀

to values below that of TBTO. In diatoms, the decrease in the activity for very long alkyl chains

was only evident in Nitzschia sp. Low-μM or even nM IC₅₀values (compounds 18 and 20, P.

tricornutum) were recorded. These results in microalgae contrast with the relatively high MICs

(50 μM, compounds 9–16) necessary to suppress macroalgal spore germination. Only 20

achieved an activity equivalent to that of TBTO. In A. salina tests, the activity increased step-

wise with the length of the hydrocarbon chain. In fact, compounds 17 and 18 exhibited a very

potent activity, with LC₅₀values below 500 nM, and all compounds with C₉alkyl chains or

higher surpassed the activity of the TBTO control, with the abovementioned exception of com-

pound 19, which was inactive.

The mode of action of cationic biocides has been intensively described, primarily as antibac-

terial agents, and it is often depicted in six sequential stages: 1) adsorption on the bacterial cell

wall, caused by electrostatic interaction between the negatively-charged bacterial surface and

the cationic compound [9]; 2) diffusion through the cell wall; 3) attachment to the cell mem-

brane; 4) lysis of the cell membrane; 5) leakage of the cytoplasmic contents and 6) cell death

[54]. The ability to kill microbial cells by contact has promoted the use of quaternary ammoni-

um and phosphonium compounds as disinfectants and antiseptics in solution over decades or,

more recently, incorporated into materials surfaces and polymers for the avoidance of microbi-

al biofilm formation [55].

There is a clear correlation between the length of the alkyl chain and the biocidal activity

(Table 1 and Fig 3). This correlation was highly dependent on the target cell. This phenomenon

has often been linked to the increased hydrophobicity of the compounds with longer alkyl

chains and the subsequent decrease in the critical micellar concentration (cmc), thus facilitat-

ing a surfactant mode of action [19]. It should be recalled that the IC₅₀values determined for

the products used in this study are well below their cmcs [56], thus cell death is unrelated to a

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

9 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

10 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 3. Correlation between the activity and the cLogP of the compounds evaluated in this study. Data correspond to three representative organisms:

Vibrio alginolyticus (blue squares), Fusarium sp. (red triangles) and Phaeodactylum tricornutum (green circles). Dotted lines show the cLogP of each

compound, whose ID. number is indicated above. Missing symbols indicate lack of activity.

doi:10.1371/journal.pone.0123652.g003

surfactant activity. Fig 3 shows the activity of the compounds in function of their cLogP value in

three tested microbial cells. For instance, for the Gram-negative bacterium V. alginolyticus, the

optimal activity is achieved by compound 15, whereas alkyl chains of C₁₈and above caused com-

plete loss of activity. These findings are in fair agreement with previously reported data [57]. For

the filamentous fungus Aspergillus sp., a similar trend is observed in the homologous series, al-

though the compounds with the longest alkyl chains remained active. However, for the diatom

P. tricornutum the activity improved as cLogP increased. This is likely due to the different nature

and chemical composition of the microbial cell wall in each case. Nonetheless, cLogP alone is

not the only factor affecting the biocidal effect of quaternary phosphonium compounds in solu-

tion. Two examples are evidenced in this study: 19 and 20. The incorporation of a carboxyl moi-

ety at the end of the alkyl chain (19) causes a decrease in cLogP from 9.06 (11) to 7.74 (19). This

value is still above the ‘virtual threshold’ established by heptyl triphenylphosphonium bromide

(8) as discussed previously. However, 19 is inactive, a fact likely derived from a different state of

aggregation of the compound in solution. At alkaline pH (7–8), the carboxyl group exists as a

carboxylate and may interact with the positively charged triphenylphosphonium moiety. As

alkyl-triphenylphosphonium compounds need to trespass the biological membranes, this config-

uration leads to the inactivation of the compound. Conversely, 20 has two positive charges and

interacts strongly with the negatively-charged cell membranes.

Quorum sensing inhibition

Biofilm formation is a first stage in the biofouling process (Fig 1). The development of biofilms

is a crucial step in the whole phenomenon as they alter the microtopograhy and physicochemi-

cal properties of the surface and, as a consequence of microbial metabolism and signaling,

they change the chemical nature of the surface and produce an array of chemical cues that can

deter, but more frequently attract, the arrival of secondary and tertiary colonizers [58].

QS, or bacterial intercellular communication, regulates bacterial gene expression as a func-

tion of population density [59]. The process relies on the production, excretion and detection

of signaling molecules (generally, small molecules such as acyl homoserine lactones in Gram-

negatives and small peptides in Gram-positives) so that, when a threshold extracellular

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

11 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

concentration of AIs is achieved, bacterial gene expression is altered. Among the QS-regulated

processes are biofilm formation and maturation [60–63]. There is clear evidence that the settle-

ment of macrofouling organisms is modulated by the development of bacterial biofilms, as it is

the case, for instance, of the polychaete Hydroides elegans [64,65], or Ulva species, whose zoo-

spores can detect acyl homoserine lactones as positive cues for settlement [66]. For this reason,

QS inhibitors have been proposed over the last few years as a mean to disrupt, or at least delay,

biofouling [67,68]. QS inhibitors are an attractive antifouling treatment since they constitute a

means of controlling biofilm formation without exerting a selective pressure (i.e. toxicity) on

bacterial populations, thus avoiding bacterial resistance towards biocidal treatments. Indeed,

bacterial cells in biofilms are up to 1000 times more resistant—or, more precisely, more toler-

ant- to chemical treatment than their planktonic counterparts [69].

To evaluate the effect of the alkyl triphenylphosphonium salts on QS-regulated phenotypes,

an initial screening at the cut-off concentration of 100 μM was conducted with C. violaceum

CVO26. The QS model of C. violaceum is relatively simple and consists of an acyl-homoserine

lactone synthase (CviI) that produces the AI molecule N-hexanoyl homoserine lactone (HHL,

strain ATCC 31532), or N-3-hydroxydecanoyl homoserine lactone (strain ATCC 12472) [70]

which is recognized by the cytoplasmic receptor CviR (Fig 4). The reporter strain CVO26 is a

CviI::mini-Tn5 mutant of strain ATCC 31532 and thus recognizes HHL. CviR is a DNA-bind-

ing transcription factor that activates the expression of the genes encoding the production of

violacein, the characteristic purple alkaloid that gives name to the species [70].

Interestingly, an inverse correlation between the length of the alkyl chain and the inhibition

of the QS-regulated phenotype was found. Indeed, the inhibitions observed in violacein pro-

duction for long alkyl chains (10–18, 20) were clearly correlated to the biocidal effect exerted

by these compounds, with ‘therapeutic ratios’ (IC₅₀for growth inhibition/IC₅₀for violacein in-

hibition) around 1, in general (Table 2). Note that, as the length of the hydrocarbon chain be-

comes shorter, this ratio increased (8–9). Compounds 1 and 2 were inactive. However, four

compounds (3–5, 7) displayed an attractive profile as QS inhibitors, with IC₅₀values for bacte-

rial growth inhibition above the highest test concentration and consequently high therapeutic

indexes (Table 2). To evaluate these compounds in further detail, 40-mM stock solutions were

prepared in water to avoid the use of high concentrations of DMSO in the bioassays. As com-

pound 7 has a poor solubility in water a 200-mM stock in DMSO was prepared. Serial two-fold

dilutions (500–3.9 μM) were assayed. Consequently, the highest amount of DMSO used in the

assay was 0.25% (v/v). The recorded IC₅₀values for growth inhibitions were all above 500 μM,

whereas violacein inhibitions were in the range 52.9–142.2 μM (Table 2 and Fig 5). In terms of

growth kinetics, concentrations up to 250 μM of these compounds did not display any signifi-

cant effect (Fig 6 and S2 Fig). Growth inhibitory effects are only observed at 500 μM, particu-

larly for compound 7 (Fig 6D).

