Biosafety assessment of Acinetobacter strains isolated from the Three Gorges Reservoir region in nematode Caenorhabditis elegans

Acinetobacter has been frequently detected in backwater areas of the Three Gorges Reservoir (TGR) region. We here employed Caenorhabditis elegans to perform biosafety assessment of Acinetobacter strains isolated from backwater area in the TGR region. Among 21 isolates and 5 reference strains of Acinetobacter, exposure to Acinetobacter strains of AC1, AC15, AC18, AC21, A. baumannii ATCC 19606T, A. junii NH88-14, and A. lwoffii DSM 2403T resulted in significant decrease in locomotion behavior and reduction in lifespan of Caenorhabditis elegans. In nematodes, exposure to Acinetobacter strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii and A. lwoffii also resulted in significant reactive oxygen species (ROS) production. Moreover, exposure to Acinetobacter isolates of AC1, AC15, AC18, and AC21 led to significant increase in expressions of both SOD-3::GFP and some antimicrobial genes (lys-1, spp-12, lys-7, dod-6, spp-1, dod-22, lys-8, and/or F55G11.4) in nematodes. The Acinetobacter isolates of AC1, AC15, AC18, and AC21 had different morphological, biochemical, phylogenetical, and virulence gene properties. Our results suggested that exposure risk of some Acinetobacter strains isolated from the TGR region exists for environmental organisms and human health. In addition, C. elegans is useful to assess biosafety of Acinetobacter isolates from the environment.

Effect of exposure to different Acinetobacter strains isolated from the TGR region and reference strains in inducing activation of oxidative stress of nematodes. Oxidative stress is one cellular contributor to toxicity of exposure to toxicants or stresses [25][26][27] . We further employed the ROS production to examine effect of Acinetobacter strains in inducing oxidative stress. Acinetobacter strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii, and A. lwoffii for 24-h resulted in obvious induction of ROS production (Fig. 4A).
Effect of exposure to different Acinetobacter strains isolated from the TGR region on expressions of antimicrobial genes in nematodes. In nematodes, intestine is the important organ to activate innate immune response to pathogen infection 9 . F55G11. 4, dod-22, lys-8, lys-1, spp-12, lys-7, dod-6, and spp-1 are most studied intestinal anti-microbial genes [28][29][30][31][32][33][34] . We next selected these 8 intestinal antimicrobial genes to determine effect of different Acinetobacter strains isolated from the TGR region on innate immune response. The increase in these 8 intestinal antimicrobial genes function to be against pathogen infection and environmental stress [28][29][30][31][32][33][34] . After exposure to Acinetobacter strains of AC1, AC15, AC18, or AC21 for 24-h, expressions of some of these antimicrobial genes could be noticeably increased. Among these 8 antimicrobial genes, exposure to strain AC1 significantly increased the expressions of spp-1, lys -8, lys-7,  Morphological and biochemical properties of Acinetobacter strains of AC1, AC15, AC18, and AC21. For the Acinetobacter strains of AC1, AC15, AC18, and AC21, they did not show obvious difference in morphological properties of cell shape, arrangement of cell, Gram staining, and colony morphology ( Table 1). The Acinetobacter strains of AC1, AC15, AC18, and AC21 also did not exhibit the obvious difference in biochemical properties of hydrothion, phenylalanine, gluconate, oxidase, nitrate reduction, catalase, peptone water, semi-solid agar, glucose, ornithine, raffinose, sorbitol, side calendula, and xylose (Table 1). Different from this, the Acinetobacter strains of AC1 and AC21 showed the negative reactions for the biochemical properties of l-arginine, l-lactic acid, d-fucose, l-histidine, l-malic acid, and d-serine (Table 1). The Acinetobacter strains Effect of exposure to different Acinetobacter strains isolated from the TGR region and reference strains on locomotion behavior in wild-type nematodes. The L4-larvae nematodes were exposed to Acinetobacter for 24-h. Control, unexposed nematodes. Bars represent means ± SD. **P < 0.01 versus control.  www.nature.com/scientificreports/ of AC15 and AC18 exhibited the positive reactions for the biochemical properties of l-arginine, l-lactic acid, d-fucose, l-histidine, l-malic acid, and d-serine (Table 1). Additionally, the Acinetobacter strains of AC1 and AC21 showed the negative reactions for the biochemical properties of glucopeptone water, citrate, and gelation, whereas the Acinetobacter strain of AC15 exhibited the positive reactions for the biochemical properties of glucopeptone water, citrate, and gelation ( Table 1).  Effect of exposure to different Acinetobacter strains isolated from the TGR region on expressions of antimicrobial genes in wild-type nematodes. The L4-larvae nematodes were exposed to Acinetobacters for 24-h. Control, unexposed nematodes. Bars represent means ± SD. **P < 0.01 versus control. www.nature.com/scientificreports/ of virulence genes in tested Acinetobacter strains was different and pathogenic Acinetobacter strains generally had more virulence genes than nonpathogenic strains (Table 2). 10 or more virulence genes were detected from pathogenic strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii and A. lwoffii (Table 2).

