Pro-Antimicrobial Networks via Degradable Acetals (PANDAs) UsingThiol−Ene PhotopolymerizationDahlia N. Amato,†,‡ Douglas V. Amato,†,‡ Olga V. Mavrodi,§ William B. Martin,‡ Sarah N. Swilley,‡

Keith H. Parsons,‡ Dmitri V. Mavrodi,*,§ and Derek L. Patton*,‡

‡School of Polymers and High Performance Materials and §Department of Biological Sciences, The University of SouthernMississippi, Hattiesburg, Mississippi 39406, United States

*S Supporting Information

ABSTRACT: We describe the synthesis of pro-antimicrobialnetworks via degradable acetals (PANDAs) as a new paradigmfor sequestration and triggered release of volatile, bioactivealdehydes. PANDAs derived from diallyl p-chlorobenzalde-hyde acetal degrade and release p-chlorobenzaldehyde as anantibacterial and antifungal agent under mild conditions (pH7.4/high humidity). We show that PANDAs enable facileaccess to materials with tunable release profiles, potentantimicrobial activity without triggering antimicrobial resist-ance, and minimal cytotoxicity.

With 700000 annual deaths globally, antimicrobialresistance is an escalating crisis that threatens the

sustainability of public health and agricultural ecosystems.1 Asthe effectiveness of antibiotics has precipitously declined, agrowing interest in alternative antibiotic scaffolds hasproliferated. In this direction, aromatic terpene aldehydes,major phytochemical constituents of plant derived essential oils(EOs), are known to exhibit potent and broad spectrumantibacterial and antifungal activities in both liquid and gaseousstates.2,3 However, practical application of terpenes remains achallenge stemming from their poor water solubility, volatility,and chemical instability. Many strategies have been reported tosequester or encapsulate terpenes within films or colloidalsystems;4,5 however, these strategies often suffer deficienciessuch as low loading, poor encapsulation efficiencies, necessity oforganic processing solvents, and uncontrolled burst releaseprofiles.6

“Polyactives”, polymeric pro-drugs that undergo degradationto release active therapeutic agents, address the deficiencies insequestering EO constituents by providing high loading,chemical stability, and tailored release kinetics.7 Recently,linear polyactives of phytochemicals and other syntheticanalogs have been used for the extended delivery ofantimicrobials,8,9 antioxidants,10,11 anti-inflammatories,12 hor-mones,13 and anticancer therapeutics.14,15 While successfullydemonstrated, tuning the degradation of linear polyactivesremains a challenge as degradation rates are determined bymanipulating coupled parameters such as crystallinity, polymermolecular weight, functionality, and hydropathy.16 Alterna-tively, cross-linked polyactives (i.e., thermosets) offer uniqueopportunities to tailor release profiles by simple manipulation

of cross-link density, monomer concentration, or monomermolecular weight.17 Recently, Matras and Chatterjee developedbiodegradable cross-linked polyactive esters that challenge theperformance of traditional linear polymers by offering greatercontrol over degradation rates, mechanical properties, andrelease kinetics.18 Although these cross-linked polyestersshowed high biocompatibility and are capable of releasinganticancer, anti-inflammatory, or antimicrobial com-pounds,19−21 the lengthy polymerization times (>1 day) atelevated temperatures (>130 °C) precludes polymerization inthe presence of cells, in vivo, or in 3D printing applications.Furthermore, polyesters with sustained degradation typicallyexhibit localized accumulation of acidic byproducts leading toinflammatory response.22 Polyactive acetals are promisingalternatives as acetals readily hydrolyze under mild aqueousconditions into pH neutral byproducts (alcohols andaldehydes).23

Inspired by polyacetal hydrogels,24,25 the work hereindescribes the synthesis of pro-antimicrobial networks viadegradable acetals (PANDAs) derived from p-chlorobenzalde-hyde (pCB) using thiol−ene photopolymerization, an approachdesigned to address many of the aforementioned challengeswith sequestration of terpene aldehydes. Thiol−ene photo-polymerization offers rapid cure kinetics at room temperature,low oxygen inhibition, homogeneous network formation, andhigh monomer conversion.26 Thiol−ene, as a step-additionpolymerization, ensures that nearly every cross-link junction