To check whether the compounds were able to thwart other QS-regulated phenotypes and

gain knowledge on their molecular targets, products 3–5 and 7 were evaluated in the more

complex QS model of Vibrio harveyi (Fig 4B). In V. harveyi, three AIs are synthesized by LuxM

(produces the HAI-1 signal, an acyl homoserine lactone), LuxS (AI-2 signal, a furanosyl borate

diester) and CqsA (CAI-1 signal, an α-amino ketone), respectively. These signals are believed

to mediate intra-species, inter-species and intra-genera communication, respectively [71].

These AIs are detected by three membrane sensors: LuxN, LuxPQ and CqsS, respectively. At

low AI concentrations, the receptors phosphorylate the σ⁵⁴-dependent response regulator

LuxO by means of the phosphotransferase LuxU, which activates the production of Quorum-

regulatory RNAs (Qrr sRNAs), four of which, together with the chaperone Hfq, target and de-

stabilize the mRNA that encodes the master regulator LuxR. On the contrary, at high AI

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

12 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

13 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 4. Quorum sensing circuits of Chromobacterium violaceum (A) and Vibrio harveyi (B). In C. violaceum ATCC 31532 (A), the synthase CviI

produces the AI molecule HHL that is recognized by the cytoplasmatic receptor CviR. When bound to HHL, CviR dimerizes and binds DNA, leading to the

expression of QS-regulated genes, including those involved in violacein production. In V. harveyi (B), three different AIs, synthesized by LuxM, LuxS and

CqsA are recognized by the transmembrane two-component receptors LuxN, LuxPQ and CqsS, respectively. At low AI concentrations, these receptors act

as kinases, phosphorylating LuxU and subsequently the σ⁵⁴-dependent response regulator LuxO. The phosphorylated LuxO activate the transcription of Qrr

sRNAs that together with the chaperone Hfq, destabilize the mluxR RNA. At high AI concentrations, the receptors switch to phosphatases and the expression

of the master regulator LuxR is allowed.

doi:10.1371/journal.pone.0123652.g004

concentrations, the QS receptors dephosphorylate LuxO via LuxU, allowing the expression of

more than 100 genes, those encoding luciferase amongst them [72].

Compounds 3–5 and 7 were evaluated in V. harveyi WT and three mutant strains: BB886

(luxPQ::Tn5Kan), unable to detect AI-2; BB170 (luxN::Tn5Kan), unable to detect HAI-1; and

BB721 (luxO::Tn5lacZ), a LuxO null mutant that produces maximal luminescence per cell con-

stitutively. In order to take into account the cumulative effects of the compounds, biolumines-

cence and growth were measured simultaneously every 15 min over 18 h (Fig 7). Accordingly,

IC₅₀values for bacterial luminescence and growth were obtained by integration of the areas

under the curves for each tested dose (Table 3). For clarity of the graphs, error bars has been

omitted in Fig 7 and only the mean curves are shown. A detailed version of this Fig. showing

the dispersion of the results is provided as S3 Fig. Several deductions can be inferred from

these data:

First, regarding the growth curves, some toxicity was observed at the highest dose (500 μM),

evidenced as a growth delay, for compounds 4 and 5 (Fig 7E–7H and 7I–7L). This effect was

Table 2. Half-maximal inhibitory concentrations (μM) for the tested compounds on the growth and

violacein production of C. violaceum CVO26.

Compound

IC₅₀(Growth inhibition)

IC₅₀(Violacein inhibition)

Ratio (GI/VI)

1

>100

>500

>100

>500

52.5

16.0

11.9

4.9

>100

92.3

142.2

136.8

>100

52.9

19.0

7.6

-

2

-

3

>5.4

>3.5

>3.7

-

4

5

6

7

>9.5

2.8

2.1

1.6

1.4

1.6

0.8

0.5

2.2

0.8

1.0

1.1

-

8

9

10

11

12

13

14

15

16

17

18

19

20

7.7

3.5

10.3

3.2

6.6

3.9

2.3

4.6

2.6

1.2

3.2

4.2

7.0

40.1

>100

12.8

35.2

>100

16.7

0.8

The index on the right column is calculated as the ratio between the IC₅₀value for growth inhibition (GI)

and that for violacein (VI).

doi:10.1371/journal.pone.0123652.t002

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

14 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 5. Dose-response curves for compounds 3–5 and 7 on C. violaceum CVO26 growth (A) and

violacein synthesis (B). Data represent the mean SD (N = 3).

doi:10.1371/journal.pone.0123652.g005

more prominent for compound 7 (Fig 7M–7P) but notably lower for compound 3, which in-

deed did not cause any substantial effect on bacterial growth (Fig 7A–7D). In addition, inhibi-

tions in bioluminescence were not strictly proportional to the inhibitions impaired on bacterial

growth, evidenced by the bioluminescence:growth ratios shown in Fig 8. Thus, toxicity does

not explain by itself the effects observed in the luminescent phenotype. Second, according to

the IC₅₀data for the different signaling pathways presented in Table 3, although the IC₅₀values

obtained for inhibitions in the AI-2-mediated QS were slightly lower than those obtained for

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

15 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 6. Growth curves of C. violaceum CVO26 in the presence of compounds 3 (A), 4 (B), 5 (C) and 7 (D). Serial two-fold dilutions of the compounds from

500 to 7.8 μM were tested. A detailed version of this Fig is provided (S2 Fig).

doi:10.1371/journal.pone.0123652.g006

that mediated by HAI-1, there seem to be no clear preference of the compounds for one or an-

other receptor, further confirmed by the luminescence inhibitions observed for the LuxO null

mutant. Since V. harveyi BB721 is constitutively bright, we deduce that any luminescence inhi-

bition in this strain, if observed, should be due to interaction of the test substance with ele-

ments downstream LuxO (although strictly, an inhibition caused by direct interaction of the

compounds with the luciferase enzyme cannot be discarded with these data; the possibility of

multiple targets cannot be discarded either). Third, it is particularly evident a non-monotonic

dose-response relationship for compound 5 (Fig 7I–7L and Fig 8) which is in marked contrast

with the monotonic trend of the other three phosphonium bromides. This non-monotonic be-

havior is more evident in the HAI^-and AI2^-mutants (Fig 7J and 7K), and less accused in the

WT and LuxO^-strains (Fig 7I and 7L), suggesting a preferential interaction with the LuxN and

LuxPQ receptors. Since these are membrane-bound proteins, this evidence agrees with the

mode of action of triphenylphosphonium compounds, which target and trespass the cell

membranes.

