Selective Sensing and Imaging of Penicillium italicum Spores and

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

Selective Sensing and Imaging of Penicillium italicum Spores and Hyphae using Carbohydrate-Lectin Interactions Idris Yazgan, Jing Zhang, Victor Murithi Kariuki, Ayfer Akgul, Lauren Cronmiller, Ali Akgul, Francis Osonga, Abbey McMahon, Yang Gao, Gaddi Eshun, Seokheun Choi, and Omowunmi A. Sadik ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.7b00934 • Publication Date (Web): 20 Feb 2018 Downloaded from http://pubs.acs.org on February 23, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Sensors is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Selective Sensing and Imaging of Penicillium italicum Spores and Hyphae using Carbohydrate-Lectin Interactions Idris Yazgana,‡, Jing Zhanga ,Victor Kariukia, Ayfer Akgulb, Lauren E Cronmillera, Ali Akgulc, Francis Osongaa, Abbey McMahona, Yang Gaod, Gaddi Eshuna, Seokheun Choid*, Omowunmi A Sadika* a

Department of Chemistry, Center for Research in Advanced Sensing Technologies &

Environmental Sustainability (CREATES), State University of New York at Binghamton, Binghamton, NY 13902-6000. b

Department of Clinical Sciences, College of Veterinary Medicine, Mississippi State University,

Starkville, P. O. Box 6100, MS 39762-6100 c

Department of Basic Sciences, College of Veterinary Medicine, Mississippi State University,

Box 9820, Starkville, MS 39762-9601 d

Department of Electrical and Computer Engineering, Center for Research in Advanced Sensing

Technologies & Environmental Sustainability (CREATES), State University of New York at Binghamton, 13902-6000

*Corresponding Author:[email protected] ‡

Current address-

Department of Biology, Faculty of Art and Science, Kastamonu University, Kastamonu-Turkiye

ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract The blue-green mold Penicillium italicum is among the most problematic post-harvest plant infections limiting the integrity of citrus and many other crops during storage and transportation, but there is no sensor for its on-site or field detection. We hereby, for the first time, report the development of novel biomolecular sensor for assessing the presence of Penicillium italicum spores and hyphae using carbohydrate-lectin recognitions. Two approaches were developed: (i) lateral tests using stand-alone poly (amic) acid (PAA) membranes and glass surfaces and (ii) quantitative tests on 96-well polystyrene plates and paper electrodes. In both cases, the surfaces were functionalized with novel derivatized sugar based ligands while staining was performed with gold nanoparticles. Both approaches provided strong signals for 104 spores/mL of P.italicum isolated from experimentally infected lemons as the lowest-reliable concentration. The 96-well plate-based gave the most sensitive detection with a 4 x 102 spores/mL limit of detection, a linear dynamic range between 2.9 x103 - 6.02 x104 spores/mL (R2 0.9939) and standard deviation of less than 5% for five replicate measurements. The selectivity of the ligands was tested against Trichaptum biforme, Glomerulla cingulata (Colletotrichum gloeosporioides) and Aspergillus nidulans fungi species. The highest selectivity was obtained using the sugarbased gold-nanoparticles towards both the spores and hyphae of P.italicum. The advanced specificity was provided by the utilized sugar ligands employed in the synthesis of gold nanoparticles and was independent from size and shapes of the AuNPs. Accuracy of the sensor response showed dramatic dependence on the sample preparation. In the case of 5-10 min centrifugation at 600 rpm, the spores can be isolated free from hyphae and conidiophore, for which spiked recovery was up to 95% (std ±4). In contrast, for gravity-based precipitation of hyphae, the spiked recovery was 88% (std 11).

Keywords: Field applications, Penicillium italicum, spores, hyphae, sugar ligands, gold nanoparticles, lectins

φ ACS Paragon Plus Environment

Page 2 of 21

Page 3 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Penicillium italicum is among the most problematic molds causing the loss of approximately 25% citrus production worldwide and its infection is commonly seen during storage and transportation[1,2]. P. italicum, widely seen as blue mold, causes the highest economical postharvest problems for citrus fruits. It is an absolute wound-pathogen, which only infects wounded citrus plants. The disease can be spread by fruit-to-fruit contact. P. italicum can specifically grow at or below 10 ºC, so its presence and infection is not prevented with cold-storage. This is what makes them infectious even under common protected storage conditions[3].

Numerous studies have been performed to eliminate and prevent the growth of P. italicum in citrus species[1,2] but there has been no studies reported on the detection of P.italicum species other than phylogenetic approaches[4]. In contrast, automated vision systems are applied to monitor infected citrus plants during storage and transportation[3], which are unable to detect species specific monitoring of the presence of P.italicum. Moreover, conventional culture methods are time consuming while other approaches such as ELISA, PCR, confocal and fluorescence microscopy and SPR are not easily accessible and may sometimes be laborintensive and costly[5]. On-site detection of fungal pathogens may be used to overcome the difficulties faced with the conventional techniques for species such as Aspergillus spp. and Cryptococcus spp.[6]. Currently, there is no study reporting the detection of P.italicum for onsite or field applications.

