Cellular Response of Escherichia coli to Photocatalysis: Flagellar

Subscriber access provided by Iowa State University | Library

Article

Cellular Response of Escherichia coli to Photocatalysis: Flagellar Assembly Variation and Beyond Jingtao Zhang, Xueying Wang, Xinying Suo, Xing Liu, Bingkun Liu, Mingming Yuan, Guanglu Wang, Chengzhen Liang, and Hengzhen Shi ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.8b08475 • Publication Date (Web): 05 Feb 2019 Downloaded from http://pubs.acs.org on February 6, 2019

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.

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 42 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 Nano

ACS Paragon Plus Environment

ACS Nano 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

Cellular Response of Escherichia coli to Photocatalysis: Flagellar Assembly Variation and Beyond Jingtao Zhang*†, Xueying Wang†, Xinying Suo†, Xing Liu†, Bingkun Liu‡, Mingming Yuan†, Guanglu Wang†, Chengzhen Liang*§, and Hengzhen Shi*‡ †

Collaborative Innovation Centre of Food Production and Safety, Henan Key

Laboratory of Cold Chain Food Quality and Safety Control, School of Food and Bioengineering, Zhengzhou University of Light Industry, Zhengzhou 450002, China ‡

School of Material and Chemical Engineering, Zhengzhou University of Light

Industry, Zhengzhou 450002, China §

Biotechnology Research Institute, Chinese Academy of Agricultural Sciences,

Beijing 100081, China. Corresponding Author: *

E-mail: [email protected].

*

E-mail: [email protected].

*

E-mail: [email protected]

ABSTRACT Bacterial cells can be inactivated by external reactive oxygen species (ROS) produced by semiconductor photocatalysis. However, little is known about cellular responses to photocatalysis. For a better understanding of this issue, one strain of Escherichia coli (E. coli, hereafter named as MT), which has an increased ability to metabolize carbon sources, was screened out from the wild type (WT) E. coli K12 by repeated exposure to photocatalysis with palladium oxide modified nitrogen-doped 1


Page 2 of 42

Page 3 of 42 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 Nano

titanium dioxide (TiON/PdO). In this study, transcriptome sequencing (RNA-seq) of the WT and MT strains that were exposed or unexposed to photocatalysis were carried out. Cellular responses to photocatalysis were inferred from the functions of genes whose transcripts were either increased or decreased. Upregulation of expression of bacterial flagellar assembly genes used for chemotaxis was detected in cells exposed to semi-lethal photocatalytic conditions of the WT E. coli. Increased capability to degrade superoxide radicals and decreased bacterial flagellar assembly and chemotaxis were observed in MT E. coli compared to WT cells. We conclude that the differences in motility and intracellular ROS between MT and WT are directly related to survivability of E. coli during exposure to photodisinfection.

KEYWORDS: photocatalysis, cellular response, RNA-seq, flagellar assembly, metabolic alteration Besides antibiotics, several methods have been pursued for antagonism against pathogenic bacteria, such as bacteriocins,1 metal nanoparticles,2 and photocatalytic disinfection3,4 in anticipation of the need to bypass antibiotic resistance in these microorganisms. Among them, photocatalytic disinfection has attracted attention for advantages5 specific to this technique. The effectiveness of photocatalytic disinfection is exerted through the redox reactions caused by electron (e-) hole (h+) pairs generated by light excitation of photocatalysts. The disinfection reagents of reactive oxygen species (ROS) are produced through the reactions between holes and electrons. Due to this ROS-generating property, photocatalytic disinfection could avoid or minimize the potential formation of toxic disinfection by-products (DBPs) and may potentially eliminate a whole spectrum of pathogens effectively. 6 2


ACS Nano 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

Among various photocatalysts exhibiting antibacterial activity against a range of organisms, titanium dioxide (TiO 2 ) and derivative materials have been extensively studied as the most popular photocatalyst.4,7 These materials exhibit desirable properties for a disinfectant, including low toxicity and thermal and chemical stability.8,9 A number of studies have reported that TiO 2 and TiO 2 -based materials can produce ROS, including hydroxyl radicals (•OH), superoxide anion radicals (•O 2 −), and others, under UV or visible light illumination, which carry out the photocatalytic effects.5,