Received: January 5, 2017Accepted: February 1, 2017Published: February 6, 2017

Letter

pubs.acs.org/macroletters

© 2017 American Chemical Society 171 DOI: 10.1021/acsmacrolett.7b00009ACS Macro Lett. 2017, 6, 171−175

contains a degradable, pCB-derived acetal. Therefore, PANDAsare molecularly designed to undergo complete degradationupon hydrolysis resulting in the release of pCB as an activeantimicrobial and antifungal agent, and the generation ofinactive low molecular weight degradation byproducts (Scheme1).

Initially, we synthesized an acyclic diallyl p-chlorobenzalde-hyde acetal (pCBA) derived from pCB. For proof-of-concept,pCB was selected from several possible benzaldehydederivatives with known antimicrobial activity.27,28 Benzylicaldehydes derivatives are known to exert antimicrobial activitythrough a variety of mechanisms, specifically, inhibition of H+-ATPase-mediated proton pumping activity, Schiff base/thiazolidine formation with proteins/peptides, and membranedisruption.29,30 We synthesized pCBA in acceptable yields viaacid-catalyzed condensation of pCB with allyl alcohol (1H/13CNMR, Figures S1 and S2). The antimicrobial activity of pCBAand pCB were evaluated via a zone of inhibition (ZOI) assaywith clinically isolated strains of Escherichia coli ATCC 43895[serotype 95 O157:H7] (E. coli ATCC 43895), Staphylococcusaureus RN6390 (S. aureus RN6390), Burkholderia cenocepaciaK56−2 (B. cenocepacia K56−2), and Pseudomonas aeruginosaPAO1 (P. aeruginosa PAO1) (Figure S3). At the highest testedconcentration (10 mg mL−1), pCBA exhibited no apparentantimicrobial activity. In contrast, pCB showed significantinhibition at 0.5 mg mL−1, results that indicate the acyclic acetalfunctions as a pro-antimicrobial compound with antimicrobialefficacy only upon hydrolysis to the aldehyde.UV-curable PANDA resins were then formulated with a

photoinitiator, pentaerythritol tetramercaptopropionate(PETMP), and varied concentrations of pCBA relative tonondegradable 1,3,5-triallyl-1,3,5-triazine-2,4,6-trione (TTT)with 1:1 alkene/thiol mole stoichiometry (Scheme 1).Following exploratory experiments, we focused primarily onPANDAs synthesized from a 90:10 pCBA/TTT monomer feedas proof-of-principle. This composition results in PANDAmaterials comprised of ∼45 wt % pCBA. PANDAs werephotopolymerized in the absence of solvent under a medium-pressure UV light (200 mW cm−2). Polymerization kinetics,investigated using real-time FTIR, were rapid with near

quantitative conversions (>95%) observed for both thiol andalkene functional groups within 5 min (Figure 1a). The

resulting PANDAs are transparent, low modulus thermosetswith a glass transition temperature (Tg) at 0.5 °C (Figure 1b).Although not presented in detail here, the concentration ofTTT can be used to facilely tune the physical properties of thenetwork (i.e., Tg, flexibility, degradation, etc.).PANDA degradation and subsequent release of pCB were

investigated at pH 7.4 and 6.0 by submerging sample disks (25mm3) in phosphate buffer solution (PBS) containing octanol topartition the aldehyde. Aliquots from the octanol phase wereanalyzed via UV−vis spectroscopy to determine the concen-tration of pCB released over time. Figure 1c shows the pCBrelease profile at pH 7.4 and 6.0. At physiological pH, a 20 h lagtime was observed before pCB was slowly released from thepolymer network, with 14% (1.07 mg mL−1) and 53% (4.19 mgmL−1) pCB release observed at 24 and 48 h, respectively.Similar trends in degradation and pCB release were observed atpH 6. After 120 h, 72% and 80% pCB release was observed forpH 7.4 and 6.0, respectively. Given that the rate of acetalhydrolysis should be proportional to the H3O

+ concentration,31

the rather small increase in rate as a function of pH indicatesthat incorporation of the acetal into a hydrophobic polymernetwork drastically decreases the observed rate of hydrolysis.Kim et al.25 attributed similar trends in degradation ofhydrophobic acetal networks to slow diffusion of aqueousbuffer into the network.The photo series in Figure 1d shows the sample disks

immersed in PBS buffer solution (pH 7.4) as a function of time.