Lectins are specific proteins that selectively bind to free and/or bound carbohydrate residues[7]. Fungal lectins, as with most lectins, show high specificity towards their targets. For example, a lectin from Rhizopus stolonifer binds to GlcNAc-L-fucose 100-times stronger than individual GlcNAc residue[8]. In general, galactose and lactose derivatives have been reported to show strong affinity towards fungal lectins[9]. Among the characterized lectins, there has been no lectin reported to be specific to P. italicum[7]. While no lectin-carbohydrate based-study has been reported in detection of fungi [10], the utilization of the sugar ligands to target lectins on bacteria in biosensor applications have been successfully shown by few studies[11–15]. Functionalized nanoparticles [16,17](i.e. antibiotic functionalization) have been shown as selective and sensitive agents to detect bacteria including Mycobacterium avium[18] and Salmonella[19] and fungi [18].

χ ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In this study, mannose, lactose and galactose derivatives and gold nanoparticles synthesized with sugar-ligands were used to selectively capture and stain P. italicum spores and mycelium. 104 spores/mL of P.italicum from experimentally infected lemon was successfully detected by the sensors. This level is 100-times less than its infectious dose initiating necrosis on citrus. The specificity of the ligands was tested against Trichaptum biforme, Aspergillus nidulans and Glomerulla cingulate, for which relatively low recognition was obtained. To the best of our knowledge, this is the first study reporting the selective and sensitive detection of P.italicum spores and hyphae with high throughput platforms.

EXPERIMENTAL SECTION Materials. Unless otherwise stated, all the chemicals utilized in the study were purchased from SigmaAldrich (MO, the USA). 18.2 MΩ pure water was obtained via Barnstead D50280 nanopurewater systems. Potato-dextrose agar (PDA) was wet and freshly prepared according to the manufacturer’s recipe. Phosphate buffer saline (PBS) buffers at different pH values (pH 6.0 to 8.0) were prepared by dissolving 137 mM NaCl and 10 mM KH2PO4 in pure-water. In the case of acetate buffer, 50 mM acetate buffer was prepared from NaCOOCH3. Adjustments of the pH were performed with 10 mM NaOH and/or HCl.

Synthesis of Sugar Derivatives with Corresponding Gold Nanoparticle Conjugates

The detailed synthesis of all sugar ligands and sugar mediated gold-nanoparticles (AuNPs) will be reported elsewhere. Sugars were modified via reductive amination method as reported earlier[13]. Briefly, the ligands and sugar moieties were mixed in 50:50 (Acetic acid:water, v/v) for 1 h, followed by the introduction of dimethylamino borane to the media for 8 h (Figure 1). The product was finally isolated from the media using Flash-chromatography.

ψ ACS Paragon Plus Environment

Page 4 of 21

Page 5 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Figure 1: Synthesis of derivatized sugar ligands. R2; H, COOH, H2NC6H4NH2, O2SC6H4NH2; R1/R3: OH, H; R4: H, C6H11O5

Synthesis of Poly (amic) Acid Membranes Solutions of viscous Poly (amic) acid (PAA) polymers were prepared from 4,4’-oxydianiline and pyromellitic dianhydride. The solutions were further used to prepare stand-alone PAA membranes, via controlled evaporation mediation method as detailed elsewhere[20].

Propagation of P. italicum P.italicum is classified as Biosafety Level 1 (BSL 1) organisms in accordance with the American Type Culture Collection[21]. PDA was utilized to propagate P.italicum spores, mycelia and hyphae. 104 spores/mL of P.italicum inoculum was utilized for propagation at 24 °C for 3-5 day incubation (Figure S1).

Preparation of Fungal Solutions for Biodetection Fungal inoculums grown on PDA agar were collected with 5 mL of 0.2 µm filtered nanopure water into sterile centrifuge-tube, followed by vortexing at 1200 rpm for 5-10 min. The time requirement of vortexing depends on the type of fungus. For example, P.italicum required 10 min while Glomerulla cingulate needed only 5 min. Vortexing was applied to detach the spores from each other. Then, the samples were left for 20 min at room temperature to allow the precipitation of hyphae and/or medium remnants. The number of spores was counted with hemocytometer, and dilutions were made in buffers as detailed under each detection method.