10

Previous reports11-14 on the mechanisms of photocatalytic disinfection

found that damage to the outer and inner cell membranes, a reduction in the enzymatic activity of coenzyme A, as well as DNA and protein damage all contribute to the efficacy of this treatment. However, little is known about the bacterial cellular responses to oxidative stress before the cell is killed under photocatalysis. Using an Affymetrix GeneChip for Pseudomonas aeruginosa, Kubacka et al. found that photodisinfection by TiO 2 -based nanocomposite films triggered a decrease in expression of genes specific for regulatory, signaling, and growth functions in P. aeruginosa PAO1 cells after two minutes under UV irradiation.15 Leung et al. found that in protein expression profiles of E. coli XL1-Blue exposed to photocatalysis with TiO 2 nanoparticles antibacterial activity did not significantly correlate with up-regulation of ROS-related proteins.16 In this study we examine the effects of photocatalysis on cellular responses in the model organism Escherichia coli (E. coli) K12 using whole transcriptome analysis. Photodisinfection can kill pathogenic microorganisms as well as those that are 3


Page 4 of 42

Page 5 of 42 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 Nano

resistant to antibiotics. Information about resistance of pathogenic microorganisms to photodisinfection is limited, particularly when this technique is applied in a clinical setting. To further our understanding of the salient mechanisms by which photocatalysis acts on bacterial cells as well as exploring whether microorganisms can develop resistance to photodisinfection, we generated derivative isolates of the original E. coli wild type (WT) through repeated exposure of this strain to photocatalytic irradiation and then looked for morphological changes in a photocatalysis-resistant mutant isolate (MT). The RNA-seq transcriptomics experiment is based on the detection of almost all transcribed genes under a given treatment by sequencing the sum of all mRNA collected at a given time. Transcriptomic experiments may alleviate the shortage of in vivo assays through investigation of specific pathways17 that can reveal cellular responses under these disinfection conditions. There is a growing body of work that uses transcriptomics to investigate cellular responses to nanomaterials. Massich et al. studied the biological response of HeLa cells to gold nanoparticles,18 and found that cells can recognize and react to the presence of the particles. Bouwmeester et al. demonstrated that silver nanoparticles induced clear changes in gene expression over a range of stress responses including oxidative stress, endoplasmic stress response, and apoptosis.19 Mortimer et al. showed that multiwall carbon nanotubes induced pronounced transcriptomic responses in P. aeruginosa PG201 such as changes in general stress response, sulfur metabolism, and nitrogen metabolism.20 In this report, RNA sequencing experiments were used to analyze differences in the 4


ACS Nano 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

WT and MT bacterial transcriptional response when exposed to semi-lethal photocatalysis conditions compared with that of unexposed cells. Through analysis of these differences we identified the main pathways for stress response by photocatalysis-treated E. coli cells, as well as similarities and differences between WT and MT strains. This research provides insight into mechanisms of photocatalytic disinfection, thus establishing a working genetic model for this sterilization technique, as well as serving as a guide for completely killing pathogenic bacteria to prevent resistant strains produced when this technique is applied in a clinical setting. Results and Discussion TiON/PdO nanoparticles were an excellent photocatalyst under visible light irradiation that could avoid cell damage from UV light. We used TiON/PdO as a photocatalyst in this study. Previous reports5,21 have shown that ROS, such as hydroxyl (•OH) and superoxide (•O 2 -) radicals, produced by TiO 2 nanoparticle photocatalysis can cause bacterial inactivation. The presence of •OH and •O 2 - in the TiON/PdO photocatalytic system was confirmed using a DMPO spin-trapping ESR technique. No obvious signals were detected in either ESR spectrum in the dark (Fig. 1a and 1b). However, signals of both DMPO-•O 2 - and DMPO-•OH were clearly observed under visible light illumination. Moreover, the signals of DMPO-•O 2 - and DMPO-•OH gradually increased with increasing irradiation time. These results confirm that both •OH and •O 2 - were produced by the TiON/PdO photocatalyst when exposed to light.

