A Specific High Toxicity of Xinjunan (Dioctyldiethylenetriamine) to Xanthomonas by Affecting the Iron Metabolism

ABSTRACT Xanthomonas spp. encompass a wide range of phytopathogens that brings great economic losses to various crops. Rational use of pesticides is one of the effective means to control the diseases. Xinjunan (Dioctyldiethylenetriamine) is structurally unrelated to traditional bactericides, and is used to control fungal, bacterial, and viral diseases with their unknown mode of actions. Here, we found that Xinjunan had a specific high toxicity toward Xanthomonas spp., especially to the Xanthomonas oryzae pv. oryzae (Xoo), the causal agent of rice bacterial leaf blight. Transmission electron microscope (TEM) confirmed its bactericidal effect by morphological changes, including cytoplasmic vacuolation and cell wall degradation. DNA synthesis was significantly inhibited, and the inhibitory effect enhanced with the increase of the chemical concentration. However, the synthesis of protein and EPS was not affected. RNA-seq revealed differentially expressed genes (DEGs) particularly enriched in iron uptake, which was subsequently confirmed by siderophore detection, intracellular Fe content and iron-uptake related genes transcriptional level. The laser confocal scanning microscopy and growth curve monitoring of the cell viability in response to different Fe condition proved that the Xinjunan activity relied on the addition of iron. Taken together, we speculated that Xinjunan exerted bactericidal effect by affecting cellular iron metabolism as a novel mode of action. IMPORTANCE Sustainable chemical control for rice bacterial leaf blight caused by Xanthomonas oryzae pv. oryzae need to be developed due to limited bactericides with high efficiency, low cost, and low toxicity in China. This present study verified a broad-spectrum fungicide named Xinjunan possessing a specific high toxicity to Xanthomonas pathogens, which were further confirmed by affecting the cellular iron metabolism of Xoo as a novel mode of action. These findings will contribute to the application of the compound in the field control of Xanthomonas spp.-caused diseases, and be directive for future development of novel specific drugs for the control of severe bacterial diseases based on this novel mode of action.

respiration was earlier reported as mode of action of Xinjunan (25), but the specific physiochemical mechanism is still not clear, and these effects may not be the initial action sites. On the other hand, in vitro experiment showed that the toxicity of Xinjunan to different types of microorganisms varied greatly, suggesting that there were significant differences in the mode of action on different microorganisms (16). Thus, it is necessary to clarify the mode of action of Xinjunan on specific microorganisms for its effective use.
The aims of this study were: (i) to determine the toxicity of Xinjunan against Xanthomonas species and compare with other plant pathogens; (ii) investigate the physiochemical changes of Xanthomonas at cellular and molecular levels in response to Xinjunan treatment; and (iii) to study the specific mode of action of Xinjunan on Xanthomonas spp.

RESULTS
Xinjunan exhibited specific high toxicity against Xanthomonas spp. Nineteen strains, including plant-pathogenic fungi, bacteria, and plant growth-promoting rhizobacteria (PGPR), were selected to evaluate the antimicrobial activity of Xinjunan. The effective concentration for 50% inhibition (EC 50 ) of the chemical on each strain was measured as a reduction in cell density or colony diameter after incubation. EC 50 values of Xinjunan to 19 strains differed from 0.33 to 62.66 mg/mL, as shown in Table 1. Xinjunan showed the strongest effect against Xanthomonas spp. with EC 50 ranging from 0.33 to 0.80 mg/mL. This specific high toxicity suggests that Xinjunan may have a unique mode of action to Xanthomonas spp. Meanwhile, MIC of Xinjunan against Xanthomonas spp. was also determined by agar dilution method as shown in Table 2. Xinjunan showed bactericidal effect against X. oryzae pv. oryzae (Xoo) strains with an MIC of 16 mg/mL, which was significantly lower than others. Furthermore, we tested the toxicity of Xinjunan on 10 different Xanthomonas oryzae strains. The EC 50 values varied from 0.33 to 5.03 mg/mL and the average EC 50 was 0.996 mg/mL (Table S1). Thus, we took Xoo as a model for the following studies.