Design of Sensing Platforms for P.italicum Detection

ω ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Four different surfaces were tested for the biomolecular development. These include borosilicate glass, gold-coated paper, PAA membranes and polystyrene plates, whose principles are illustrated in Figure 2. A variety of ligands were developed and tested to target P.italicum. Dglucosamine, L-cysteine, 3-aminoquinoline and 4/5-aminosalicylic acid were utilized without further modifications. D-galactose, D-mannose and D-lactose were modified with aminosalicylic acids, diamines, sulfonyldiamines and anilines. Only the best modified sugars were employed throughout the study. The synthesized sugar ligands were used as reducing and stabilizing agents for the synthesis of gold nanoparticles. These sugar based gold nanoparticles were labeled as sbAuNP1, sb-NP2, sb-NP3 and sb-NP4. The differences in these labels were based on their surface chemistry; size and shape of the nanoparticles were not different. The surface chemistry of the nanoparticles was unique based on the types of sugars employed but the size and shape was the same. All the nanoparticles were spherical with 6-10 nm size. The surface chemistries were 5(N-galactosyl)-salicylic acid, 5-(N-mannosyl)-salicylic acid, 5-(N-lactosyl)-salicylic acid and (N,N’-digalactosyl) ethylenediamine for sb-AuNP1, sb-NP2, sb-NP3 and sb-NP4, respectively.

Paper-based electrodes 100 nm gold layers were e-beamed onto glossy photography-paper [ATC Orion 8-E, E-beam evaporator, AJA International]. Free thiol group containing modified sugar, galactose-4thioaniline, dissolved in ethanol to introduce onto the gold surface. Then, the coated electrode was placed into the vial containing 104 P. italicum spores/mL [pH 7.6 PBS buffer] for 6h with continuous stirring (120 rpm) at 35 °C. Unbounded and weakly bounded P.italicum spores were removed with rinsing thrice with pH 7.6 PBS buffer. 10 mM K3(CN)6 in 1 M NaNO3 solution was used as electroactive species to monitor binding on gold electrode. Silver and titanium wires were used for reference and auxiliary electrodes, respectively. Measurements were carried out using Autolab potentiostat PGstat302n equipped with Novo 1.11 software [FL, the USA].

Silica Glass Slides The detection of P.italicum using activated glass surfaces was based on a “yes/no” response. Glass slides were cleaned with 50:50 (v/v), HCl:MeOH, followed by free-amino or mercapto group containing silanization were performed as reported elsewhere[22,23]. The amino-group functionalized glass surface was activated with 10 % glutaraldehyde (GA) in pure-water for 1 h.

ϊ ACS Paragon Plus Environment

Page 6 of 21

Page 7 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

The mercapto-group activated glass-surface was activated with N-γ-maleimidobutyryloxysuccinimide ester (GMBS) chemistry. This was dissolved in pH 7.4 PBS buffer, followed by activation with primary amino groups for 30 min. The activated molecules (N-lactosyl-p-aniline, NLPDA) were then introduced to 3-mercaptopropyl trimethoxysilane activated glass surfaces, where active end groups of GMBS bound to free thiol group [18]. Then, from 103 spores/mL to 105 spores/mL of P.italicum, suspended in pH 7.4 PBS buffer or pH 5.5 acetate buffer were added to the activated glass surfaces for 4-6 h incubation under 35 °C and 60-rpm continuousstirring conditions. Under non-stirring conditions, incubation must be carried out between 12-16 h. Any unbound P.italicum spores were subsequently rinsed away through sonication in purewater for 3 min. Finally, gold-nanoparticles were added to each sample for 8h incubation. At the end of each step, the glass and the membranes were rinsed 5-times with pure-water to eliminate unbound molecules and the fungal spores.

Poly (amic) acid Membranes Similar to the activated glass supports, the detection of P.italicum on the PAA membranes were only tested for “yes/no” responses under the tested conditions for the same reason. Transparent PAA membranes were prepared using controlled-vaporization method[20,24]. In order to enhance contrast for possible P.italicum binding related discoloration, coloring dyes (i.e. rhodamine 6G) were added to the PAA membranes, or the developed gold nanoparticles were utilized to stain the captured spores/hyphae on none-treated PAA membranes. In all cases, 10% GA was utilized to activate the PAA surface for 2h. This was followed by the introduction of the ligand, NLPDA. Any remaining active GA residues were de-activated with the addition of glycine. Spore concentrations ranging from 103 spores/mL to 105 spores/mL were added to the membrane surfaces for between 4 h to 8 h incubation. Limited numbers of hyphae and conidiophore were accompanied to spores for non-intended cases. In the case of non-stirring conditions, incubation needed to be extended up to 12h.

Polystyrene 96-wells Plates 96-well polystyrene plates provide high-throughput screening since this allows the testing of different concentrations and multiple assays simultaneously. Hence 96-well plate was extensively utilized in the study for testing different concentrations of P.italicum for different

ϋ ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ligands and selectivity tests. Microtiter® 96-well plate was first treated with 40 mM KOH for 2 hours to increase presence of carboxyl groups, followed by activated with 50 mM EDC/NHS for 2 h. N-lactosyl-p-aniline (NLPDA) at 5 mg/mL and 1 mg/mL concentrations was incubated for 2 h. Penicillium italicum, dissolved in 50 mM pH 5.5 acetate-buffer at 6x104 to 3x103 spores/mL were incubated for 6h at 35 °C under continuous stirring. At the final step, a special sugar functionalized AuNPs were introduced for 4h under continuous shaking at 35 ºC.