5


Page 6 of 42

Page 7 of 42 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 Nano

The colony morphotype of WT and MT E. coli strains on LB medium plates showed that after 36 hours of incubation at 37 °C the WT cells formed larger colonies than the MT cells (Fig. 1c). In addition to differences in size, MT colonies had an irregular margin compared to the round WT colonies. Photodisinfection treatment by TiON/PdO for WT and MT E. coli cells was carried out under the same conditions. The photodisinfection treatments were carried out at the concentration of 1 mg/mL photocatalyst and with the initial concentration of 107 CFU/mL E. coli under ~1 mW/cm2 visible light (400 - 700 nm) irradiation. We found that the WT E. coli was more sensitive than the MT E. coli to photodisinfection (Fig 1d). After visible light irradiation for 40 min, almost all of the WT E. coli cells (~2 ×107 CFU/mL) were inactivated, which is a seven-log reduction in viable cells prior to treatment. However, there was only about five-log reduction in the bacterial load of MT E. coli cells from ~2 ×107 CFU/mL to ~2 ×102 CFU/mL (see Fig. 1d). The MT strain also showed metabolic differences from the WT strain. Bacterial mutant isolates generated by mild exposure to photocatalytic disinfection were tested for utilization of 31 carbon sources (see table S1) using the Biolog EcoPlateTM-based redox technique, developed for strain identification.22,23 For further confirmation of differences in carbon metabolism by the MT strain, we used the BIOLOG EcoPlateTM to compare the abilities of WT and MT to utilize various carbon sources. The MT strain showed an increased capacity for utilization of 10 carbon sources compared to WT (Fig. 2). Furthermore, MT was significantly more efficient than WT at utilization of glucose-1-phosphate, pyruvic acid methyl ester, and D-malic acid carbon sources. 6


ACS Nano 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

From the above tests we demonstrated that MT cells are not just different from the WT cells morphologically and metabolically but also in their tolerance to photodisinfection. To systematically analyze differences in gene transcription between MT and WT cells exposed to TiON/PdO photocatalysis, we performed RNA-Seq using the MT50 and WT50 cells (E. coli cells with ~50% survival under TiON/PdO photodisinfection). In parallel, RNA was also collected from MT100 and WT100 (E. coli cells with ~100% survival rate in the absence of photodisinfection treatment). Both the WT and MT cells showed changes in gene expression patterns after treatment with photocatalysis (Fig. S1i and S1ii (a) and (b), respectively). Furthermore, gene expression in MT cells was different from that of WT with or without photocatalysis (see Fig. S1ii (c) and (d)). Although MT cells showed a higher rate of survivability against photodisinfection than WT (Fig 1d), there were only down-regulated differentially expressed genes (DEGs) in MT cells compared with WT under photocatalysis. In this study, DEGs were filtered with the criteria of |log 2 FC|>1, adjusted P value ≤0.01 (Table S2). The number of DEGs between MT50 vs MT100 and WT50 vs WT100 was 3365 and 3307, respectively (see Fig. S1ⅲ (a)). The ratio of the up-regulated DEGs of MT50 vs MT100 was similar to that of the WT50 vs WT100 comparison with 55.6% showing higher expression and 11.2% of the genes showing down-regulation. Interestingly, the ratio of the up-regulated DEGs in the MT100 vs WT100 comparison was 62.5% which is much higher than the 37.5% ratio of down-regulated DEGs (see Fig. S1ⅲ (b)), while in MT50 vs WT50 comparisons there were no up-regulated DEGs. Thus, there are more up-regulated 7


Page 8 of 42

Page 9 of 42 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 Nano

than down-regulated DEGs in E. coli cells treated with photocatalysis. This may be an active response process for the E. coli under photodisinfection and the MT/WT strains may express more genes to fight harmful environments. After many photodisinfection treatments, the MT strains may enhance the active response ability with the higher (62.5%) up-regulated DEGs than down-regulated DEGs (37.5%) compared with WT without photocatalysis treatment. To validate these DEGs, a quantitative real-time (qRT)-PCR assay was performed for 15 randomly selected genes with changes in their expression levels between MT50 and WT50 (Table. S1). The regression slope for RNA-Seq vs qRT-PCR is close to 1, showing a strong positive correlation between RNA-Seq data and qRT-PCR data, and thus validating the RNA-seq data (Fig. S2). Previous reports have found that the cell membrane can be damaged by photocatalysis,24 and so in this work we focused on outer cell membrane-related genes (Table S3) due to their importance in maintaining the integrity of cells. Due to the position of cell membrane proteins within the cell wall, they are likely to sustain damage when the outer cell wall is compromised by interaction with ROS. Damage to these membrane proteins will manifest as heightened expression of the genes encoding them in an effort by the cell to maintain functionality and integrity of the membrane following photocatalysis. Five genes related to out membrane proteins, including ompL (P76773 porin), ompT (P09169 protease), ompW (P0A915 membrane protein), ompG (P76045 membrane protein), and ompN (P77747 porin), were up-regulated in comparisons of MT50 and WT50 with MT100 and WT100 (Fig 3a). 8