Discussion
Acinetobacter has attracted significant attention because it is ubiquitous in nature and commonly found in soil, water and hospital 35 . Many Acinetobacter species can cause serious nosocomial infections in medicine and actively participate the nutrient cycle in the ecosystem 36 . Due to the clinical and ecological importance of Acinetobacter, it is proposed as a model microorganism for environmental microbiological studies, pathogenicity tests, and industrial production of chemicals 18 . Nevertheless, many research areas including biosafety, natural transformation, biodegradation, and important physiological characteristics have been limitedly investigated or neglected. We here performed a biosafety evaluation of Acinetobacter strains isolated from backwater area in the TGR region and 5 reference strains of Acinetobacter species in nematode C. elegans. The high prevalence pathogens exist in the backwater area of the TGR region 7,8 . The reason to carry out the biosafety assessment of Acinetobacter strains is that the Acinetobacter has been one of dominant microorganisms in the TGR region 6 , and Acinetobacter isolated form the water in the TGR arises most frequently in our study. The reasons to use C. elegans are that it is very sensitive to various environmental exposures, and can be employed as an ideal model for the study on the pathogenesis of human pathogens, and the mechanisms in host-microbe interactions 9,13,14,16,37 . More importantly, we previously have systematically performed the biosafety evaluation of water samples from the TGR region in both flood season and quiet season 19,20 . The reasons to select 5 reference strains of A. baumannii, A. lwoffii, A. junii, A. haemolyticus, and A. johnsonii to expose C. elegans are that the genus of Acinetobacter comprises 68 species with validly-published names (https:// apps. szu. cz/ anemec/ Class ifica tion. pdf, May 25, 2021) and these 5 reference speices are important clinical microorganisms 21,38 , and A. baumannii ATCC 19606 T is a model strain of pathogenic bacteria causing nosocomial infection 39 , followed by the non-A. baumannii species A. haemolyticus, A. junii, A. johnsonii, and A. lwofii 21 . Table 1. Biochemical properties of four Acinetobacter strains isolated from the TGR region. "+" stands for positive; "−" stands for negative; "±" stands for not applicable.  www.nature.com/scientificreports/ Our previous studies have suggested that both solid phase and liquid phase could contribute to toxicity induction of surface water sample collected from backwater areas in the TGR region 19,20 . In the liquid phase, the potential toxicants were suggested as the organic pollutants 20 . In this study, using lifespan as the toxicity assessment endpoint, we found that four (AC1, AC15, AC18, and AC21) of the isolated and examined Acinetobacter strains and tree reference strains of A. baumannii, A. junii, and A. lwoffii significantly reduced lifespan (Fig. 2). Using a more sensitive endpoint of locomotion behavior, we also observed the significant decrease in locomotion behavior after exposure to Acinetobacter strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii, or A. lwoffii (Fig. 2), which further confirmed the detected toxic effect of exposure to Acinetobacter strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii, and A. lwoffii on nematodes. These observations suggested that some of the Acinetobacter strains at the backwater area in the TGR region have the exposure risk to environmental organisms and human health. Nevertheless, not all the Acinetobacter strains at the backwater area in the TGR region potentially induced toxicity on environmental organisms. Our data indicated a crucial role of environmental pathogens in contributing to toxicity induction in the solid phase of water sample in backwater area in TGR region.