Scheme 1. Synthesis of PANDAs and Major DegradationByproducts

Figure 1. Cure kinetics, thermomechanical properties, release kinetics,and degradation behavior of PANDAs. (a) Conversion kinetics for90% pCBA resins cured at 200 mW cm−2 UV light. (b) Representativethermomechanical plot of the 90% pCBA PANDA. (c) Correlationbetween pCB released and incubation time at pH 6.0 and 7.4. (d)Time-lapse macroscopic images of degradation of 90% pCBA diskssubmerged in PBS (pH 7.4) and (e) placed within a 90% humiditychamber under N2 at 25 °C (Image contrast enhanced for visibility).Error bars indicate the SD (n = 5).

ACS Macro Letters Letter

DOI: 10.1021/acsmacrolett.7b00009ACS Macro Lett. 2017, 6, 171−175

172

Shortly after immersion, the samples change from transparentto opaque white, but retain their shape with recedingdimensions over time. The observed release and degradationbehavior are consistent with a surface erosion process (typicalof polyacetals), and may be expected given that these materialsare relatively hydrophobic (static water contact angle ≈90°).32,33 The degradation behavior of these materials becomesexceedingly more interesting when exposed to water vapor(rather than bulk immersion). Figure 1e shows a photo series ofPANDA disks subjected to a gentle flow of 90% humiditywithin a scintillation vial (using N2 as a carrier for water vaporto avoid acidification from CO2). Within 2 h, the PANDA diskhydrolyzed to a point where viscous flow of the sample wasobserved. At 4 h, the degradation byproducts are shown flowingdown the supporting substrate. The rapid progression from aninfinite polymer network to viscous flow of low molecularweight degradation products is a consequence of the thiol−enestep polyaddition mechanism, a process that ensures mostcross-link junctions contain a degradable linkage and precludesthe formation of high MW linear degradation products found inchain growth networks.34,35 Notably, these observations pointto a rapid bulk degradation process and suggest a vastdifference in water vapor permeability (under high humidity)and liquid water permeability (under immersion). While theseobservations appear to be unique for degradable networksbased on current literature, the difference in permeability ofwater vapor and bulk water is a phenomenon commonlyobserved in cross-linked poly(dimethylsiloxane).36,37 Further-more, qualitative observations of PANDAs exposed to HClvapor led to rapid bulk degradation, while exposure toammonium hydroxide vapor led to retarded degradation(Figure S4). Experiments are underway to quantify thedegradation behavior under vapor conditions.The synthetic design of PANDAs is motivated by the

potential use of these materials in antimicrobial/antifungalapplications. The antimicrobial activity of PANDAs was initiallyevaluated via a ZOI assay with clinically isolated strains of E. coliATCC 43895 [serotype O157:H7], S. aureus RN6390, B.cenocepacia K56−2, and P. aeruginosa PAO1. As shown inFigure 2a, PANDAs containing 0% pCBA showed no ZOI, asexpected, and serve as a control for the assay. PANDAscontaining >60% pCBA exhibited ZOIs, with larger ZOIsobserved for all bacteria with increasing concentration of pCBA(relative to TTT) in the polymer network. The antimicrobialefficacy of the 90% pCBA PANDA material was furtherinvestigated using a terminal dilution assay to quantify killkinetics. Two 25 mm3 90% pCBA PANDA disks (3.6 mg mL−1