ACS Nano 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

Dartigalongue et al. showed that the porin OmpL has resistance to sublethal concentrations of oxidizing agents.25 The protease OmpT belongs to the omptin family of proteases and functions as a defense mechanism in E. coli.26 OmpG was the main porin in E. coli and could respond to environmental changes.27 OmpN may be involved in sugar translocation, signal transduction, anchoring the flagellum, and various cellular processes.28 The biological function of OmpW is not clear, but current data suggest that it is a versatile protein.29 It is involved in the protection of bacteria against various forms of environmental stress, such as osmosis30 and oxidation.31 However, three outer membrane-related genes, ompX (P0A917 membrane protein), ompA (P0A910 membrane protein), and ompC (Q1PI90 membrane protein) were found to be down-regulated. OmpX was shown to be involved in the invasion of host cells and bacterial defense against the complement systems of the host.32 OmpX was also closely related to OmpA.33 OmpA serves a number of functions in E. coli. Its biophysical role is as a receptor for bacteriophages and bacteriocins. OmpA has been demonstrated to be responsible for bacterial survival in blood via the evasion of complement attack and suppression of immune cells. OmpA is also involved in maintaining the stability and integrity of the bacterial membrane.34 The expression of ompC is regulated by osmolarity.35 The loss of ompC in E. coli reduces its ability to invade intestinal cells.36 The up-regulated membrane genes could be due to photocatalytic damage to the membrane and lead to cellular defense against photodisinfection, while the down-regulated genes could be due to the loss of some unnecessary invasion function in this environment or the changes in cellular 9


Page 10 of 42

Page 11 of 42 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 Nano

osmolarity.37 These findings led us to more closely examine the expression of genes responsible for maintenance of osmotic potential across the cell membrane. Expression of the genes osmY (P0AFH8 periplasmic protein), osmB (P0ADA7 lipoprotein), osmC (P0C0L2 peroxiredoxin OsmC), osmE (P0ADB1 transcriptional regulator), and osmF (P33362 osmoprotectant uptake system substrate-binding protein) were shown to be down-regulated in MT50 and WT50 compared with MT100 and WT100, respectively (Fig. 3b and Table S3). In comparisons of MT100 vs WT100, all of the above five membrane osmotically-related genes were up-regulated. In previous reports, OsmY, OsmB, OsmC, and OsmF were osmotically-inducible proteins.38,39 Those genes were down-regulated in MT50 and WT50, but up-regulated in MT100 compared with WT100, which could indicate mutations that were induced by selection pressure from photocatalysis. In previous reports, intracellular ROS could be increased under photocatalysis.40,41 To explore the connection between osmotic pressure and ROS, we examined expression of genes encoding superoxide radical degradation enzymes as a measure for probing levels of intracellular ROS (Fig 3c and Table S3). In comparisons of MT50 and WT50 with MT100 and WT100, five genes encoding the super-oxide radical inactivation enzymes catalase HPII, superoxide dismutase [Mn], superoxide dismutase [Cu-Zn], superoxide dismutase [Fe], and catalase-peroxidase16,42,43 were all down-regulated. However, by comparing MT100 with WT100, four genes were found to be significantly up-regulated. In addition, there was no significant difference in expression of these five genes between treatments of MT50 vs WT50, suggesting that 10


ACS Nano 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

they are highly upregulated in WT strains during oxidative stress. This finding shows that changes in the MT strain may enhance its capacity for handling ROS and tolerance to photodisinfection. To better understand the overall patterns of gene expression, i.e., to identify pathways that were up- or down-regulated in response to photocatalysis for each strain under both treatment conditions, the Kyoto Encyclopedia of Genes and Genomes (KEGG) and BioCyc pathway enrichment analyses were performed. All the pathways that were significantly enriched by up- or down-regulated genes are shown in Table S4, while Fig. 4 shows a selection of the most interesting pathways. Under photocatalysis, only three pathways were significantly enriched with up-regulated genes in comparisons of WT50 and WT100. These pathways include phenylalanine metabolism, ascorbate and aldarate metabolism, and flagellar assembly. Only one pathway, phenylalanine metabolism, was significantly enriched for up-regulated genes in MT50 vs MT100 comparisons. Phenylalanine metabolism and ascorbate and aldarate metabolism have been reported to play a role in the stress response.44-46 Moreover, the phenylalanine metabolism pathway was also significantly enriched for up-regulated genes in MT100 samples compared to WT100 (Fig S3 and Table S3). Pathway enrichment analysis also yielded the finding that the flagella assembly pathway was significantly enriched for up-regulated genes in WT50 vs WT100 comparisons (Fig. 4). This discovery is important in light of understanding that bacterial motility depends on flagellar biosynthesis47 and that E. coli cells subjected to photocatalysis depend heavily on chemotaxis to escape high concentrations of toxins, 11