There are very few reports of QS inhibition by quaternary ‘-onium’ salts. For instance, qua-

ternary ammonium compounds such as calmidazolium has been proven to interfere with QS

in V. harveyi [73], whereas Peach and co-workers highlighted two miltefosine-related

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

16 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 7. Bioluminescence (solid lines) and growth curves (dotted lines) for compounds 3–5 and 7 in V. harveyi. Compound 3 (A-D); compound 4 (E-H);

compound 5 (I-L); compound 7 (M-P). Serial two-fold dilutions of the compounds from 500 to 7.8 μM were tested.

doi:10.1371/journal.pone.0123652.g007

compounds as hits in a high-throughput screening of biofilm inhibitors in V. choleare [74]. In

that case, however, the authors did not study if biofilm inhibition was caused by QS disruption

or through any other mechanism. To the best of our knowledge, this is the first report of QS

disruption by phosphonium compounds. In the present study, compounds 3–5 and 7 did not

cause a substantially different response in either the LuxN receptor nor in the LuxPQ receptor.

Evidence of QS disruption downstream LuxO further supports the existence of more than one

target in the QS signaling circuit. Interestingly, the effect of compound 5 followed a non-mono-

tonic trend. This kind of response was unexpected for a QS disruptor, but it is not uncommon

in pharmacology, in particular for endocrine disruptors and hormones [75]. Among the causes

attributed for this kind of response, one of the most common is nonselectivity [75]. Even

Table 3. IC₅₀values (μM) for luminescence and growth inhibitions caused by compounds 3–5 and 7 in

V. harveyi WT and reporter strains.

Compound

Luminescence

WT

Growth

BB886

BB170

BB721

3

296.7

>500

205.1

265.2

306.9

>500

179.4

251.0

225.8

>500

128.6

348.5

>500

276.4

>500

4

5*

7

*Non-monotonic response.

doi:10.1371/journal.pone.0123652.t003

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

17 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 8. Relative luminescence ratios (Light units/OD₆₀₀) for compounds 3–5 and 7 in V. harveyi. WT (A), BB886 (B), BB170 (C) and BB721 (D). Serial

two-fold dilutions of the compounds from 500 to 7.8 μM were tested.

doi:10.1371/journal.pone.0123652.g008

though evidence of QS disruption, unrelated to toxic effects, is deduced from the present find-

ings, the precise mechanisms describing how these compounds thwart bacterial cell-to-cell

communication remain to be elucidated. It is worth to recall that, since quaternary phosphoni-

um compounds have shown a promising profile as therapeutics and drug carriers [76–82], the

present findings immediately suggest a potential use as coadjuvants in antibiotherapy or other

antibacterial treatments (e.g. luminescent vibriosis in aquaculture [83]).

Tyrosinase inhibition

Tyrosinases (polyphenol oxidases) are ubiquitous enzymes whose function is the catalysis of

the hydroxilation of monophenols to o-monophenols, and the subsequent oxidation of the lat-

ter to o-quinones [84,85]. Tyrosinases play a key role in many biochemical processes, from the

biosynthesis of pigments such as melanin to sclerotisation of insect cuticules or the production

of biological adhesives such as those of barnacles and mussels [85,86]. Mussels become at-

tached to rocks in the intertidal zone through the production of adhesive threads termed byssi.

Byssogenesis occurs in three stages [87]: 1) the mussel explores a surface with a muscular

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

18 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

organ termed ‘foot’ and, when a suitable place is found, the mussel cleans the surface of unat-

tached matter and gets prepared for byssal secretion with contractional movements of the foot;

2) the foot is pressed tightly to the surface and a pull of adhesive proteins rich in DOPA resi-

dues is secreted, forming a byssal thread and a terminal plaque within few minutes; 3) the foot

is retracted and the process is repeated to form more threads (S1 Video). Tyrosinases are pres-

ent in the proteinaceous secretions of the foot glands and are involved in the curation (cross-

linking) of the polymeric adhesive through the oxidation of DOPA (post-transcriptionally con-

versed to tyrosine) to o-quinone [87,88].

Tyrosinase inhibition is thus a key target for the laboratory testing of antifouling molecules.

Although Mytilus sp. phenoloxidase differs to some extent from that of the widespread com-

mercial source (Agaricus bisporus), the latter was used for the in-vitro testing of the phosphoni-

um compounds since a) it is non-expensive and commercially available, thus making it suitable

for screening purposes, and b) it has been previously proposed for antifouling testing [89].

Under our experimental conditions, the kinetic parameters K_Mand v_maxobtained for the cata-

lytic conversion of L-Dopa to dopaquinone were 0.8–1.0 mM and 0.07–0.09 AU min^-1. As

Table 4 shows, only compounds with alkyl chains of C₁₂-C₁₉exhibited IC₅₀values below the

cut-off concentration of 100 μM. A steep shift in the activity is observed for lengths above C₁₄,

with a performance similar to that of the known tyrosinase inhibitor kojic acid. However, com-

pound 20 did not display relevant inhibitions at the tested concentrations. Thus, it is deduced

that the inhibitory activity is caused by the interaction of the alkyl chain with the enzyme,

which is reverted by the presence of a second cationic moiety.

Compound 16 was selected to further characterize the type of inhibition displayed by the

triphenylphosphonium salts. Lineweaver-Burk plots of the inhibited enzyme by a dose around

the IC₅₀(15 μM) and double (30 μM) in comparison to the uninhibited enzyme yielded a

group of straight lines that converged at their intersection with the x axis (Fig 9). This pattern

corresponds to a non-competitive inhibition, in which the inhibitor has the same affinity for

the free enzyme than for the enzyme-substrate complex. These results suggest that compound

16 inhibits tyrosinase activity by causing a conformational change derived from its binding to

an allosteric site.

Fluorescence of enzymes is very sensitive to conformational changes of the protein struc-

ture. The interaction of tyrosinase inhibitors can be evaluated by the analysis of the intrinsic ty-

rosinase fluorescence when tyrosinase and inhibitor (quencher) molecules are mixed in a

solution. In particular, we have evaluated the effect of the tyrosinase inhibitor 16. Quenching

of tyrosinase fluorescence is expected as a consequence of conformational alterations induced

Table 4. Tyrosinase IC₅₀values (μM) for the compounds tested in this study.

Compound

IC₅₀(μM)

1–11

>100

94.5

78.9

34.7

28.4

17.5

15.0

12.3

>100

16.5

12

13

14

15

16

17

18

19

20

Kojic acid

doi:10.1371/journal.pone.0123652.t004

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

19 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 9. Lineweaver-Burk plots for tyrosinase inhibition in the presence of compound 16. Data represent

the mean SD (N = 3).

doi:10.1371/journal.pone.0123652.g009

by the quencher [90,91]. Tryptophan residues are basically responsible of the tyrosinase fluo-

rescence when the compound is optically excited at around 280 nm. Fig 10 shows the fluores-

cence emission spectra of tyrosinase enzymes mixed with different quencher concentrations.

The quencher concentrations chosen in this study ranged from zero (native enzyme) to 30 μM.

This upper limit is well below the cmc threshold, which is estimated to be around 100 μM [56].

In this way, we assure that the reported tyrosinase fluorescence quenching is due to enzyme-in-

hibitor interaction and it is not related to any detergent-like side effect. In addition to this, a de-

natured enzyme sample was included in the study for comparison purposes. Quenching of the

fluorescence intensity is observed as the concentration of the inhibitor increases. In some cases,

a red-shift is expected in the maximum of the emission wavelength, which is associated to a

higher exposure of the tryptophan residues in the distorted protein structure to the polar

Fig 10. Fluorescence spectra of tyrosinase under 280 nm excitation at different concentrations of 16. The denatured and uninhibited enzyme were

included as controls. Inset: Stern-Volmer plot of the fluorescence quenching.

doi:10.1371/journal.pone.0123652.g010

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

20 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

environment of the solvent [92]. However, there is no significant red-shift in the emission spec-

tra, which indicates that the tryptophan residues are not very much exposed to the

solvent environment.