Effect of Xinjunan on Xoo cell morphology. The growth curve was determined by adding a series of Xinjunan with concentration gradients. The growth rate of Xoo was completely inhibited while culturing in media with Xinjunan at MIC. Xinjunan had no  S2). Based on these results, Xinjunan with a concentration of 1/4 MIC and 0.5 h/4 h were selected as treatment condition for cell structure observation. Transmission electron microscope (TEM) was carried out to evaluate the level of cell wall damage and intracellular modification in Xoo PXO99 A treated by Xinjunan. Xoo cells exposed to 1% dimethylformamide (DMF) served as the negative control, which showed an intact outline of cell wall and a peptidoglycan layer (Fig. 1a). Xoo cells treated with Xinjunan at 1/4 MIC for 0.5 h were partial sunken or corrugated (Fig. 1b), and these situations were further exacerbated after treated for 4 h. In addition, cytoplasmic vacuolation and cell wall degradation can be found in some cells after 4 h (Fig. 1c). TEM results confirmed the bactericidal activity of the chemical, and suggested that it should take a certain time for Xinjunan to produce a marked effect. Xinjunan inhibited DNA synthesis of Xoo. In order to study the effect of Xinjunan on macromolecules synthesis of Xoo, different concentrations of Xinjunan were set up to investigate the DNA/protein/exopolysaccharides (EPS) contents variation in Xoo cells. As shown in Fig. 2, Xinjunan has a significant inhibitory effect on DNA synthesis of Xoo. Under 0.2 mg/mL treatment, the inhibition rate of DNA synthesis was 13.5%, and in a certain range of concentration, the inhibition effect improved with the concentration of agent increasing. In the same condition, Xinjunan had little effect on the synthesis of extracellular polysaccharide or protein, indicating that the drug may exert a specific action on the Xoo cells by inhibiting the biosynthesis of bacterial DNA.
Effect of Xinjunan on Fe homeostasis of Xoo. Based on the above findings, we aimed to figure out the target genes that might be interfered by Xinjunan. We first adopted RNA-seq to determine the overall situation of transcriptional changes under    S3 to 4). The gene ontology (GO) enrichment analysis was applied to describe characteristics and reaction network of DEGs. GO terms with corrected P value less than 0.01 were considered significantly enriched by differential expressed genes. The results showed that the DEGs enriched not only in response to stress responses, but also, in particularly, involved in iron uptake (Fig. 3). To verify these results, we subsequently detected the siderophore production using a Chrome azural S (CAS) agar plate. Results showed that when Xinjunan was added in vitro, the siderophore produced by Xanthomonas was significantly increased, and the degree of increase was positively correlated with the concentration of this chemical (Fig. 4a). Meanwhile, we measured the intracellular iron content of Xoo in iron-depleted or ironreplete conditions using inductively coupled plasma optical emission spectrometry (ICP-OES). After growth for 3 h in the iron-depleted MMX medium, the iron accumulation in cells of Xoo strains remained the same level either treated with Xinjunan or not. However, in the iron-replete medium (MMX 1 100 mM FeCl 3 ), the iron concentration in the Xoo were significantly increased by 2.38-fold under Xinjunan treatment (Fig. 4b). When additional dipyridyl (DIPY), the specific chelating agent for iron ions, was added into the medium, the intracellular iron level immediately reduced to the same level as treated with Fe alone. Further investigation regarding iron-uptaking genes transcription level was also determined by quantitative PCR. FeoB, pvsA, and pvsB were involved in Xanthomonas iron Specific Toxicity of Xinjunan against Xanthomonas Microbiology Spectrum transportation and siderophore synthesis, whereas fur plays negative regulatory role on iron uptake. Real-time RT-PCR analysis of these genes revealed that feoB, pvsA, and pvsB expression levels were higher, whereas fur expression was downregulated compared with control ( Fig. 4c). Taken together, these results supported the view that Xinjunan could affect Fe homeostasis of Xoo by disrupting bacterial iron uptake. Xinjunan activity is associated with environmental Fe content. To further confirm the role of Fe content in Xinjunan anti-bacterial activity, we adopted confocal laser scanning microscopy to visualize the cell viability in response to Xinjunan treatment with or without Fe. Bacteria with intact cell membranes stain fluorescent green, whereas bacteria with damaged membranes stain fluorescent red. When Xoo were solely treated with Xinjunan, the cells exhibited the same level of green fluorescence compared to mock (Fig. 5) in both 0.5 h and 3 h treatment. However, when added 100 mM FeCl 3 together with Xinjunan, the majority of bacteria cells became yellow-red after 0.5 h treatment, and completely turned into red "dead" cells after 3 h. When additional DIPY, the specific chelating agent for iron ions, was introduced, the Xoo cells remained green for the whole time. In addition, we also investigated the growth curve in response to different Fe condition. The results were consistent with the cell viability observation (Fig. 6). When Xinjunan was added alone, there was little inhibition compared with the control. However, when additional Fe 31 was added, an obvious inhibitory effect could be observed, and when DIPY was added simultaneously, the inhibitory effect disappeared. Thus, we concluded that the environmental Fe content is necessary for the bactericidal activity of Xinjunan.