Page 12 of 42

Page 13 of 42 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 Nano

ROS, or other harmful environments.48 Interestingly, flagella assembly genes that are up-regulated in comparisons between treatments for the mutant strain are significantly down-regulated in comparisons between MT50 and WT50 or between MT100 and WT100, indicating that expression of these genes may be impaired in the mutant relative to the wild type (Fig. 4), possibly as a result of repeated selection pressure under photocatalysis. The mutant strain may decrease its response ability of flagellar assembly and chemotaxis after exposure to photodisinfection. Some of the flagellar assembly and chemotaxis-related genes such as those in the C ring (fliG, fliM, and fliN),49 MS ring (fliF),50 the export gate (fliO, fliP, fliQ, and fliR)51,52 and the ATPase ring complex (fliI, fliH, and fliJ)53 were down-regulated in the mutation strain under photodisinfection treatment of MT50 compared with MT100 (Fig. S4 and Table S3). The FliG, FliM, and FliN genes forming the C ring, which is responsible not only for torque generation but also for switching the direction of motor rotation and flagellar assembly,54,55 were shown to be down-regulated in MT50 compared with MT100 (Fig. S4 and Table S3). Thus, the mutant strain may change part of its response to photodisinfection in flagellar assembly and chemotaxis. Predictably, the flagella assembly pathway was significantly enriched for up-regulated genes in the WT50 vs WT100 comparison, indicating that this pathway is part of a normal, chemotactic stress response to photocatalysis for E. coli cells. Fig. 5 is a schematic of a flagellum mapped with up-regulated and down-regulated genes from WT50 vs WT100 and MT100 vs WT100 comparisons. Expression levels of the flagellar assembly and bacterial chemotaxis-related genes 12


ACS Nano 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

from each strain and treatment (WT50, MT50, WT100, and MT100) are shown in Fig. S4. The differences in motility between MT and WT cells was further confirmed by puncture tests (Fig. 6a). In this assay the movement ability of the WT cells was better than that of MT cells, which is consistent with the results of transcriptomic and pathway enrichment analysis. The intracellular ROS content was assayed by H 2 DCFDA (Fig. 6b). Results of this experiment show that levels of intracellular ROS in WT100 and MT100 cells are at a low level in the absence of exposure to photocatalysis, with ROS levels in the mutant strain lower than that of wild type. However, under photocatalytic conditions the ROS levels increased for both strains, though MT50 exhibited lower ROS than WT50 cells. These results support the findings of transcriptomic analysis. In addition, the levels of intracellular ROS in WT-Control and MT-Control were slightly higher than in WT100 and MT100 cells, which indicated that the visible light may have a weak influence on the intracellular ROS of the cell. The intracellular ROS fluorescent images of WT-Control, MT-Control, WT100, MT100, WT50, and MT50 cells confirmed the quantified result of fluorescence spectrophotometry (Fig. 6c). We used qRT-PCR to further test eight selected DEGs related to superoxide radical degradation and flagellar assembly of the MT and WT cell without using a photocatalyst but with only visible light irradiation as control. The result of qRT-PCR demonstrated that the expression levels of MT-Control vs MT100 and WT-Control vs WT100 were not significantly different