Fluorescence quenching can be described by the Stern-Volmer equation:

F₀

¼ 1 þ K_SV½Qꢀ

ð1Þ

F

where F₀and F are the ﬂuorescence intensities before and after the addition of the quencher, re-

spectively, [Q] is the concentration of the quencher, and K_SVis the Stern-Volmer quenching

constant, which indicates the sensitivity of the enzyme to the quencher. A linear dependence of

the ratio F₀/F on the quencher concentration can be observed in the inset of Fig 10. This is nor-

mally an indication of a single class of ﬂuorophore, tryptophan residues, which are equally acces-

sible to the quencher. The K_SVconstant of about 4490 M^-1has been obtained from the best

ﬁtting of the Stern-Volmer plot to a linear equation. Two possible quenching mechanisms pro-

vide linear Stern-Volmer plots. On the one hand, collisional quenching of the ﬂuorescence oc-

curs when the quencher diffuses to the ﬂuorophore during the lifetime of its excited state. The

excited molecule returns to its ground state without emission of radiation due to contact with

the quencher. This is a dynamic process, in which there is no permanent distortion of the pro-

tein. On the other hand, static quenching happens when a molecular complex is formed between

the ﬂuorophore and the quencher. In order to distinguish which of the two mechanisms is re-

sponsible of the tyrosinase ﬂuorescence quenching, lifetime measurements can be performed.

In the case of collisional quenching (dynamic quenching) the lifetime of the excited state of

the fluorophore decreases with the quencher concentration according to the following equa-

tion:

t₀

t

¼ 1 þ K_SV½Qꢀ

ð2Þ

Consequently,

F₀t₀

¼

ð3Þ

F

t

This means that a shortening of the lifetime equal to that of the ﬂuorescence intensity shoud

be expected.

However, if static quenching is occurring, the enzyme-quencher complexes are non-fluores-

cent. This means that the fluorescence detected comes from proteins which have not interacted

with the quencher and, consequently, their lifetime remains the same.

The decay of the PL has been measured under excitation at 280 nm and detection at the

maximum emission wavelength (Fig 11). A similar exponential decay curve has been measured

for the native enzyme and for the enzyme-inhibitor solutions up to the highest quencher con-

centration used (30 μM). This result indicates that static quenching is responsible of the fluo-

rescence quenching observed. When the data are fitted to an exponential decay curve the best

fit is obtained for a lifetime value of about 2.3 ns. In contrast, a significant drop of the fluores-

cence lifetime to 0.75 ns is measured for the denatured enzyme.

The fact that the lifetime of the enzyme-quencher solutions is the same than that of the

native enzyme indicates that the quenching of the fluorescence is due to a static process,

which occurs by the formation of a permanent enzyme-quencher complex which is non-fluo-

rescent. The short decay of the denatured protein is probably due to the unfolding of the en-

zyme, which exposes the tryptophan residues to the solvent environment introducing new

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

21 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 11. Decay of the fluorescence of tyrosinase tryptophan residues at different concentrations of

compound 16. IRF is the instrumental response function.

doi:10.1371/journal.pone.0123652.g011

non-radiative relaxation mechanisms, which are responsible of the shortening of the lifetime

and decrease of the intensity of the fluorescence.

Mussel foot retraction assay

At this point, two groups of triphenylphosphonium compounds can be distinguished: those in-

hibiting QS-regulated phenotypes, with a non-toxic mode of action (3–5, 7) and those with a

broad-spectrum biocidal and tyrosinase inhibitory activity (12–18). To investigate the behavior

of both kinds of compounds as mussel repellents, a representative compound of each class was

selected to be tested in M. galloprovincialis. Thus, compounds 3 and 16 were selected for fur-

ther evaluation. As described before, mussels explore substrata with their feet before byssal for-

mation. Consequently, substances causing a repellent response (evidenced by foot retraction)

are likely to deter mussels from settling.

Both 3 and 16 caused rather similar mussel foot retraction behaviors, with a clear dose-de-

pendent effect (Fig 12). The highest test concentration (200 μM) caused the strongest effect for

both tested compounds (94–95%). Mean increments between consecutive increasing tests on

the whole range of concentrations represented approximately two-fold increase for both com-

pounds with maxima of four-fold and three-fold increase from 12.5 to 25 μM and 6.25 to

12μM for both 16 and 3, respectively. The concentration of both compounds required to cause

foot repulsion in 50% of mussels was 47 and 34 μM for 3 and 16, respectively (Fig 12). Visual

evidence of these results in comparison to the positive (CuSO₄) and negative (FSW) control

can be observed in S2–S4 Videos.

Conclusions

In this study, the antifouling profile 20 triphenylphosphonium salts has been comprehensively

evaluated (Fig 13). The activity displayed by these compounds fall into four main categories:

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

22 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 12. Dose-response curves for compunds 3 and 16 in mussel (Mytilus galloprovincialis) foot

retraction assays.

doi:10.1371/journal.pone.0123652.g012

1. Broad-spectrum biocides against marine microorganisms, macroalgal spores and inverte-

brates. These compounds interacted strongly with the biological membranes due to their

high lipophilicity (> C₇) or positive charge (20).

2. Tyrosinase inhibition, caused by compounds with alkyl chains above C₁₁. This activity was

notably higher for compounds 16–18, with IC₅₀values similar to that of kojic acid, in the

low-micromolar range. The type of inhibition was non-competitive. Fluorescence spectros-

copy analyses confirmed that the inhibition was due to a static process, caused by the forma-

tion of a permanent enzyme-quencher complex.

3. QS inhibition. Four compounds in this series (3–5 and 7) exhibited a promising behavior as

QS disruptors in two bacterial models: C. violaceum and V. harveyi. These compounds were

non-toxic to bacteria at inhibitory concentrations and, in particular, compound 3 did not

exert any toxic effect up to 500 μM. The evaluation in V. harveyi mutants suggests non-spec-

ificity, likely with multiple molecular targets, including elements downstream LuxO.

4. Mussel foot repellents. Compounds 3 (a QS inhibitor) and 16 (a broad-spectrum biocide

and potent tyrosinase inhibitor) were tested in mussel foot retraction assays with Mytilus

galloprovincialis as test organism. Both compounds exhibited a similar behavior, with effec-

tive doses of 47 and 34 μM, respectively.

Recall that compounds with negatively-charged functionalities at the end of the alkyl chain

(6, 19) were inactive, probably as a consequence of a different arrangement in solution.

The findings reported in this study widen the scope of use of triphenylphosphonium com-

pounds as antifouling additives. However, it is worth to recall that the activities reported herein

have been demonstrated with molecules in solution and it remains to test whether they are able

to exhibit the same properties in functionalized material surfaces for antifouling protection.

Since their biocidal properties has been exploited over decades to confer antimicrobial proper-

ties on different kinds of materials, it is likely that the new activities presented herein for this

family of compounds will be retained as well.

Whereas the use of biocides is environmentally safe only when they are immobilized (e.g.

functionalized surfaces) or they are easily biodegraded and released in a controlled way (e.g.

self-polishing coatings), the use of non-toxic inhibitors of key processes for the biological

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

23 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

Fig 13. Summary of this study. The different colors highlight the main bioactivities and the structure-activity relationships of the tested

triphenylphosphonium salts.

doi:10.1371/journal.pone.0123652.g013

colonization of material surfaces is more versatile and desirable. In this regard, it is particularly

prospective the discovery of the ability of certain triphenylphosphonium compounds to disrupt

bacterial cell-to-cell communication without exerting significant toxicities. We are aware that

these properties could be also exploited in other fields not necessarily related to marine anti-

fouling protection.