DISCUSSION
Chemical control is still an important tool for BLB management due to the lack of stable resistant cultivars against the disease in China. A bacterial blight resistance analysis of China National authorized rice varieties showed that only 78 varieties (4.3%) showed disease resistance among the 1,805 tested varieties in a regional trial, accounting for 4.3% (26). Bismerthiazol had been the most commonly used bactericide for controlling BLB in China since 1970s; however, bismerthiazol-resistant Xoo has been reported at various fields in different regions of China (10,(27)(28)(29). The use of some other bactericides like copper agent and agricultural streptomycin are recently forbidden or restricted in China. As a result, the effective BLB controlling agents are very few in the market. Therefore, introduction of novel bactericides for sustainable BLB control is recommended.
Despite Xinjunan being applied in agricultural use for almost 30 years in China, it is mainly used to control fungal diseases, especially for fruit trees diseases (16,30). Meanwhile, little is known about the mode of action of Xinjunan on different types of pathogens. In this study, Xinjunan seems to be more effective against bacterial pathogens, indicating that Xinjunan may be a good candidate for controlling bacterial diseases. The results revealed that Xanthomonas spp., among all tested pathogens, were particularly sensitive to this chemical. Considering the differences in structure and chemical properties compared to other commonly used bactericides, it was suggested that the Xinjunan might have a unique mode of action against Xanthomonas and is safe to the nontarget organisms. On the other hand, it might be difficult to produce cross-resistance with previous bactericides. In addition, this compound can be synthesized from diethylenetriamine and bromoalkane in one step with a yield high up to 87.6% (31), which indicated that this bactericide has great The current study confirmed the addition of Xinjunan-caused morphological changes in Xoo cells, mainly the cell wall structure damage, resulting in leakage of cellular components and cytoplasmic vacuolation. The results showed that the compound had bactericidal effect rather than bacteriostatic effect. However, this kind of damage seems to take certain time, and the degree is not as strong as other reported drugs. For instance, terpenoids target the cytoplasmic membrane, which could disrupt lipid-protein interaction, increase membrane permeability, and ultimately destroy cell integrity (32). Exposure to diffcidin or bacilysin makes cell walls loose and porous, distorting from their normal shape or even be ruptured (33). In connection with the latter results that the Xinjunan did not damage the cell integrity without iron or with extra DIPY addition, the quantitative detection results showed that the chemical could inhibit Xoo nucleic acid synthesis. It is reasonable to believe that Xinjunan may not directly damage and interact with the cell wall structure  of Xoo but interfere with the related pathways of cell wall synthesis by affecting the expression of specific genes. RNA-seq of Xoo in response to Xinjunan treatment offered us a global view to seek potential mode of action. As showed by GO enrichment analysis, the DEGs are interestingly involved in iron uptake terms. Iron is vital in the progress of life; excessive or insufficient intake of iron will both affect cell metabolism (34,35). This was consistent with the results that Xinjunan inhibited Xoo nucleic acid synthesis.
Based on the above findings, we sequentially verified the effect of Xinjunan on cell growth and viability under strict control of iron environment. LCSM images of Xoo cells intuitively appeared almost entirely green in the untreated and sole addition of Xinjunan groups, indicating most cells remained alive and intact under iron-depleted condition. By contrast, cells exposed to Xinjunan together with Fe for 4 h would lead to red dead cells. Additionally, when iron chelating agent was added, almost all Xoo cells remained green, implying that the bactericidal activity was closely related with Fe content. This meaningful discovery will guide the scientific use of this chemical, and furthermore, will provide a possibility of compound synergism. In summary, the specific high bactericidal effect of Xinjunan on Xoo was achieved by affecting the cellular iron metabolism (Fig. 7). When Xoo is treated with Xinjunan, a large amount of environmental Fe 31 entered and accumulated in the cells, which may convert into Fe 21 and destroy the intracellular iron homeostasis. It was reported when the concentration of Fe 21 in cells is too high, ferrous ions catalyze Fenton reaction to produce hydroxyl radicals with high activity, which destroy DNA and lead to the cell death. To our knowledge, this was a novel mode of action that has not been recorded in the fungicide resistance action committee (FRAC). Further exploration of the action site/target of the chemical will not only be beneficial to consolidate the research foundation of action mechanism of fungicides, but also be directive for future development of novel specific drugs to control the diseases caused by Xanthomonas species.