13


Page 14 of 42

Page 15 of 42 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 Nano

(Fig S5). Herein, our results should be directly related to the ROS activation and the light should have no significant influence on the proposed mechanism. Under photocatalysis of MT50 and WT50, compared with MT100 and WT100, respectively,

many

pathways

such

as

electron

transport

chain/oxidative

phosphorylation, microbial metabolism in diverse environments, pentose phosphate pathway, pyruvate metabolism, fatty acid metabolism, glycolysis, TCA cycle, and glutathione metabolism were significantly enriched in down-regulated genes. However, after prolonged and repeated exposure to photocatalysis (67 treatments) most of these pathways were significantly enriched for up-regulated genes in MT100 when compared with WT100 (see Fig. 4 and Table S4). These pathways, including oxidative phosphorylation, superoxide radical degradation, fatty acid degradation, and TCA cycle appear to be enhanced in the mutant MT100 cells compared to wild type but are inhibited overall during photocatalytic stress. Conversely, the pathways of flagellar assembly, bacterial chemotaxis, and purine metabolism (Table S4) of the MT100 cells are decreased compared with WT100. Similarly, MT colonies were small with irregular margins when cultured on solid media (Fig. 1c), supporting our enrichment analysis showing that the metabolism of these cells has been compromised in the absence of stress. However, the reward for MT cells is an improved insensitivity to photodisinfection compared to WT cells (Fig. 1d). Based on the results of the transcriptomics analysis, we propose that in MT cells not exposed to photocatalysis, the enhanced metabolic pathways of oxidative phosphorylation, fatty acid degradation, TCA cycle, and others listed above are helpful for the cell to 14


ACS Nano 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

increase capacity for superoxide radical degradation as the chief strategy for gaining insensitivity to photocatalysis. Conclusions In conclusion, we report the cellular response of E. coli to photocatalysis of TiON/PdO nanoparticles under visible light illumination. We obtained a mutant strain (MT) with a small colony size and irregular margin morphology by repeated exposure to photocatalytic disinfection using a TiON/PdO catalyst. RNA-seq experiments were then conducted to compare the transcriptional responses of the WT and MT strains when exposed (WT50 and MT50) or unexposed (WT100 and MT100) to photocatalysis. Under photocatalysis treatment, the metabolic processes of the MT cell such as oxidative phosphorylation, TCA cycle, glycolysis, pyruvate, fatty acid, and glutathione were decreased. Chemotactic motility was enhanced through upregulation of genes in the flagellar assembly pathway during stress response in the mutant strain produced under photocatalytic selection pressure. However, this same strain showed an inhibited metabolism and a decreased ability to move compared to wild type cells. These results suggest that E. coli may have adapted its response to photocatalysis away from escape toward an enhanced ability for superoxide radical degradation following repeated exposure to and selection by photodisinfection. This work forms a foundation for development of more sophisticated and effective photocatalytic disinfection treatments. Materials and Methods

15


Page 16 of 42

Page 17 of 42 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 Nano

Photocatalyst synthesis and identification of radicals The palladium oxide-modified, nitrogen-doped titanium dioxide (TiON/PdO) nanopowder was synthesized by a sol-gel protocol according to previously reported methods,24, 56 and as briefly described below. A mixture of titanium tetraisopropoxide and tetramethylammonium hydroxide (mol ratio at 5:1) was first made in absolute ethanol. Then, a proper amount of Pd(acac) 2 dissolved in CH 2 Cl 2 was added (Pd/Ti mol ratio at 0.5%). The mixture obtained was loosely covered and stirred until a homogenous gel formed. The hydrolysis of precursors was initiated by exposure to moisture in air. The gel was aged in air for several days to allow further hydrolysis and drying. The xerogel was then crushed into a fine powder and calcinated at 500°C in air for 5 h to obtain the desired nanoparticle photocatalysts. The formation of •OH and •O 2 - radicals from the TiON/PdO substrate was characterized using a JES-FA200 (JEOL, Japan) Electron Spin Resonance (ESR) instrument at room temperature. The radicals’ spin, trapped by 5,5-Dimethyl-1-pyrroline N-oxide (DMPO), was measured as ESR signals. The involvement of •O 2 - was examined in methanol due to the instability of •O 2 - in water, and the involvement of •OH was detected in ultrapure water. In the samples, 50 mg of the TiON/PdO photocatalyst was mixed with 50 mL of a DMPO (100 mM) solution. Spectra were recorded after 5 min and 10 min irradiation by a 300 W xenon lamp with a UV cut-off filter and an infrared reflection filter to provide visible light (400-700 nm). The sample without irradiation was used as a negative control in these experiments. Bacterial culture for photocatalysis 16