The promising profile of these non-toxic compounds was reinforced by the similar response

that compound 3 caused in mussel foot retraction tests in comparison to compound 16, which

would implicate such non-toxic product selection for fouling impact analysis as a powerful

strategy in terms of environmental impact versus broad-spectrum product efficiency balance.

Those innocuous products for the biological environment that exert powerful repelling actions

on the mussel feet (at least comparable to other components with high biocidal action) might

be desirable for dealing with fouling impact. Nonetheless, it is necessary to keep in mind the

gap between the antifouling impact and the mussel foot-repelling actions we are checking,

which are more comparable to a negative chemotactic response rather than an inhibitory effect

on mussel attachment.

Overall, this study re-focus on the antifouling properties of alkyl triphenylphosphonium

compounds from a different perspective, including new biological targets such as tyrosinase in-

hibition or bacterial intercellular communication. As far as we are aware, this study reports for

the first time the QS inhibitory properties of phosphonium compounds, which deserve a par-

ticular in-depth evaluation.

Supporting Information

S1 Fig. Arrangement of M. galloprovincialis individuals employed in foot retraction assays.

The posterior adductor muscle is cutted to open both valves (A), and the foot-retracting assay

is conducted by dipping the test solutions onto the animal’s feet (B).

(TIF)

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

24 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

S2 Fig. Growth curves of C. violaceum CVO26 in the presence of compounds 3 (A), 4 (B), 5

(C) and 7 (D). Serial two-fold dilutions of the compounds from 500 to 7.8 μM were tested.

Data represent the mean SD (N = 3).

(TIF)

S3 Fig. Bioluminescence (solid lines) and growth curves (dotted lines) for compounds 3–5

and 7 in V. harveyi. Compound 3 (A-D); compound 4 (E-H); compound 5 (I-L); compound 7

(M-P). Serial two-fold dilutions of the compounds from 500 to 7.8 μM were tested. Data repre-

sent the mean SD (N = 3).

(TIF)

S1 File. General methodological information and spectroscopic data of the compounds syn-

thesized for this study.

(DOCX)

S1 Video. Time-lapse sequence of byssal thread formation. The sequence, that covers 4–5 h

of continuous recording, shows clearly how the mussel extends its foot to explore the surface

and produces a byssal thread.

(MP4)

S2 Video. Mussel foot-retraction response to FSW (negative control).

(MP4)

S3 Video. Mussel foot-retraction response to 1000 ppm CuSO₄(positive control).

(MP4)

S4 Video. Mussel foot-retraction response to 200 μM of compound 3.

(MP4)

Acknowledgments

A.J.M-R. holds a 2+2 fellowship from PLOCAN. The authors are indebted to Dr. Tom Defoirdt

(Department of Animal Production, University of Ghent) for his guidance and valuable advice

in the experimental set-up with Vibrio harveyi. Dr. José M. Siverio (Department of Biochemis-

try, Microbiology, Cellular Biology and Genetics, University of La Laguna) is acknowledged for

his helpful comments regarding the enzymatic assays. Dr. Gercende Courtois (University of

Las Palmas de Gran Canaria) and Dr. Ángel Trigos (University of Veracruz) are acknowledged

for providing the diatom and and fungal strains referenced in this study. The authors are

thankful to Mrs. Elsa Silva Caride for her labor in the mussel foot repulsion tests, and to Dr.

Cirilo Pérez for his assistance in the preparation of the synthetic compounds.

Author Contributions

Conceived and designed the experiments: AJM-R JJF. Performed the experiments: AJM-R. An-

alyzed the data: AJM-R JJF. Contributed reagents/materials/analysis tools: VSM MN JJF.

Wrote the paper: AJM-R JJF. Conducted the mussel foot repulsion tests: JMFB. Analyzed the

fluorescence quenching data: FL. Proposed the use of G. oxysperma, collected the specimens

and induced spore release: MS. Drafted the manuscript: AJM-R.

References

1. Yebra D, Kiil S, Dam-Johansen K. Antifouling technology—past, present and future steps towards effi-

cient and environmentally friendly antifouling coatings. Prog Org coatings. 2004; 50: 75–104.

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

25 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

2. Hertiani T, Edrada-Ebel R, Ortlepp S, van Soest RWM, de Voogd NJ, Wray V, et al. From anti-fouling to

biofilm inhibition: new cytotoxic secondary metabolites from two Indonesian Agelas sponges. Bioorg

Med Chem. 2010; 18: 1297–1311. doi: 10.1016/j.bmc.2009.12.028 PMID: 20061160

3. Qian PY, Xu SY. Antifouling Activity of Marine Natural Products. In: Fattorusso E, Gerwick WH, Taglia-

latela-Scafati O, editors. Handbook of Marine Natural Products. Dordrecht, The Netherlands: Springer;

2012. pp. 749–821.

4. Budd M, Beaumont A. High mortality of the larvae of the common mussel at low concentrations of tribu-

tyltin. Mar Pollut Bull. 1984; 15: 402–405.

5. Maguire RJ. Environmental aspects of tributyltin. Appl Organomet Chem. 1987; 1: 475–498.

6. Valkirs AO, Seligman PF, Davidson BM. Sublethal growth effects and mortality to marine bivalves from

long-term exposure to tributyltin. Chemosphere. 1987; 16: 201–220.

7. Jellali R, Campistron I, Pasetto P, Laguerre A, Gohier F, Hellio C, et al. Antifouling activity of novel poly-

isoprene-based coatings made from photocurable natural rubber derived oligomers. Prog Org Coat-

ings. 2013; 76: 1203–1214.

8. Lejars M, Margaillan A, Bressy C. Fouling release coatings: a nontoxic alternative to biocidal antifouling

coatings. Chem Rev. 2012; 112: 4347–4390. doi: 10.1021/cr200350v PMID: 22578131

9. Majumdar P, Lee E, Patel N, Ward K, Stafslien SJ, Daniels J, et al. Combinatorial materials research

applied to the development of new surface coatings IX: an investigation of novel antifouling/fouling-re-

lease coatings containing quaternary ammonium salt groups. Biofouling. 2008; 24: 185–200. doi: 10.

1080/08927010801894660 PMID: 18368587

10. Long CJ, Finlay JA, Callow ME, Callow JA, Brennan AB. Engineered antifouling microtopographies:

mapping preferential and inhibitory microenvironments for zoospore attachment. Biofouling. 2010; 26:

941–952. doi: 10.1080/08927014.2010.531390 PMID: 21038153

11. Decker JT, Kirschner CM, Long CJ, Finlay JA, Callow ME, Callow JA, et al. Engineered antifouling

microtopographies: an energetic model that predicts cell attachment. Langmuir. 2013; 29:13023–

13030. doi: 10.1021/la402952u PMID: 24044383

12. Olsen SM, Kristensen JB, Laursen BS, Pedersen LT, Dam-Johansen K, Kiil S. Antifouling effect of hy-

drogen peroxide release from enzymatic marine coatings: Exposure testing under equatorial and Medi-

terranean conditions. Prog Org Coatings. 2010; 68: 248–257.