We further observed the significant ROS production in nematodes exposed to Acinetobacter strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii, or A. lwoffii (Fig. 4A), which suggested the oxidative stress activated by exposure to these Acinetobacter strains. Meanwhile, we also detected the significant increase in SOD-3::GFP expression after exposure to Acinetobacter strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii, or A. lwoffii (Fig. 4B), which further confirmed the oxidative stress activated by exposure to these Acinetobacter strains. These results suggested the close association of the toxic effects of exposure to Acinetobacter strains (AC1, AC15, AC18, AC21, A. baumannii, A. junii, and A. lwoffii) with oxidative stress activation. Nevertheless, we did not detect the decrease in SOD-3::GFP expression after exposure to the above pathogenic Acinetobacter strains. This may be largely due to the short exposure duration (24-h) for these pathogenic Acinetobacter strains. Usually, long-term exposure to toxicants at high concentrations causes decrease in SOD-3::GFP expression 9 . Exposure to nanopolystyrene (1-10 μg/L) caused increase in SOD-3::GFP expression, whereas exposure to nanopolystyrene (1000 μg/L) resulted in decrease in SOD-3::GFP expression 40 .
In nematodes, we further found that exposure to Acinetobacter strains of AC1, AC15, AC18, and AC21 induced increase in expressions of some antimicrobial genes (spp-1, dod-22, lys-8, lys-7, spp-12, dod-6, lys-1, and/or F55G11.4) (Fig. 5). Meanwhile, a pronounced increase in Acinetobacter colony-forming unit (CFU) was observed in nematodes infected with AC1, AC15, AC18, and AC21 (Fig. S1). The increase in these antimicrobial genes mediated a protective response to pathogen infection and environmental toxicants 9,28-34 . These antimicrobial genes can be expressed in the intestine (https:// wormb ase. org). The reason to select these intestinal antimicrobial genes is that the ROS production is mainly activated in the intestine 25 . Similarly, we also did not observe the suppression in expressions of these antimicrobial genes in nematodes exposed to Acinetobacter strains of AC1, AC15, AC18, or AC21, which is also largely due to the performed short exposure duration (24-h) in nematodes. Moreover, we found that exposure to Acinetobacter strains of AC1, AC15, AC18, and AC21 induced the different dysregulation of examined antimicrobial genes (Fig. 5). Exposure to AC1 could cause the increase in expressions of all 8 examined antimicrobial genes, and exposure to AC21 resulted in the increase in expressions of 7 examined antimicrobial genes (Fig. 5). In contrast, exposure to AC15 could cause the increase in expressions of only 4 examined antimicrobial genes, and exposure to AC18 could result in the increase in expressions of only 3 examined antimicrobial genes (Fig. 5). These results implied that Acinetobacter strains of AC1 and AC21 might cause the more severe toxicity at least at some aspects than Acinetobacter strains of AC15 and AC18.
In this study, we provide some lines of evidence to show the important value of C. elegans for assessing biosafety of Acinetobacter strains isolated from the TGR region. Nevertheless, C. elegans only has simple developmental structures, and dose not have some organs (such as heart, liver, lung, and kidney) observed in mammals. Table 2. The presence of main virulence genes in pathogenic and nonpathogenic Acinetobacter strains. "+" stands for positive; "−" stands for negative.  www.nature.com/scientificreports/ Therefore, the further biosafety assessment experiments in mammals for the identified four Acinetobacter strains are still needed. We also examined morphological and biochemical properties of Acinetobacter strains of AC1, AC15, AC18, and AC21. However, we did not observe the obvious difference in morphological properties of cell shape, arrangement of cell, Gram staining, and colony morphology among the examined Acinetobacter strains of AC1, AC15, AC18, and AC21 (Table 1). In contrast, the observed difference in toxicity of Acinetobacter strains of AC1, AC15, AC18, and AC21 on nematodes might be related to the difference in some biochemical properties among the examined Acinetobacter strains of AC1, AC15, AC18, and AC21. For example, we observed the obvious difference in biochemical properties of l-arginine, l-lactic acid, d-fucose, l-histidine, l-malic acid, and d-serine in the Acinetobacter strains of AC1 and AC21 from those in the Acinetobacter strains of AC15 and AC18 (Table 1). To clarify if they share virulence factors that better induce to the nematode intestinal antimicrobial response, 14 main virulence genes (Table S4) of pathogenic strains of AC1, AC15, AC18, AC21, A. baumannii, A. junii and A. lwoffii and nonpathogenic strains of AC2, AC12, AC14, AC17, A. haemolyticus, and A. johnsonii were detected by PCR. The results showed that pathogenic Acinetobacter strains generally had more virulence genes than nonpathogenic strains (Table 2), and AC1 and AC21, AC15 and AC18 shared more of the same virulence genes, but nonpathogenic strains of AC2 and AC12 also had 11 and 14 virulence genes, respectively. Nevertheless, the exact underlying mechanism still needs the further careful examination.