ACS Nano 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

E. coli (K12, ATCC 15597) cells were inoculated in 5 mL of Luria-Bertani (LB) medium and incubated at 37 °C overnight. Cells were then washed twice with a phosphate buffer solution (PBS, pH 7.0) and suspended in 5 mL PBS (~109 CFU/mL) for measurement of photocatalytic disinfection. Photocatalytic disinfection performance Photocatalytic disinfection protocols followed methods used in our previous work,57 briefly described below. Ten mg TiON/PdO nanopowder was mixed ultrasonically for 10 min with 9.9 mL PBS buffer solution. Then, 0.1 mL of the above E. coli suspension was added to the above prepared mixture at a concentration of 107 CFU/mL for photocatalytic disinfection experiments. A 300 W xenon lamp with a UV cut-off filter (λ>400 nm) and an infrared reflection filter were used as the visible light (400-700 nm) source for photocatalysis. In this study, the light intensity striking the cells was ~1 mW/cm2. A 0.1 mL aliquot of the cell suspension was sampled at regular time intervals. Following appropriate dilution with PBS buffer solution, the 0.1 mL aliquots were spread onto LB agar plate and incubated at 37 °C for 24 h. The number of viable cells were counted as colony-forming units (CFUs). All analyses were conducted in triplicate. MT cell selection After photodisinfection, 20 isolated E. coli colonies were picked from the LB agar plate and incubated in 5 mL LB liquid medium at 37 °C overnight. Those colonies were used in the subsequent photocatalytic disinfection experiment. We repeated this 17


Page 18 of 42

Page 19 of 42 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 Nano

test 67 times and selected a small colony with an irregular margin morphology (thus named MT), for further investigation of the metabolic and transcriptional basis for tolerance to photodisinfection. Biolog EcoPlateTM assay To test metabolic capabilities of the selected isolates, 10 μL of the overnight-cultured WT or MT cells were added to every well of Biolog EcoPlateTM plates. The initial optical densities (ODs) of every well were measured before inoculation. The plates were incubated at 37 °C for up to 48 h. The OD at 590 nm (OD590) in each well was recorded every 12 h using a Biolog microstation and associated software (Biolog OmniLog version 4.1). The metabolic activity of WT and MT in the Biolog EcoPlateTM was expressed as the average well color development (AWCD) and calculated as reference.22 RNA sequencing and data analysis RNA was isolated from MT50 and WT50 cells (E. coli cells with ~50% survival under TiON/PdO photodisinfection) using a TRIzol kit (Invitrogen, USA), following the manufacturer’s protocol. In parallel, the RNA was also isolated from MT100 and WT100 (E. coli cells with ~100% survival and without photodisinfection treatment). All experiments were performed in triplicate for each of three biological replicates. The concentration and purity of total RNA were assessed by a Nanodrop2000 (Thermo Fisher Scientific, MA, USA), and the integrity of total RNA was evaluated on 2% agarose gels and with an RNA Nano 6000 Assay Kit for the Bioanalyzer 2100 18


ACS Nano 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

system (Agilent Technologies, CA, USA). mRNA purification and cDNA synthesis were performed using 10 μg total RNA per sample, and rRNA was removed using an Epicentre Ribo-zero rRNA Kit (Epicentre, USA). Sequencing libraries were generated using NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (#E7530L, NEB, USA) following the manufacturer’s protocol and index barcodes were added to attribute sequences to each sample. Library construction was checked by Qubit® RNA Assay Kit in a Qubit® 3.0 for preliminary quantification. Insert size was assessed using the Agilent Bioanalyzer 2100 system (Agilent Technologies, CA, USA) and accurately quantified by StepOnePlusTM Real-Time PCR System (Thermo Fisher Scientific, MA, USA) (library valid concentration>10 nM). After the library passed all tests, the samples were sequenced using the Illumina HiSeqTM X TEN platform. Clean sequencing reads were obtained by removal of raw data reads containing more than one low quality (Q-value ≤ 20) base, reads with 5' primer contaminants, reads without 3' primer, reads lacking the insert, reads with a poly A tail, and reads shorter than 18 nt. Clean reads were aligned with the reference genome (ftp://ftp.ensemblgenomes.org/pub/release-32/bacteria/fasta/bacteria_98_collection/es cherichia_coli_k_12_gca_000974405). The unique mapped reads of RNA sequence were used for further analysis. Transcript abundance and Spearman correlation coefficient analysis was calculated by Fragments per kilo-base per million reads (FPKM), which was measured by the software RSEM.58 Transcripts with a Padj

Cellular Response of Escherichia coli to Photocatalysis: Flagellar

Recommend Documents