13. Regina VR, Søhoel H, Lokanathan AR, Bischoff C, Kingshott P, Revsbech NP, et al. Entrapment of

subtilisin in ceramic sol-gel coating for antifouling applications. ACS Appl Mater Interfaces. 2012; 4:

5915–5921. doi: 10.1021/am301554m PMID: 23020255

14. Burgess JG, Boyd KG, Armstrong E, Jiang Z, Yan L, Berggren M, et al. The development of a marine

natural product-based antifouling paint. Biofouling. 2003; 19 Suppl: 197–205. PMID: 14618721

15. Fusetani N. Antifouling marine natural products. Nat Prod Rep. 2011; 28: 400–410. doi: 10.1039/

c0np00034e PMID: 21082120

16. Wahl M, Goecke F, Labes A, Dobretsov S, Weinberger F. The second skin: ecological role of epibiotic

biofilms on marine organisms. Front Microbiol. 2012; 3: 292. doi: 10.3389/fmicb.2012.00292 PMID:

22936927

17. Dobretsov S, Teplitski M, Bayer M, Gunasekera S, Proksch P, Paul VJ. Inhibition of marine biofouling

by bacterial quorum sensing inhibitors. Biofouling. 2011; 27: 893–905. doi: 10.1080/08927014.2011.

609616 PMID: 21882898

18. Moore LE, Ledder RG, Gilbert P, McBain AJ. In vitro study of the effect of cationic biocides on bacterial

population dynamics and susceptibility. Appl Environ Microbiol. 2008. 74: 4825–4834. doi: 10.1128/

AEM.00573-08 PMID: 18515475

19. Kanazawa A, Ikeda T, Endo T. Synthesis and antimicrobial activity of dimethyl- and trimethyl-substitut-

ed phosphonium salts with alkyl chains of various lengths. Antimicrob Agents Chemother. 1994; 38:

945–952. PMID: 8067774

20. Kanazawa A, Ikeda T, Endo T. Novel polycationic biocides: synthesis and antibacterial activity of poly-

meric phosphonium salts. J Polym Sci Part A Polym Chem. 1993; 31: 335–343.

21. Hemp ST, Zhang M, Allen MH, Cheng S, Moore RB, Long TE. Comparing ammonium and phosphoni-

um polymerized ionic liquids: thermal analysis, conductivity, and morphology. Macromol Chem Phys.

2013; 214: 2099–2107.

22. Chang H-I, Yang M-S, Liang M. The synthesis, characterization and antibacterial activity of quaternized

poly(2,6-dimethyl-1,4-phenylene oxide)s modified with ammonium and phosphonium salts. React

Funct Polym. 2010; 70: 944–950.

23. Kumar V, Malhotra SV. Study on the potential anti-cancer activity of phosphonium and ammonium-

based ionic liquids. Bioorg Med Chem Lett. 2009; 19: 4643–4646. doi: 10.1016/j.bmcl.2009.06.086

PMID: 19615902

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

26 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

24. García-Argüelles S, Serrano MC, Gutiérrez MC, Ferrer ML, Yuste L, Rojo F, et al. Deep eutectic sol-

vent-assisted synthesis of biodegradable polyesters with antibacterial properties. Langmuir. 2013; 29:

9525–9534. doi: 10.1021/la401353r PMID: 23808373

25. Kenawy E- R, Abdel-Hay FI, El-Magd AA, Mahmoud Y. Biologically active polymers: VII. Synthesis and

antimicrobial activity of some crosslinked copolymers with quaternary ammonium and phosphonium

groups. React Funct Polym. 2006; 66: 419–429.

26. Anthierens T, Billiet L, Devlieghere F, Du Prez F. Poly(butylene adipate) functionalized with quaternary

phosphonium groups as potential antimicrobial packaging material. Innov Food Sci Emerg Technol.

2012; 15: 81–85.

27. Hatakeyama ES, Ju H, Gabriel CJ, Lohr JL, Bara JE, Noble RD, et al. New protein-resistant coatings

for water filtration membranes based on quaternary ammonium and phosphonium polymers. J Memb

Sci. 2009; 330: 104–116.

28. Wang K, Zeng Y, He L, Yao J, Suresh AK, Bellare J, et al. Evaluation of quaternary phosphonium-

based polymer membranes for desalination application. Desalination. 2012; 292: 119–123.

29. Li C, Liu Y, Zeng Q-Y, Ao N-J. Preparation and antimicrobial activity of quaternary phosphonium modi-

fied epoxidized natural rubber. Mater Lett. 2013; 93: 145–148.

30. Wu T, Xie A-G, Tan S-Z, Cai X. Antimicrobial effects of quaternary phosphonium salt intercalated clay

minerals on Escherichia coli and Staphylococci aureus. Colloids Surf B Biointerfaces. 2011; 86: 232–

236. doi: 10.1016/j.colsurfb.2011.04.009 PMID: 21536418

31. Cai X, Zhang J, Ouyang Y, Ma D, Tan S, Peng Y. Bacteria-adsorbed palygorskite stabilizes the quater-

nary phosphonium salt with specific-targeting capability, long-term antibacterial activity, and lower cyto-

toxicity. Langmuir. 2013; 29: 5279–5285. doi: 10.1021/la400824f PMID: 23547786

32. Smith RAJ, Porteous CM, Gane AM, Murphy MP. Delivery of bioactive molecules to mitochondria in

vivo. Proc Natl Acad Sci U S A. 2003; 100: 5407–5412. PMID: 12697897

33. Biswas S, Dodwadkar NS, Piroyan A, Torchilin VP. Surface conjugation of triphenylphosphonium to tar-

get poly(amidoamine) dendrimers to mitochondria. Biomaterials. 2012; 33: 4773–4782. doi: 10.1016/j.

biomaterials.2012.03.032 PMID: 22469294

34. Bansal R, Tripathi SK, Gupta KC, Kumar P. Lipophilic and cationic triphenylphosphonium grafted linear

polyethylenimine polymers for efficient gene delivery to mammalian cells. J Mater Chem. 2012; 22:

25427–25436.

35. Brumaghim J, Michels M, Raymond K. Hydrophobic chemistry in aqueous solution: stabilization and

stereoselective encapsulation of phosphonium guests in a supramolecular host. European J Org

Chem. 2004; 22: 4552–4559.

36. Uraguchi D, Ueki Y, Ooi T. Controlled assembly of chiral tetraaminophosphonium aryloxide-arylhydrox-

ide(s) in solution. Angew Chem Int Ed Engl. 2011; 50: 3681–3683. doi: 10.1002/anie.201007752

PMID: 21413109

37. Kanazawa A, Ikeda T, Abe J. Polar thin films produced by phosphonium liquid crystals: two-dimension-

al self-assembled ionic layers with spontaneous polarization. Angew Chem Int Ed Engl. 2000; 39:

612–615. PMID: 10671275

38. Yang Z, Jin X, Guaciaro M, Molino BF. Asymmetric synthesis and absolute configuration of streptophe-

nazine G. J Org Chem. 2012; 77: 3191–3196. doi: 10.1021/jo202642a PMID: 22432723

39. Clinical and Laboratory Standards Institute. Methods for dilution antimicrobial susceptibility tests for

bacteria that grow aerobically. Approved Standard. 9th ed., volume 32. Wayne, Philadelphia: Clinical

and Laboratory Standards Institute; 2012.

40. Martín-Rodríguez AJ, Reyes F, Martín J, Pérez-Yépez J, León-Barrios M, Couttolenc A, et al. Inhibition

of bacterial quorum sensing by extracts from aquatic fungi: first report from marine endophytes. Mar.