Together, we performed a biosafety assessment of Acinetobacter strains isolated from backwater area in TGR region in nematodes. Among the isolated Acinetobacter strains, we identified four Acinetobacter strains with the potential to cause toxic effects on nematodes, such as the reduction in lifespan and the decrease in locomotion behavior. The observed toxic effects of Acinetobacter strains were associated with activation of oxidative stress. Moreover, exposure to toxic Acinetobacter strains caused the increase in some antimicrobial genes, suggesting the activation of innate immune response of animals against the Acinetobacter exposure. Considering the fact that we know little about the environmental Acinetobacter pathogens in the TGR region, our data provide important suggestion for exposure risk of certain Acinetobacter strains in the TGR region to environmental animals and human health. Our data has further implied that, after the long-term exposure, the Acinetobacter pathogens are potentially enriched in intestine and cause toxic effects by affecting immune response in environmental animals and human. In the future, we will further identify virulence and resistance factors and perform the sequencing for the identified four Acinetobacter strains isolated from the TGR region.

Methods
Water sampling. The water sample was collected in backwater area (N108° 23′ 25″, E30° 47′ 45″) in Wanzhou, Chongqing in the flood season 20 . The reason to select this season is that the bacterioplankton community is generally higher in this season than that in the impoundment season 6 . The detailed properties of collected surface water sample have been described previously 20 . Water sample was collected and stored as described 41 . In brief, the equal volumes (10 L) were collected from the depths of 0.5, 5, 10 m in the backwater area site. Water samples were used for the isolation of Acinetobacter after mixing fully in the sterile bucket, and water samples were stored at 0 °C after collection.
Acinetobacter isolation, identification, and preservation. The mixed water sample was diluted serially (1, 10 −1 , and 10 −2 ), inoculated into LB medium, and incubated at 37 ± 0.5 °C for 24 h. Subculture and purification of bacterial colonies were carried out by the streak plate method. For the purified bacterial isolates, genomic DNA of different bacterial isolates was extracted using bacterial genomic DNA extraction kit (TIAN-GEN, Beijing, China) according to the manufacturer's instructions. The complete bacterial 16S rRNA gene was amplified with the primer set 27 F and 1492R. PCR products were visualized using 1% agarose gels stained with ethidium bromide. Positive amplicons were quantified using a PicoGreen dsDNA Assay kit (Invitrogen, CA, USA). Purified products were sequenced and analyzed by Magigen (Guangzhou Magigen Biotechnology Co., Ltd., China). Phylogenetic tree was constructed using the Mega 5.0 program using the neighbor-joining (N-J) method with a 1000-bootstrap.
All identified Acinetobacter strains were preserved by freeze drying 42 . The exponential phase cells of Acinetobacter strains grown in LB medium for 18 h were suspended in aseptic no-fat skimmed milk with an initial cell concentration of 10 8 -10 9 CFU/mL. The bacterial mixture within ampoules vials was frozen at − 20 °C for 2 h, followed by − 80 °C for 12 h. After that, they were loaded onto the freeze dryer. Both primary drying and secondary drying for 25 h after the freezing were performed. The freeze-dried products were packaged in blister packs and stored in the refrigerator at − 80 °C.
When needed, freeze-dried powders were diluted with sterilized water, and then the suspensions were streakinoculated onto a LB medium using an inoculation loop. A single colony was inoculated into sterilized LB broth and the bacteria grew to the log phase in a constant temperature oscillator at 37 °C for the use 43,44 . Analysis of Acinetobacter properties. Different Acinetobacter strains inoculated on broth agar medium were incubated for 24 h at 37 °C 45 . Primary identification and characterization of different Acinetobacter strains were performed to determine cell shape, arrangement of cell, gram staining, and colony morphology using UVsolo 2 touch (Analytik Jena AG, Germany) 46  Maintenance of C. elegans. CF1553/muIs84[SOD-3:GFP] and wild-type N2 were used. Normal nematode growth media (NGM) plates were used to maintain nematodes 48 . To prepare synchronized L4-larvae, gravid worms were first treated with bleaching solution (0.45 M NaOH and 2% HOCl) 49 . The released eggs were let to further develop into the L4-larvae population.