Conclusion. In this research, we proved Xinjunan seems to be more effective against bacterial pathogens, especially to Xanthomonas spp., indicating there might be unique target site(s) of this bactericide. We investigated the physiochemical changes of Xanthomonas at cellular and molecular levels in response to Xinjunan treatment,

Specific Toxicity of Xinjunan against Xanthomonas
Microbiology Spectrum which made up a deficiency in related field, and suggested Xinjunan may not directly damage the cell structure but interfere with the related pathways by affecting the expression of specific genes. The mode of action of Xinjunan was investigated for the first time, which confirmed the specific high bactericidal effect of Xinjunan on Xoo was achieved by affecting the cellular iron metabolism. This is not only the first time to establish the specific high toxicity/physiochemical effect of Xinjunan to Xanthomonas pathogens, but also the first time to reveal a novel mode of action of this unique bactericide.

MATERIALS AND METHODS
Reagents, strains, and culture conditions. In this study, 95% Xinjunan (Dioctyldiethylenetriamine) C 20 H 45 N 3 , CAS No. 57413-95-3 was provided by Shandong Vicome Greenland Chemical Co., Ltd. Fungal and bacterial strains used in this study were stored in our lab. Generally, fungal strains were cultured in PDA medium at 28°C. Bacterial strains were cultured in NB medium with shaking (180 rpm). For Fe detection or cell viability experiments, Xoo strains were inoculated into MMX media after enrichment culture. CAS medium was prepared by adding 10 mL of CAS solution to 100 mL of Silva Buddenhagen (SB) agar.
EC 50 determination. For fungal pathogens, the EC 50 were determined in vitro by transferring plugs (5 mm in diameter) of mycelium from the leading edge of an actively growing colony to a series of PDA plates containing different concentration of Xinjunan. PDA medium containing DMF (dimethylformamide), the solvent of Xinjunan, was used as control. The diameters (minus the diameter of the inoculation plug) of the colonies were measured after incubation for 2 to 4 days at 25°C in darkness. The growth inhibition as percentage of control was calculated. The median effective Xinjunan concentration (EC 50 ) for the isolates were calculated based on linear regression of colony diameter on log-transformed fungicide concentration. For each concentration, three replications were conducted.
For bacterial strains, the EC 50 values were determined on the basis of growth inhibition, as described earlier (29). A quantity of 50 mL of the bacterial suspension was added to 25 mL of NB medium in 50 mL Erlenmeyer flasks containing diluted concentrations of Xinjunan. NB medium containing DMF (dimethylformamide), the solvent of Xinjunan, was used as control. The optical density of the suspensions in all flasks was measured when the optical density of the suspension in the control flask increased to about 10 8 CFU mL 21 . The log of percentage inhibition based on optical density was regressed on the log of compound concentration, and EC 50 values were calculated.
MIC determination. NA agar medium was mixed with Xinjunan in the culture dish with the final concentration of 0 to 64 mg/L. The bacteria cells were cultured in a flask containing NB medium until late exponential growth stage. Then, 2 mL of the tested bacterial suspensions with different dilution fold (1Â, 3Â, and 9Â) was spotted on the media and incubated at 28°C for further 48 h. NA medium containing DMF was used as control. The MIC is defined as the lowest antimicrobial concentration of Xinjunan that prevents visible growth of the organism investigated after approximately 48 h of incubation.
Growth curves. The growth curves were established as previously described with minor modifications (36). Briefly, Xoo was cultured in NB liquid medium with continuous shaking at 28°C until exponential growth phase. The culture with fresh NB medium was adjusted to a final OD of approximately 0.4. Xinjunan was added to each well to yield final concentrations of EC 50 to 1/16 MIC,1/8 MIC, 1/4 MIC, 1/2 MIC, and 1 MIC, while NB containing 1% DMF (vol/vol) was utilized as a negative control. Subsequently, samples were incubated at 28°C under constant shaking, and cell growth was assessed by measuring the optical density at 600 nm from 0 h to 12 h using a microplate reader (Biotek, Winooski, VT).