Drugs. 2014; 12: 5503–5526. doi: 10.3390/md12115503 PMID: 25415350

41. Arendrup MC, Cuenca-Estrella M, Lass-Flörl C, and the Subcommittee on Antifungal Susceptibility

Testing (AFST) of the ESCMID European Committee for Antimicrobial Susceptibility Testing

(EUCAST). Method for the determination of broth dilution minimum inhibitory concentrations of antifun-

gal agents for conidia forming moulds. EUCAST Definitive Document EDef 9.2. 2014.

42. Rambali B. Susceptibility testing of pathogenic fungi with itraconazole: a process analysis of test vari-

ables. J Antimicrob Chemother. 2001; 48: 163–177. PMID: 11481285

43. Wellburn AR. The spectral determination of chlorophylls a and b, as well as total carotenoids, using var-

ious solvents with spectrophotometers of different resolution. J Plant Physiol. 1994; 144:307–313.

44. Warren CR. Rapid Measurement of chlorophylls with a microplate reader. J Plant Nutr. 2008; 31:

1321–1332.

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

27 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

45. Pellizzari F, Oliveira MC, Medeiros AS, Yokoya NS, Oliveira EC. Morphology, ontogeny, and phyloge-

netic position of Gayralia brasiliensis sp. nov. (Ulotrichales, Chlorophyta) from the southern coast of Bra-

zil. Bot Mar. 2013; 56: 197–205.

46. Callow ME, Callow JA, Pickett-Heaps JD, Wetherbee R. Primary adhesion of Enteromorpha (Chloro-

phyta, Ulvales) propagules: quantitative settlement studies and video microscopy. J Phycol. 1997; 33:

938–947.

47. Fletcher RL. A bioassay technique using the marine fouling green alga Enteromorpha. Int Biodeterior.

1989; 25: 407–422.

48. Chambers LD, Hellio C, Stokes KR, Dennington SP, Goodes LR, Wood RJK, et al. Investigation of

Chondrus crispus as a potential source of new antifouling agents. Int Biodeterior Biodegradation. 2011.

65: 939–946.

49. Vandeputte OM, Kiendrebeogo M, Rajaonson S, Diallo B, Mol A, El Jaziri M, et al. Identification of cate-

chin as one of the flavonoids from Combretum albiflorum bark extract that reduces the production of

quorum-sensing-controlled virulence factors in Pseudomonas aeruginosa PAO1. Appl Environ Micro-

biol. 2010; 76: 243–253. doi: 10.1128/AEM.01059-09 PMID: 19854927

50. Martinelli D, Grossmann G, Séquin U, Brandl H, Bachofen R. Effects of natural and chemically synthe-

sized furanones on quorum sensing in Chromobacterium violaceum. BMC Microbiol. 2004; 4: 25.

PMID: 15233843

51. Greenberg EP, Hastings JW, Ulitzur S. Induction of luciferase synthesis in Beneckea harveyi by other

marine bacteria. Arch Microbiol. 1979; 120: 87–91.

52. Souza PM, Elias ST, Simeoni LA, de Paula JE, Gomes SM, Guerra ENS, et al. Plants from Brazilian

Cerrado with potent tyrosinase inhibitory activity. PLoS One 2012; 7: e48589. doi: 10.1371/journal.

pone.0048589 PMID: 23173036

53. Hayashi Y, Miki W. A newly developed bioassay system for antifouling substances using the blue mus-

sel, Mytilus edulis galloprovincialis. J Mar Biotechnol. 1996; 43: 127–130.

54. Anderson EB. Synthesis and Non-Covalent Interactions of Novel Phosphonium-Containing Polymers.

Ph.D. Thesis. Virginia Polytechnic Institute and State University. 2010. Avalibale: http://scholar.lib.vt.

edu/theses/available/etd-09012010-115556/.

55. Siedenbiedel F, Tiller JC. Antimicrobial polymers in solution and on surfaces: overview and functional

principles. Polymers. 2012; 4: 46–71.

56. Prasad M, Moulik SP, Bengal W, Macdonald A, Palepu R. Self-aggregation of Alkyl (C10-, C12-, C14-,

and C16-) triphenyl phosphonium bromides and their 1:1 molar mixtures in aqueous medium: a thermo-

dynamic study J Phys Chem B. 2004; 108: 355–362.

57. Gilbert P, Moore LE. Cationic antiseptics: diversity of action under a common epithet. J Appl Microbiol.

2005; 99: 703–715. PMID: 16162221

58. Qian P-Y, Lau SCK, Dahms H-U, Dobretsov S, Harder T. Marine biofilms as mediators of colonization

by marine macroorganisms: implications for antifouling and aquaculture. Mar Biotechnol. 2007; 9:

399–410. PMID: 17497196

59. Rutherford ST, Bassler BL. Bacterial quorum sensing: its role in virulence and possibilities for its con-

trol. Cold Spring Harb Perspect Med. 2012; 2: a012427 doi: 10.1101/cshperspect.a012427 PMID:

23125205

60. Busscher HJ, van der Mei HC. How do bacteria know they are on a surface and regulate their response

to an adhering state? PLoS Pathog. 2012; 8: e1002440. doi: 10.1371/journal.ppat.1002440 PMID:

22291589

61. Shrout JD, Tolker-Nielsen T, Givskov M, Parsek MR. The contribution of cell-cell signaling and motility

to bacterial biofilm formation. MRS Bull. 2011; 36: 367–373. PMID: 22053126

62. Atkinson S, Cámara M, Williams P. N-Acylhomoserine lactones, quorum sensing, and biofilm develop-

ment in Gram-negative bacteria. In: Kjelleberg S, Givskov M, editors. The biofilm mode of life: Mecha-

nisms and adaptations. Wymondham, UK: Horizon Bioscience; 2007. pp. 95–122.

63. Parsek MR, Greenberg EP. Sociomicrobiology: the connections between quorum sensing and biofilms.

Trends Microbiol. 2005; 13: 27–33. PMID: 15639629

64. Huang Y-L, Dobretsov S, Ki J-S, Yang L-H, Qian P-Y. Presence of acyl-homoserine lactone in subtidal

biofilm and the implication in larval behavioral response in the polychaete Hydroides elegans. Microb

Ecol. 2007; 54: 384–392. PMID: 17394040

65. Huang Y-L, Dobretsov S, Xiong H, Qian P-Y. Effect of biofilm formation by Pseudoalteromonas spon-

giae on induction of larval settlement of the polychaete Hydroides elegans. Appl Environ Microbiol.