pCBA released within 24 h) exhibited more than a 5 logreduction in bacteria count in <12 h against E. coli, S. aureus,and other pathogens (Figure 2b), which translates into killefficiencies of >99.999%. In interest of developing strategiesand materials to mitigate the development of antimicrobialresistance, we evaluated the potential emergence of bacterialresistance of P. aeruginosa against pCBA-based PANDAs usinga serial passage mutagenesis assay. For comparison, the sameassay was performed using tetracycline. Figure 2c shows theinability of P. aeruginosa to develop resistance toward pCBreleased from PANDA disks after 20 serial passages as indicatedby the absence of an increase in the MIC (25 mm3 disk ≈ 1.3mg mL−1 pCB released within 24 h). We note that theconcentration of pCB released from the disk after 24 h iscomparable to the small molecule MIC at 1.25 mg mL−1. Incontrast, the MIC value for tetracycline increased after only two

passages, with a 10-fold increase in MIC after 20 passages. Theresults suggest that P. aeruginosa has less propensity to developresistance against the pCB released from pCBA-basedPANDAs, which highlights the potential of PANDAs for useas part of a broader strategy to slow or mitigate thedevelopment of antimicrobial resistance.To further probe the broad-spectrum activity of pCBA

PANDAs, we performed zone of inhibition experiments againsttwo opportunistic pathogenic fungi [Candida albicans (C.albicans) and Trihoderma harzianum (T. harzianum)].38,39

Placement of a 25 mm3 90% pCBA PANDA disk resulted incomplete inhibition of both fungi whereas the same size diskscontaining 0% pCBA exhibited no measurable antimicrobialactivity (Figure 2d). Borrowing from the known inhibitoryactivity of volatile EOs, we exploited the inherent volatility ofpCB to inhibit C. albicans, a pathogenic fungus responsible forup to 65% of all candidiasis cases in humans, via a split plateinhibition volatility assay.39 As shown in Figure 2e, 25 mm3

90% pCBA PANDA disks were placed on one side of a 9 cmsplit Petri dish with C. albicans plated on agar on the oppositeside. The dish was sealed and incubated for 30 days. In theabsence of PANDA disks (control), zero inhibition wasobserved. Partial inhibition was observed with one disk, whilethe growth of C. albicans was completely inhibited by two disksfor up to 30 days. These experiments exploit the humidity-

Figure 2. Antimicrobial activity and cytocompatibility assays. Errorbars indicate the SD (n = 5). (a) Correlation between zone ofinhibition and % pCBA in disk after 24 h incubation. (b) Kill kineticsof four bacteria in the presence of 90% pCBA PANDA via terminaldilution assay. (c) Bacterial resistance study of the 90% pCBA PANDAand tetracycline against P. aeruginosa. (d) Zone of inhibition diffusionassay of 0% and 90% pCBA PANDAs against C. albicans and T.harzianum. (e) Split plate diffusion assay of 90% pCBA PANDA disksagainst C. albicans after 30 days. (f) Cell viability assay for control(DMSO), pCB, and degraded 90% pCBA PANDA.

ACS Macro Letters Letter

DOI: 10.1021/acsmacrolett.7b00009ACS Macro Lett. 2017, 6, 171−175

173

triggered degradation and volatility of pCB for noncontactcontrol of fungus. Finally, KB cells (a type of HeLa cells),derived from a glandular cancer of the cervix were used as amodel cell line for cytotoxicity.40 Incubation of the KB cellswith a degraded 90% pCBA disk (e.g., media containingPANDA degradation products after 7 days at pH 7.4, equivalentto 2.5 mg mL−1 pCB) and small molecule pCB at the sameconcentration showed no significant toxicity (>90% viability) toKB cells (Figure 2f).In summary, we have demonstrated the proof-of-principle for

pro-antimicrobial networks via degradable acetals. PANDAsundergo degradation under mild conditions to release pCB, asmall molecule active that serves as a potent antimicrobial agentagainst a broad spectrum of microbes. The humidity triggereddegradation and volatile release of pCB was exploited to inhibitthe growth of C. albicans without the need for direct contact.Importantly, the broad spectrum antimicrobial efficacy wasachieved with minimal cytotoxicity toward KB HeLa cells.Solvent-free, room-temperature, photopolymerization providesa rapid, one-pot approach to cross-linked polyactives withexcellent potential for advanced processing techniques,including 3D printing. The PANDA approach presented hereis readily applicable to other bioactive aldehydes pointing to thepotential use of these materials within pharmaceutical,biomedical, and agricultural industries.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acsmacro-lett.7b00009.