Selectivity of the developed sugar ligands and sugar ligand gold nanoparticles Selectivity is one of the most essential parameters in biomolecular sensing. Aspergillus nidulans, Trichaptum biforme and Glomerulla cingulate were utilized in the selectivity test. Relative response of the sensor was calculated by accepting the sensor response for P.italicum as 100 obtained while assigning the rest as relative to P.italicum. Since P.italicum is among the penicillin producing fungi, the common human mediated contaminants were tested including E.coli, Staphylococcus epidermidis and Citrobacter freundii to determine if they can co-exist with P.italicum.

Application in Real Citrus Samples Since P.italicum is one of the main post-harvest contaminant of lemons, infected lemon samples were used for method validation. Lemons were purchased from a local grocery, and rinsed thrice with 70% ethanol to disinfect any bacteria or fungi. 105 spores in 100 µL PBS buffer were placed on to the samples, which were then kept in air-tight container for 72 hours. Grown P.italicum were then isolated and used to validate the developed sensor. Stand-alone rhodamine 6G (R6G) modified PAA membranes and 96-well polystyrene plates were used in real-sample applications.

RESULTS Selective Detection and Staining of P. italicum In this study, sugar ligands and sugar-based AuNPs (sb-AUNPs) were utilized to selectively capture and stain the target fungi. Therefore, utilization of sb-AUNPs was employed to generate additional selectivity. For this purpose, both the derivatized sugars and sb-AUNPs were coated on different surfaces including paper electrode, glass and membrane surfaces or polystyrene 96well-plates. Figure 2 illustrates the summary of the steps and approaches applied on the different

ό ACS Paragon Plus Environment

Page 8 of 21

Page 9 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

sensing surfaces. Paper-electrodes with its gold-layer served as sensitive surface to monitor the fungal binding using the redox currents of potassium ferricyanide monitored with cyclicvoltammetry (CV). Glass and membrane surfaces provided visible and/or light microscopy monitoring of the microbial binding. Fungal binding was monitored for Polystyrene surface, 96well plate, via microplate reader at 490-595 nm. CV was applied to evaluate the P.italicum binding for gold-modified paper electrodes. 4-(Ngalactosyl)-mercaptophenyl and 4-(N-mannosyl)-mercaptophenyl were utilized as sugar ligands. 4-(N-galactosyl)-mercaptophenyl provided strong binding for 104 spores/mL of P.italicum concentration while the surface was not capable of any strong affinity for A.nidulans spores at 104 spores/mL. Capturing P.italicum on the modified gold surface resulted in about 95% decrease in oxidation current and about 50% shift of oxidation peak voltage towards higher voltage while the shifts were minimal for A.nidulans. Redox signals followed similar trend as shown in Supplementary Figure S2.

ύ ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2: Principle of P.italicum biomolecular sensing on different substrates. (A) Gold-coated paper electrode, (B) Stand-alone PAA membrane, (C) 3-mercaptopropyl-3-methoxysilane and (D) 3-aminopropyl-3-methoxysilane activated borosilicate surface, and (E) carboxylated polystyrene surface. Blue circle depicts thiol-gold binding; purple circle depicts glutaraldehydeamino group binding; green circle depicts GMBS chemistry and red circle depicts EDC/NHS chemistry related amide-bond formation.

Membranes and glass surfaces were employed as either modified or non-modified forms. These were tested to determine if they can be utilized in low resource settings or for on-site applications. The sb-AuNP staining was employed for spores/fruiting bodies including the hyphae. Spores, fruiting bodies and hyphae of P.italicum were also captured on PAA membranes by utilizing the sugar ligands which were immobilized on the PAA surface with glutaraldehyde chemistry [Figure 2B]. The results showed that the captured hyphae/fruiting bodies of P.italicum were not stained by sb-AuNP3 while sb-AuNP1 selectively stained the captured fruiting bodies and hyphae (Supplementary Figure S3).