TEM analysis. TEM observation was performed as described by Yoshioka-Nishimura et al. (37) with some modifications. TEM was applied to examine the structural changes occurring in Xoo cells treated with Xinjunan. Xoo was cultured at the late exponential phase in 20 mL of NB at 28°C, and the cells were sedimented by centrifugation at 4°C (6,000 g, 10 min) and resuspended in PBS buffer to achieve an OD 600 of 0.4. The cell suspension was treated with Xinjunan at 1/4 MIC, and further cultured for 0.5 h and 4 h at 28°C. The treated and untreated cells were separately sedimented by centrifugation at 4°C (8,000 g, 5 min) and washed three times with PBS buffer. These cells were then fixed with 2.5% (vol/vol) glutaraldehyde overnight at 4°C and were further dehydrated through a graded series of alcohol (30%, 50%, 70%, 90%, and 100%), for 15 min. Thin sections were cut using an ultramicrotome (MT-X; RMC, Jefferson, OH, USA) and subjected to a TEM analysis at an operating voltage of 80 kV.
Determination of DNA content. A total of 1 mL exponential-phase Xoo liquid culture was inoculated to 25 mL NB broth, then Xinjunan was added to make its final concentrations to 0.2 mg/mL, 0.5 mg/mL, 1 mg/mL, 1.5 mg/mL, 2 mg/mL, respectively. Then, 24 h later, 2 mL of the bacteria were centrifuged and dyed at 50°C. DNA of the above samples was extracted using bacterial genome extraction kit (Omega Bio-Tek, Norcross, GA, USA). The DNA contents were determined by Nanodrop Spectrophotometer.
Determination of protein content. Xoo were treated with 0.2 mg/mL, 0.5 mg/mL, 1 mg/mL, 1.5 mg/mL, and 2 mg/mL Xinjunan, respectively, as we mentioned above. The liquid culture of each treatment was centrifuged at 12,000 rpm for 5 min to collect the cell pellets. The sediments were washed with normal saline for three times, and then the suspensions were adjusted to the same turbidity. A total of 2 mL of normal saline was added into 4-m suspension of each treatment, prior to ultrasonic disruption (3 min, 3 S crushing, 3 S interval), and thereafter centrifuged at 10,000 rpm for 30 min. The protein concentration (mg/mL) was measured using the Bradford protein assay kit (Beyotime Biotechnology, Shanghai, China), and bovine serum albumin was used as a reference (38).
Determination of EPS content. The polysaccharide content was determined by sulfuric acid-phenol method. Briefly, Xoo were treated with 0.2 mg/mL, 0.5 mg/mL, 1 mg/mL, 1.5 mg/mL, and 2 mg/mL Xinjunan, respectively, as we mentioned above. A total of 20 mL of bacterial broth was centrifuged at 5,000 rpm for 15 min to collect the supernatant. Three volumes of 95% ice-ethanol (-20°C) were added and mixed thoroughly, and the samples were then left overnight at 4°C. The supernatant was discarded after centrifugation and precipitation was resuspended with 8 mL sterilized water. Next, 1 mL suspension was rapidly mixed with 1 mL of 5% phenol; then, 5 mL of concentrated sulfuric acid was added for a 10 min incubation. The OD 488 value was determined on a spectrophotometer after water bath at 25°C for 15 min. Glucose gradient solutions ranging from 25 to 200 mg/mL were chosen for standard curve determination.
RNA-seq and data analysis. Exponential-phase Xoo liquid culture was inoculated to 25 mL NB broth with continuous shaking at 28°C until the OD reached 0.4. Xinjunan was added to make the final concentration of EC 50 (i.e., 0.33 mg/mL) and continued shaking for 0.5 h. Xoo cells were collected by centrifuging at 4°C and washed with PBS buffer twice. DMF was used as negative control. The samples were frozen in liquid nitrogen immediately and sent to BGI company for RNA extraction and cDNA library construction. Later, QC test samples were sequenced via Illumina HiSeq 2000.
The clean reads were obtained by filtration and then aligned to the reference sequence with SOAP aligner/SOAP2 (39). After the comparison, the distribution and coverage of reads on the reference sequence were counted to determine whether the comparison results passed the second quality control.