Experimental details for the synthesis and character-ization of pCBA, PANDAs, antifungal/antibacterialassays, and cytotoxicity toward KB cells (PDF).

■ AUTHOR INFORMATIONCorresponding Authors*E-mail: dmitri.mavrodi@usm.edu.*E-mail: derek.patton@usm.edu.

ORCIDDahlia N. Amato: 0000-0001-9831-358XDouglas V. Amato: 0000-0003-3343-2945Derek L. Patton: 0000-0002-8738-4750Author Contributions†These authors contributed equally (D.N.A. and D.V.A.).

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported in part by the National ScienceFoundation (OIA-1430364). D.N.A. and D.V.A. acknowledgestraineeship support from the NSF NRT program “Interface”(DGE-1449999) through USM. We thank Dwaine Braasch forassistance with GC-MS.

■ REFERENCES(1) O’Neil, J. Antimicrobial Resistance: Tackling a Crisis for the Healthand Wealth of Nations; Wellcome and HM Government: London,U.K., 2016.(2) Burt, S. Int. J. Food Microbiol. 2004, 94 (3), 223−253.

(3) Nielsen, P. V.; Rios, R. Int. J. Food Microbiol. 2000, 60 (2−3),219−229.(4) El Asbahani, A.; Miladi, K.; Badri, W.; Sala, M.; Addi, E.;Casabianca, H.; El Mousadik, A.; Hartmann, D.; Jilale, A.; Renaud, F.;Elaissari, A. Int. J. Pharm. 2015, 483 (1−2), 220−243.(5) Sanchez-Gonzalez, L.; Vargas, M.; Gonzalez-Martínez, C.;Chiralt, A.; Chafer, M. Food Eng. Rev. 2011, 3 (1), 1−16.(6) Majeed, H.; Bian, Y.-Y.; Ali, B.; Jamil, A.; Majeed, U.; Khan, Q.;Iqbal, K.; Shoemaker, C.; Fang, Z. RSC Adv. 2015, 5 (72), 58449−58463.(7) Stebbins, N. D.; Faig, J. J.; Yu, W.; Guliyev, R.; Uhrich, K. E.Biomater. Sci. 2015, 3 (8), 1171−1187.(8) Carbone-Howell, A. L.; Stebbins, N. D.; Uhrich, K. E.Biomacromolecules 2014, 15 (5), 1889−1895.(9) Bien-Aime, S.; Yu, W.; Uhrich, K. E. Macromol. Biosci. 2016, 16(7), 978−983.(10) Ouimet, M. A.; Griffin, J.; Carbone-Howell, A. L.; Wu, W.-H.;Stebbins, N. D.; Di, R.; Uhrich, K. E. Biomacromolecules 2013, 14 (3),854−861.(11) Ko, E.; Jeong, D.; Kim, J.; Park, S.; Khang, G.; Lee, D.Biomaterials 2014, 35 (12), 3895−3902.(12) Kwon, J.; Kim, J.; Park, S.; Khang, G.; Kang, P. M.; Lee, D.Biomacromolecules 2013, 14 (5), 1618−1626.(13) Guo, S.; Nakagawa, Y.; Barhoumi, A.; Wang, W.; Zhan, C.;Tong, R.; Santamaria, C.; Kohane, D. S. J. Am. Chem. Soc. 2016, 138(19), 6127−6130.(14) Lv, L.; Guo, Y.; Shen, Y.; Liu, J.; Zhang, W.; Zhou, D.; Guo, S.Adv. Healthcare Mater. 2015, 4 (10), 1496−1501.(15) Kim, B.; Lee, E.; Kim, Y.; Park, S.; Khang, G.; Lee, D. Adv. Funct.Mater. 2013, 23 (40), 5091−5097.(16) Burkoth, A.; Burdick, J.; Anseth, K. J. Biomed. Mater. Res. 2000,51 (3), 352−359.