Sugar-Lectin Interactions at Glass Surfaces Similar to the tested membranes, glass surface was designed as a simple platform for lateral test. Glass-surface was either modified with 3-aminopropyl-3-methoxysilane or 3-mercaptopropyl-3methoxysilane. Figure 2C/D showed the modification scheme. 4-(N-lactosyl) aniline [NLPDA] was obtained as the best ligand in comparison to L-cysteine, galactose-4-mercaptoaniline and Dglucosamine to capture P.italicum, and used for both 3-aminopropyl-3-methoxysilane or 3mercaptopropyl-3-methoxysilane activated glass surfaces. In Figure 3, NLPDA was coated on 3mercaptopropyl-3-methoxysilane activated glass-surface. Then, the dispersed spores (104 spores/mL) in acetate-buffer were introduced to the glass surface for 4 h, followed by rinsed with pure-water 5-times to remove any unbound spores. 4-sets of AuNPs were utilized to stain the immobilized spores for 8h incubation; the 4-sets of the AuNPs refer to the different sugarfunctionalized AuNPs [Figure 3A]. Right after incubation, the slide was sonicated for 5 min to remove any unbound or weakly bound AuNPs. After sonication, only one of the AuNPs provided visible color as seen circled are in Figure 3C, which was also clearly seen under lightmicroscopy at 10X [Figure 3B]. SEM images of the stained section revealed the capturing of

υτ ACS Paragon Plus Environment

Page 10 of 21

Page 11 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

P.italicum spores [Figure 3D], whose characteristic perfectly spherical form is shown in Figure 3E.

[A]

AuNP1

[B]

AuNP2

AuNP3

AuNP4

[C]

[D]

[E]

Figure 3: (A) AuNPs staining of P.italicum on glass-surface, (B) 10X light microscopy view and (C) digital camera image of AuNP3 stained P.italicum area, and their corresponding (D) 20K and (E) 50K SEM images. The glass surface was modified with NLPDA via GMBS on 3mercaptopropylmethoxysilane activated glass surface. 104 spores/mL P.italicum in pH 5.5 acetate-buffer concentration was applied to the surface, which was then stained with sugarfunctionalized gold nanoparticles.

It is noteworthy to mention that capturing P.italicum on glass surface followed unique characteristics. Clean and/or 3-mercaptosilane functionalized glass surfaces in water non-

υυ ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

specifically captured the P.italicum spores and hyphae and the spores and hyphae contacted the surface for at least 12 h. This can be attributed to presence of high sugar content of P.italicum spores and hyphae outer surfaces[25]. However, utilizing buffers did not allow this binding. In contrast, A.nidulans and G.cingulata did not show this type of strong non-specific binding. Because our goal was to develop a selective surface, the introduction of fungal spores to the surface was performed in buffers which were not allowed to dry. Light microscopy imaging was used to observe the capture/staining of 102 spores and hyphae/mL mixture on the glass surface. This approach did not provide any strong meaningful visibility or a clear indication of the selective capture of selective P.italicum.

Detection of P.italicum in 96-well plates

υφ ACS Paragon Plus Environment

Page 12 of 21

Page 13 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Figure 4: Standard graphics of P.italicum detection with 96-well plate. Wells coated with 1mg/mL of 4-(N-lactosyl) aniline (NLPDA). P.italicum spores between 104.78 and 103.47 were suspended in 50 mM pH 5.5 acetate-buffer. Absorbance was read at 560 nm. 96-well plate-based approach provided a 4 x102 spores/mL limit of detection, a linear detection range between 2.9 x103 - 6.02 x104 spores/mL (R2 0.9939) with 5% sensitivity. The narrow linear detection could be due to the fact that at higher concentrations fungal spores’ binding to the functionalized well surface is not specific, and they might be binding to the edges of the well as well. However, it was not possible to calculate a dynamic range in the of case lateral detection of the fungi due to the fact that the same surface area can be covered at different P.italicum concentration while the intensity per mm2 can be different.

When dissolved in acetate-buffer, P.italicum showed affinity towards all of the ligands including L-Cysteine and D-glucosamine. However, when pH was raised between 7.4-7.6, only NLPDA showed some degree of affinity towards P.italicum. In the case of 5 mg/mL of NLPDA coated wells, the calibration curve was not obtained at 1 mg/mL NLPDA coated wells [Figure 4]. This could be related to the fact that at high NLPDA concentrations, binding occurred at the polystyrene surface using interactions such as hydrophobic-hydrophobic and electrostatic interactions in addition to EDC/NHS directed covalent binding. Therefore, the concentration was kept minimum. Unfortunately, there is no fungal sensor using sugar-lectin interactions that we can use to compare our results. But we have provided a comparison with other analytical method; in particular a PCR-based theoretical detection limit for P.italicum recorded a detection limit of 2 x 105 conidia per sample extract [27].

Selectivity Sugar-ligand mediated capture and staining of A. nidulans, T. biforme and G.cingulata spores/hyphae were performed to evaluate selectivity of the ligands towards P.italicum. Even though limited recognition of these fungi by the ligands was monitored via light microscopy, the tested sugar-functionalized nanoparticles did not show any trend for the developed gold

υχ ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

nanoparticles. Therefore, it can be speculated that significant selectivity was obtained towards P.italicum using the gold-nanoparticles rather than the modified ligands.