2007; 73: 6284–6288. PMID: 17704279

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

28 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

66. Twigg MS, Tait K, Williams P, Atkinson S, Cámara M. Interference with the germination and growth of

Ulva zoospores by quorum-sensing molecules from Ulva-associated epiphytic bacteria. Environ Micro-

biol. 2013; 16: 445–453. doi: 10.1111/1462-2920.12203 PMID: 23879807

67. Qian P-Y, Chen L, Xu Y. Mini-review: Molecular mechanisms of antifouling compounds. Biofouling.

2013; 29: 381–400. doi: 10.1080/08927014.2013.776546 PMID: 23574197

68. Dobretsov S, Dahms H-U, Yili H, Wahl M, Qian P-Y. The effect of quorum-sensing blockers on the for-

mation of marine microbial communities and larval attachment. FEMS Microbiol Ecol. 2007; 60: 177–

188. PMID: 17371321

69. Bridier A, Briandet R, Thomas V, Dubois-Brissonnet F. Resistance of bacterial biofilms to disinfectants:

a review. Biofouling. 2011; 27: 1017–1032. doi: 10.1080/08927014.2011.626899 PMID: 22011093

70. Stauff DL, Bassler BL. Quorum sensing in Chromobacterium violaceum: DNA recognition and gene

regulation by the CviR receptor. J Bacteriol. 2011; 193: 3871–3878. doi: 10.1128/JB.05125-11 PMID:

21622734

71. Pompeani AJ, Irgon JJ, Berger MF, Bulyk ML, Wingreen NS, Bassler BL. The Vibrio harveyi master

quorum-sensing regulator, LuxR, a TetR-type protein is both an activator and a repressor: DNA recog-

nition and binding specificity at target promoters. Mol Microbiol. 2008; 70: 76–88. doi: 10.1111/j.1365-

2958.2008.06389.x PMID: 18681939

72. Jung K. Microbiology: Tuning communication fidelity. Nat Chem Biol. 2011; 7: 502–503. doi: 10.1038/

nchembio.618 PMID: 21769096

73. Soulère L, Sabbah M, Fontaine F, Queneau Y, Doutheau A. LuxR-dependent quorum sensing: Com-

puter aided discovery of new inhibitors structurally unrelated to N-acylhomoserine lactones. Bioorg

Med Chem Lett. 2010; 20: 4355–4358. doi: 10.1016/j.bmcl.2010.06.081 PMID: 20615699

74. Peach KC, Bray WM, Shikuma NJ, Gassner NC, Lokey RS, Yildiz FH, et al. An image-based 384-well

high-throughput screening method for the discovery of biofilm inhibitors in Vibrio cholerae Mol BioSyst.

2011; 7: 1176–1184. doi: 10.1039/c0mb00276c PMID: 21246108

75. Vandenberg LN, Colborn T, Hayes TB, Heindel JJ, Jacobs DR, Lee DH, et al. Hormones and endo-

crine-disrupting chemicals: low-dose effects and nonmonotonic dose responses. Endocr Rev. 2012;

33: 378–455. doi: 10.1210/er.2011-1050 PMID: 22419778

76. Solomon MA, Shah AA, D’Souza GGM. In vitro assessment of the utility of stearyl triphenyl phosphoni-

um modified liposomes in overcoming the resistance of ovarian carcinoma Ovcar-3 cells to paclitaxel.

Mitochondrion. 2013; 13: 464–472. doi: 10.1016/j.mito.2012.10.013 PMID: 23123917

77. Belikova NA, Jiang J, Stoyanovsky DA, Glumac A, Bayir H, Greenberger JS, et al. Mitochondria-tar-

geted (2-hydroxyamino-vinyl)-triphenyl-phosphonium releases NO* and protects mouse embryonic

cells against irradiation-induced apoptosis. FEBS Lett. 2009; 583: 1945–1950. doi: 10.1016/j.febslet.

2009.04.050 PMID: 19427865

78. Smith RAJ, Hartley RC, Cochemé HM, Murphy MP. Delivery of bioactive molecules to mitochondria in

vivo. Trends Pharmacol Sci. 2012; 33: 341–352. doi: 10.1016/j.tips.2012.03.010 PMID: 22521106

79. Manetta A, Gamboa G, Nasseri A, Podnos YD, Emma D, Dorion G, et al. Novel phosphonium salts dis-

play in vitro and in vivo cytotoxic activity against human ovarian cancer cell lines. Gynecol Oncol. 1996;

60: 203–212. PMID: 8631539

80. Bachowska B, Kazmierczak-Baranska J, Cieslak M, Nawrot B, Szczęsna D, Skalik J, et al. High cyto-

toxic activity of phosphonium salts and their complementary selectivity towards HeLa and K562 cancer

cells: Identification of tri-n-butyl-n-hexadecylphosphonium bromide as a highly potent anti-HeLa phos-

phonium salt. ChemistryOpen. 2012; 1: 33–38. doi: 10.1002/open.201100003 PMID: 24551490

81. Bansal R, Tripathi SK, Gupta KC, Kumar P. Lipophilic and cationic triphenylphosphonium grafted linear

polyethylenimine polymers for efficient gene delivery to mammalian cells. J Mater Chem. 2012; 22:

25427.

82. Millard M, Pathania D, Shabaik Y, Taheri L, Deng J, Neamati N. Preclinical evaluation of novel triphe-

nylphosphonium salts with broad-spectrum activity. PLoS ONE. 2010; 5: e13131. doi: 10.1371/journal.

pone.0013131 PMID: 20957228

83. Defoirdt T, Boon N, Sorgeloos P, Verstraete W, Bossier P. Alternatives to antibiotics to control bacterial

infections: luminescent vibriosis in aquaculture as an example. Trends Biotechnol. 2007; 25: 472–479.

PMID: 17719667

84. Sánchez-Amat A, Solano F, Lucas-Elío P. Finding new enzymes from bacterial physiology: a success-

ful approach illustrated by the detection of novel oxidases in Marinomonas mediterranea. Mar Drugs.

2010; 8: 519–541. doi: 10.3390/md8030519 PMID: 20411113

85. García‐Borrón J, Solano F. Molecular anatomy of tyrosinase and its related proteins: Beyond the gisti-

dine‐bound metal catalytic center. Pigment cell Res. 2002; 15: 162–173. PMID: 12028580

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

29 / 30

Antifouling Profile of Alkyl Triphenylphosphonium Salts

86. Thalmann C, Lötzbeyer T. Enzymatic cross-linking of proteins with tyrosinase. Eur Food Res Technol.

2002; 214: 276–281.

87. Silverman H, Roberto F. Understanding marine mussel adhesion. Mar Biotechnol. 2007; 9: 661–681.

PMID: 17990038

88. Bandara N, Zeng H, Wu J. Marine mussel adhesion: biochemistry, mechanisms, and biomimetics. J

Adhes Sci. 2013; 27: 1–24.

89. Dahms H-U, Hellio C. Laboratory bioassays for screening marine antifouling compounds. In Hellio C,

Yebra D, editors. Advances in Marine Antifouling Coatings and Technologies. Boca Ratón, FL: CRC

Press; 2009. pp. 275–307.

90. Lai J-Q, Li Z, Li Y-H, Yang Z. Specific ion effects of ionic liquids on enzyme activity and stability. Green

Chem. 2011; 13: 1860–1868.

91. Kim D, Park J, Kim J, Han C, Yoon J, Kim N, et al. Flavonoids as mushroom tyrosinase inhibitors: a fluo-

rescence quenching study. J Agric Food Chem. 2006; 54: 935–941. PMID: 16448205

92. Lakowicz JR. Principles of Fluorescence Spectroscopy. 3rd ed. New York: Springer; 2007.

PLOS ONE | DOI:10.1371/journal.pone.0123652 April 21, 2015

30 / 30

Article Doi

From broad-spectrum biocides to quorum sensing disruptors and mussel repellents: Antifouling profile of alkyl triphenylphosphonium salts

DOI: 10.1371/journal.pone.0123652

Source and publish data:

Authors:

Article abstract of DOI:10.1371/journal.pone.0123652

Full text of DOI:10.1371/journal.pone.0123652

Products guided by the article

R&D Labs maybe for 22884-33-9

Relevant to this article

Hot Product