(17) Rydholm, A. E.; Reddy, S. K.; Anseth, K. S.; Bowman, C. N.Biomacromolecules 2006, 7 (10), 2827−2836.(18) Chandorkar, Y.; Bhagat, R. K.; Madras, G.; Basu, B.Biomacromolecules 2014, 15 (3), 863−875.(19) Dasgupta, Q.; Madras, G.; Chatterjee, K. ACS Biomater. Sci. Eng.2016, DOI: 10.1021/acsbiomaterials.6b00680.(20) Dasgupta, Q.; Chatterjee, K. Mol. Pharmaceutics 2015, 12 (9),3479−3489.(21) Dasgupta, Q.; Madras, G.; Chatterjee, K. Acta Biomater. 2016,30, 168−176.(22) Lu, L.; Peter, S. J.; D. Lyman, M.; Lai, H.-L.; Leite, S. M.;Tamada, J. A.; Uyama, S.; Vacanti, J. P.; Robert, L.; Mikos, A. G.Biomaterials 2000, 21 (18), 1837−1845.(23) Binauld, S.; Stenzel, M. H. Chem. Commun. 2013, 49 (21),2082−2102.(24) Murthy, N.; Thng, Y. X.; Schuck, S.; Xu, M. C.; Frechet, J. M. J.J. Am. Chem. Soc. 2002, 124 (42), 12398−12399.(25) Kim, S.; Linker, O.; Garth, K.; Carter, K. R. Polym. Degrad. Stab.2015, 121, 303−310.(26) Hoyle, C. E.; Bowman, C. N. Angew. Chem., Int. Ed. 2010, 49(9), 1540−1573.(27) Friedman, M.; Henika, P.; Mandrell, R. J. Food Prot. 2003, 66(10), 1811−1821.(28) Pelttari, E.; Karhumaki, E.; Langshaw, J.; Perakyla, H.; Elo, H. Z.Naturforsch. C 2007, 62 (7−8), 487−497.(29) Shreaz, S.; Bhatia, R.; Khan, N.; Imran Ahmad, S.; Muralidhar,S.; Basir, S.; Manzoor, N.; Khan, L. Microb. Pathog. 2011, 51 (4), 277−284.(30) Xie, R.; Tu, M.; Elder, T. Energy Fuels 2016, 30 (4), 3078−3084.(31) Fife, T. H.; Jao, L. K. J. Org. Chem. 1965, 30 (5), 1492−1495.(32) Ulery, B.; Nair, L.; Laurencin, C. J. Polym. Sci., Part B: Polym.Phys. 2011, 49 (12), 832−864.(33) Heller, J. Poly(ortho esters). In Biopolymers I; Langer, R. S.,Peppas, N. A., Eds.; Springer: Berlin, Heidelberg, 1993; pp 41−92.(34) Metters, A.; Hubbell, J. Biomacromolecules 2005, 6 (1), 290−301.

ACS Macro Letters Letter

DOI: 10.1021/acsmacrolett.7b00009ACS Macro Lett. 2017, 6, 171−175

174

(35) Rydholm, A. E.; Bowman, C. N.; Anseth, K. S. Biomaterials2005, 26, 4495.(36) Randall, G. C.; Doyle, P. S. Proc. Natl. Acad. Sci. U. S. A. 2005,102 (31), 10813−10818.(37) Robb, W. L. Ann. N. Y. Acad. Sci. 1968, 146 (1), 119−137.(38) Kantarciog lu, A.; Celkan, T.; Yucel, A.; Mikami, Y.; Kurugoglu,S.; Mitani, H.; Altas, K. Med. Mycol. 2009, 47 (2), 207−215.(39) Pfaller, M.; Diekema, D.; Gibbs, D.; Newell, V.; Ellis, D.; Tullio,V.; Rodloff, A.; Fu, W.; Ling, T. J. Clin. Microbiol. 2010, 48 (4), 1366−1377.(40) Masters, J. R. W. Nat. Rev. Mol. Cell Biol. 2000, 1 (3), 233−236.

ACS Macro Letters Letter

DOI: 10.1021/acsmacrolett.7b00009ACS Macro Lett. 2017, 6, 171−175

175