Figure 5: Selectivity of the sensing platform: A) Sensor response to C.gloeosporioides, A.nidulans and T.biforme. (B) P.italicum. The polystyrene surfaces (96-well plate) were modified with NLPDA. All the fungi species were dispersed in acetate buffer. Absorbance was read at 560 nm.

Figure 5 shows the experimental conditions employed to functionalize and stain the captured fungi spores in order to objectively evaluate the selectivity of the developed sugar ligands and the synthesized sb-AuNPs. As seen from Figure 5A, at high concentrations, the detection of the fungi species were recognizable and a trend was obtained while at lower concentrations, the detection was limited and no trend could be observed. C.gloeosporioides did not provide any response for 104.7 spores/mL in the case of sb-AuNP1 staining while sb-AuNP2 staining gave responses even at 103.8 spores/mL. This indicates that the selectivity of NLPDA was lower in comparison to sb-AuNP1. In addition, A.nidulans and T.biforme did not provide any meaningful signals at lower concentrations. Figure 5B also clearly revealed that at the target concentration (104 spores/mL), 15% is the possible interference of the tested fungi.

In the case of lateral tests, the membrane surfaces provided better rejection capability towards C.gloeosporioides, A.nidulans and T.biforme when compared with the glass surfaces. At 104 spores/mL concentration, none of the fungi provided recognizable binding at the membrane

υψ ACS Paragon Plus Environment

Page 14 of 21

Page 15 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

surface. As shown in Figure S4, P.italicum did not allow the co-existence of E.coli, C.freundii and S.epidermidis. This suggests that the bacterial species which did not exhibit any resistance to penicillin are unable to interfere with the selective detection of P.italicum.

Validation in complex samples Lemon is among the major hosts of this pathogen, so lemon was utilized to mimic the possible application of the developed biosensor platforms. Sensor validation was performed for both qualitative and quantitative sensors. Infection and sample preparation of/from lemon is illustrated in Supplementary Figure S5.

Lateral test was performed with Rhodamine 6G (R6G) sequestered PAA membrane. PAA membranes are biodegradable and the functionalized membranes showed better selectivity for P.italicum in comparison to the glass surfaces, so their disposal into the environment should not result in accumulation unlike glass or other non-biodegradable surfaces. 104 spores/mL of P.italicum in acetate buffer was introduced on top of the NLPDA functionalized PAA-R6G membrane, where 6-8 h incubation was performed. As detailed in Supplementary Figure S6, captured P.italicum spores resulted in opaque white area. Light microscopy and SEM imaging endorsed that the opaque area was loaded with fungal spores. In the case of 4 x 103 spores/mL P.italicum concentrations, very small opaque areas were obtained. Since 106 spores/mL was accepted as the minimum infectious dose, the developed lateral test can be used for field tests especially in low resource areas.

Reproducibility, Accuracy and Stability In order to evaluate reproducibility, we prepared the calibration curves, where the error-bars were based on the standard-deviations. The standard deviation of the 96-well was less than 5% for five replicate measurements. Each run was carried out in 3 wells of 96-well plates, so each concentration was run 15 times. Qualitative test was performed using the 96-wells platform due to

its high-throughput characteristics. Continuous shaking under humid conditions at 35 ºC was applied to capture 104 and 4 x 103 spores/mL at initial concentrations of P.italicum spores onto NLPDA functionalized 96-well plates. Accuracy of the sensor response showed dramatic dependence on the sample preparation. In the case of 5-10 min centrifugation at 600 rpm, the

υω ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

spores can be isolated free from hyphae and conidiophore, for which spiked recovery was up to 95% (std ±4). In contrast, gravity-based precipitation of hyphae, the spiked recovery was 88% (std 11), and the differences can be assigned to matrix effect. In particular, the presence of long hyphae decreased the spiked recovery and reproducibility. This could be attributed to the fact that the nanoparticles showed strong selectivity towards the spores whereas the sugar ligands coated on the 96-well plates had less discrimination to the hyphae of P.italicum. Therefore, the spiked recovery was below 100%.

Stability of the test depends on the sb-AuNPs type. Throughout the study, we preferred to use the sb-AuNPs which did not result in color change upon interacting with fungal cells or color changes during incubation. This could be related to (i) either the sb-AuNPs did bind to outer surface of the cells or (ii) carbohydrate-decoration did not allow further changes on the nanoparticle which could result from microbial metabolism (e.g. enzymatic activity or action of extracellular products) or time effect. During the study, we observed that some of the nanoparticles were changing color when added to the fungal solution (which could be related to size and shape changes), or could be related to microbial metabolism.