Then, a series of follow-up analysis, such as gene expression level, prediction of new transcripts, annotation, and SNP detection, were carried out on the data. The DEGs among the samples were screened out from the results of gene expression. The GO enrichment analysis was applied to describe product characteristics and reaction network of DEGs. All DEGs were mapped to GO terms in the database (40,41). GO terms with corrected P value less than 0.01 were considered significantly enriched by differential expressed genes.
qRT-PCR analysis. Total RNA was extracted using Omega Bacterial RNA Kit (R6950-01). The reverse transcription reaction was performed using AMV First Strand cDNA Synthesis Kit (Sangon Biotech, B532445-0020) according to the manufacturer's instructions. The specific primer pairs for FeoB, pvsA, pvsB, fur, and 16S rRNA genes were used as listed in Table S2. The 16S rRNA gene was used as the reference gene. Real-time PCR was performed using SGExcel FastSYBR Master (Sangon Biotech, B532955-0005). The PCRs were run on CFX96 connect real-time PCR system (Bio-Rad, Hercules, CA).
Siderophore detection. Siderophore production was determined using a Chrome azural S (CAS) agar plate (42,43). CAS medium was prepared by adding 10 mL of CAS solution to 100 mL of SB agar (44). Bacterial colonies were spotted onto a CAS plate and incubated aerobically at 28°C for 2 days. The siderophore production was indicated by the presence of a yellow halo zone around the bacterial colony that was seen readily against the blue-green media background. This halo was formed upon the release of a yellowish CAS dye upon competitive binding of the bacterial siderophore with Fe 31 from the bluegreen CAS-Fe 31 complexes (43).
Bacterial cell viability determination. Cell viability was determined by LIVE/DEAD BacLight Bacterial Viability Kits L7012 (Molecular Probes, Invitrogen) according to the manufacturer's instruction. Briefly, Xoo cells were grown to an OD 600 of 1.0 in NYG medium. Cells were collected by centrifugation at 10,000 rpm for 10 min and washed three times with fresh MMX medium. The bacteria were then inoculated into fresh MMX plus 100 mM FeCl 3 (control), Xinjunan (1/4 MIC), Xinjunan (1/4 MIC) plus 100 mM FeCl 3 , Xinjunan plus 100 mM FeCl 3 and 150 mM DIPY for 30 min and 4 h. Then, the cells were collected by centrifugation at 5,000 rpm at 4°C for 10 min and the pellets were washed and resuspended in 10 mL of PBS buffer.
Next, 3 mL of the dye mixture (combined equal volumes of SYTO9 and PI) was added to each milliliter of the bacteria suspension and incubated at RT in dark for 15 min. A total of 5 mL of the stained bacterial suspension were put onto a glass slide and covered with a coverslip. Imaged cells with filters listed below: Ex/Em 480/550 for SYTO9, Ex/Em 490/635 for PI.
Determination of cellular iron concentration. Cellular iron levels in different strains were determined by inductively coupled plasma-optical emission spectroscopy ICAP 6300 (ICP-OES, Thermo Fisher Scientific, MA USA) as previously described (45). The bacterial cells were grown to an OD 600 of 1.0 in NB medium. Cells were collected by centrifugation at 8,000 rpm and washed three times with fresh MMX medium. The bacteria were then inoculated into fresh MMX plus 100 mM FeCl 3 /Xinjunan/Xinjunan media (OD 600 = 0.6). Three hours later, cells were harvested by centrifugation and washed three times with sterilized PBS buffer (KH 2 PO 4 0.27 g/L, Na 2 HPO 4 1.42 g/L, NaCl 8g/L, KCl 0.2 g/L, pH 7.4). The pelleted cells were oven-dried at 65°C for 72 h and digested with HNO 3 -HClO 4 (4:1, vol/vol). The digestions were transferred to 25 mL volumetric flasks and the final volumes were made up to 25 mL with H 2 O. The iron atom contents were measured by ICP-OES. The samples not treated by HNO 3 -HClO 4 (4:1) were used as parallel controls. The total iron concentration for each sample was calculated by dividing the iron atom value by the number of bacterial cells.
Data analysis. All analyses were conducted using SPSS 19.0 (Statistical Package for the Social Science, SPSS Inc., Chicago, IL). When ANOVAs were significant (P = 0.05), means were separated with Fisher's least significant difference (LSD).

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 0.3 MB.