Reference strains of
Acinetobacter pathogenesis assay. The L4-larvae population was exposed to different Acinetobacter strains. Different Acinetobacter strains were seeded on modified NGM containing 0.35% peptone. Exposure to different Acinetobacter strains was started by transferring nematodes onto each assay plate by adding 60 animals to each assay NGM plate. Full-lawn assay plate was used for Acinetobacter pathogenesis assay as described 50 . That is, the surface of assay NGM plate was all seeded Acinetobacter strains. The aim of using full lawn assay was to exclude the possibility of effect from the avoidance behavior of nematodes to Acinetobacter strains.
Acinetobacter CFU analysis. The method was performed basically as described 51  Lifespan assay. After exposure of L4-larvae nematodes to different Acinetobacter strains for 24-h, the survival of worms was counted every day at 20 °C 52 . If no response was observed after prodding using platinum wire, the worms were considered as dead. The animals were transferred daily during the first 7-day. For the lifespan assay, 60 animals were examined for each treatment. Three replicates were carried out. We used log-rank test to analyze the lifespan curve data. Survival curves were considered to have significant difference if P values were ≤ 0.01.

Locomotion behavior.
Locomotion behavior reflects the functional state of motor neurons 53 . Body bend and head thrash were selected as the endpoints 54 . After exposure, the worms were first washed with M9 buffer. After that, assuming that animals traveled along x axis, a body bend is defined as a change of bending direction at the mid-body. A head thrash is defined as a change of posterior bulb direction along y axis. For each treatment, 40 animals were analyzed.
Activation of oxidative stress. Production of reactive oxygen species (ROS) reflects the activation of oxidative stress 55 . The method was performed as described 56 . After the exposure to different Acinetobacter strains, the animals were labeled for 3 h using CM-H 2 DCFDA (1 µM). After that, the animals were observed at 488 nm (excitation wavelength)/510 nm (emission filter) under a laser scanning confocal microscope. Using Image J software, we semi-quantified intestinal fluorescence intensity in comparison to intestinal autofluorescence. For each treatment, 50 animals were examined. In nematodes, sod-3 encodes mitochondrial Mn-SOD 9 . Using Image J software, fluorescence intensity of SOD-3::GFP signals in the intestine was semi-quantified. For each treatment, 50 animals were examined.
Quantitative real-time polymerase chain reaction (qRT-PCR). The total RNAs of control and exposed nematodes were extracted using Trizol. Using a spectrophotometer, concentration and purity of the obtained RNAs were determined. We performed the reverse transcriptase reaction with Mastercycler gradient PCR system for cDNA synthesis. With the aid of SYBR Green qRT-PCR master mix, transcriptional expression of spp-1, lys-8, lys-7, lys-1, spp-12, dod-6, dod-22, and F55G11.4 were determined in real-time PCR system. The reference gene was tba-1. Three biological replicates were carried out. Primer information is provided in Table S3.
Polymerase chain reaction (PCR) of virulence genes. Primers were designed according to the published sequences of virulence genes of Acinetobacter on GenBank. Primer information was provided in Table S4. All experimental strains were inoculated onto beef extract agar and cultured at 37 ℃ for 24 h. A single colony was selected and inoculated in beef extract broth at 37 ℃ for 12 h, and the bacterial culture was used as the template. PCR was undertaken in a final volume of 25 μL using the PCR kit (Sangong Bioengineering (Shanghai) Co., Ltd.) with 1μL of each primer and 1 μL of the template. The thermal cycling parameters were 30 s at 98 °C for, followed 35 cycles of 5 s at 98 °C, 5 s at 59 °C and 60 s at 72 °C. 5μL PCR products were analyzed on agarose 1.4% (w/v) gels.

Statistical analysis.
Statistical analysis was carried out using SPSS Statistics 19.0 Software (SPSS Inc., USA.). Probability level of 0.01 was considered statistically significant. Using one-way analysis of variance (ANOVA), the differences between groups were tested.