CONCLUSIONS

A variety of sugar-ligands and sugar-based gold nanoparticles were developed to target the recognition and sensing of P.italicum. The ligands were immobilized onto a variety of platforms including polystyrene 96-well plates, glass surfaces, gold substrates and poly (amic) acid membrane surfaces. The developed sugar-ligands showed sensitivity and specificity towards the targets in comparison to the tested small molecules. In all cases, P.italicum was successfully detected at 1/100 of its infectious dose. The specific detection of spores and hyphae of P.italicum were demonstrated using these ligands. The sugar ligands and sugar ligands synthesized AuNPs were shown to advance the selectivity towards P.italicum by discriminating the recognition of Glomeralla cingulate, Aspergillus nidulans and Trichaptum biforme. It is noteworthy to mention that gold nanoparticles were shown to stain the fungal cells depending on size[26], but advanced specificity was provided by the utilized sugar ligands independent of size and shapes of the AuNPs. Although the membranes and glass surfaces provided only qualitative detection, further

υϊ ACS Paragon Plus Environment

Page 16 of 21

Page 17 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

improvements of these platforms design can provide semi-quantitative data and may provide cheap, an environmentally-sensitive option especially at low resource areas. In summary, the developed platforms provide inexpensive point of use and high-throughput quantitative detection of P.italicum in food samples. The methods are simple and can provide possible information on the contaminations of P.italicum in citrus samples at the field, during storage and transportation scenarios.

Supporting

Information

Available:

The

following

files

are

available

free

of

charge. Selective Sensing and Imaging of Penicillium italicum Spores and Hyphae using CarbohydrateLectin Interactions. The file includes digital image of Penicillium italicum growth and its antibacterial activity in vitro; figures of cyclic voltammetry and imaging-based detection of P.italicum and sampling of P.italicum from infected lemon.

ACKNOWLEDGEMENTS:

The authors acknowledge the National Science Foundation/Bill & Melinda Gates Foundation Grant #IOS-1543944. Dr. Ali Akgul acknowledges the Turkish National Educational Ministry for PhD scholarship.

References [1]

S. Yang, J. Zhou, D. Li, C. Shang, L. Peng, S. Pan, The structure-antifungal activity relationship of 5,7-dihydroxyflavonoids against Penicillium italicum, Food Chem. 224 (2017) 26–31. doi:10.1016/j.foodchem.2016.12.001.

[2]

S. Yang, L. Liu, D. Li, H. Xia, X. Su, L. Peng, S. Pan, Use of active extracts of poplar buds against Penicillium italicum and possible modes of action, Food Chem. 196 (2016) 610–618. doi:10.1016/j.foodchem.2015.09.101.

[3]

L. Palou, Chapter 2 – Penicillium digitatum, Penicillium italicum (Green Mold, Blue Mold), in: Postharvest Decay, 2014: pp. 45–102. doi:10.1016/B978-0-12-4115521.00002-8.

[4]

L.G. Hern?ndez-Montiel, J.L. Ochoa, Fruit Rot Caused by Penicillium italicum on Lemon

υϋ ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

( Citrus aurantifolia ) in Colima, Mexico, Plant Dis. 91 (2007) 767–767. doi:10.1094/PDIS-91-6-0767B. [5]

B. Van Grinsven, K. Eersels, O. Akkermans, S. Ellermann, A. Kordek, M. Peeters, O. Deschaume, C. Bartic, H. Dilië, E.S. Redeker, P. Wagner, T.J. Cleij, Label-Free Detection of Escherichia coli Based on Thermal Transport through Surface Imprinted Polymers, (n.d.). doi:10.1021/acssensors.6b00435.

[6]

J. Prattes, S. Heldt, S. Eigl, M. Hoenigl, Point of Care Testing for the Diagnosis of Fungal Infections: Are We There Yet?, Curr. Fungal Infect. Rep. (2016). doi:10.1007/s12281016-0254-5.

[7]

K. Hatano, K. Matsuoka, D. Terunuma, T. Shimazaki, M. Villeneuve, K. Matsuoka, K. Hatano, D. Terunuma, Y. Esumi, T. Nakamura, K. Hatano, D. Terunuma, K. Matsuoka, Y. Sakano, T. Suzuki, D. Miyamoto, M. Sugita, D. Terunuma, K. Morita, Y. Suzuki, Carbosilane glycodendrimers, Chem. Soc. Rev. 42 (2013) 4574–4598. doi:10.1039/C2CS35421G.

[8]

Y. Oda, T. Senaha, Y. Matsuno, K. Nakajima, R. Naka, M. Kinoshita, E. Honda, I. Furuta, K. Kakehi, A new fungal lectin recognizing alpha(1-6)-linked fucose in the N-glycan., J. Biol. Chem. 278 (2003) 32439–47. doi:10.1074/jbc.M305181200.

[9]

G.J.L. Ng, A.M.L. Quek, C. Cheung, T. V Arumugam, R.C.S. Seet, Stroke biomarkers in clinical practice: A critical appraisal, (2017). doi:10.1016/j.neuint.2017.01.005.

[10] M. Khater, A. De La Escosura-Muñiz, A. Merkoçi, Biosensors for plant pathogen detection, (2017). doi:10.1016/j.bios.2016.09.091. [11] E. Bulard, A. Bouchet-Spinelli, P. Chaud, R. Calemczuk, S. Fort, T. Livache, Carbohydrates as New Probes for the Identification of Closely Related Escherichia coli Strains Using Surface Plasmon Resonance Imaging, (n.d.). doi:10.1021/ac5037704. [12] L.K. Mahal, E. Schick, M.A. Bopp, M.J. Schneider, P. Oroszlan, M. Ehrat, M.R. Millar, P. Horner, R.L. Schnaar, D. Burton, et al., Catching bacteria with sugar., Chem. Biol. 11 (2004) 1602–4. doi:10.1016/j.chembiol.2004.11.017. [13] I. Yazgan, N.M. Noah, O. Toure, S. Zhang, O.A. Sadik, Biosensor for selective detection of E. coli in spinach using the strong affinity of derivatized mannose with fimbrial lectin, Biosens. Bioelectron. 61 (2014). doi:10.1016/j.bios.2014.05.008. [14] J. Gao, D. Liu, Z. Wang, Screening lectin-binding specificity of bacterium by lectin

υό ACS Paragon Plus Environment

Page 18 of 21

Page 19 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

microarray with gold nanoparticle probes, Anal. Chem. (2010). doi:10.1021/ac1022309. [15] Z. Shen, M. Huang, C. Xiao, Y. Zhang, X. Zeng, P.G. Wang, Nonlabeled quartz crystal microbalance biosensor for bacterial detection using carbohydrate and lectin recognitions, Anal. Chem. (2007). doi:10.1021/ac061986j. [16] H.J. Chung, T. Reiner, G. Budin, C. Min, M. Liong, D. Issadore, H. Lee, R. Weissleder, Ubiquitous detection of gram-positive bacteria with bioorthogonal magnetofluorescent nanoparticles, ACS Nano. (2011). doi:10.1021/nn2029692. [17] B.R. Panda, A.K. Singh, A. Ramesh, A. Chattopadhyay, Rapid estimation of bacteria by a fluorescent gold nanoparticle- polythiophene composite, Langmuir. (2008). doi:10.1021/la802171b. [18] C. Kaittanis, S.A. Naser, J.M. Perez, One-step, nanoparticle-mediated bacterial detection with magnetic relaxation, Nano Lett. (2007). doi:10.1021/nl062553z. [19] D. Kwon, J. Joo, J. Lee, K.H. Park, S. Jeon, Magnetophoretic chromatography for the detection of pathogenic bacteria with the naked eye, Anal. Chem. (2013). doi:10.1021/ac401717f. [20] I. Yazgan, Novel Poly ( Amic ) AcidMembrane Chemistries With ExperimentallyControlled Pore Size,Transport,and Disinfection Properties, State University of New York at Binghamton, 2016. [21] Penicillium italicum Wehmer, anamorph ATCC ® 48814TM, (n.d.). https://www.atcc.org/products/all/48814.aspx (accessed June 15, 2017). [22] N.M. Noah, S. Alam, O.A. Sadik, Detection of inducible nitric oxide synthase using a suite of electrochemical, fluorescence, and surface plasmon resonance biosensors, Anal. Biochem. 413 (2011) 157–163. doi:10.1016/j.ab.2011.02.010. [23] N.M. Noah, S.K. Mwilu, O.A. Sadik, A.A. Fatah, R.D. Arcilesi, Immunosensors for quantifying cyclooxygenase 2 pain biomarkers, Clin. Chim. Acta. 412 (2011) 1391–1398. doi:10.1016/j.cca.2011.04.017. [24] N. Du, C. Wong, M. Feurstein, O.A. Sadik, C. Umbach, B. Sammakia, Flexible Poly(amic acid) Conducting Polymers: Effect of Chemical Composition on Structural, Electrochemical, and Mechanical Properties, Langmuir. 26 (2010) 14194–14202. doi:10.1021/la101314j. [25] V. GRISARO, N. SHARON, R. BARKAI-GOLAN, The Chemical Composition of the

υύ ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Cell Walls of Penicillium digitatum Sacc. and Penicillium italicum Whem., J. Gen. Microbiol. 51 (1968) 145–150. doi:10.1099/00221287-51-1-145. [26] G.L. Jones, J.A. Bailey, R.J. O’Connell, Sensitive staining of fungal extracellular matrices using colloidal gold, Mycol. Res. 99 (1995) 567–573. doi:10.1016/S0953-7562(09)807156. [27] R. Haughland, M., Varma, L. Wymer, S. Vesper, Quantiiative PCR Analysis of Selected Aspergillus, Penicillium and Paecilomyces Species, System. Appl. Microbiol. 27 (2004), 198–210 .

φτ ACS Paragon Plus Environment

Page 20 of 21

Page 21 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

For TOC Only

φυ ACS Paragon Plus Environment