Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2022 Aug 25;24(11):5261–5276. doi: 10.1111/1462-2920.16165

Antidepressants promote the spread of antibiotic resistance via horizontally conjugative gene transfer

Pengbo Ding 1, Ji Lu 1, Yue Wang 1, Mark A Schembri 2, Jianhua Guo 1,✉
PMCID: PMC9804927  PMID: 36054646

Abstract

Antibiotic resistance is a global concern threatening public health. Horizontal gene transfer (HGT) between bacterial species contributes greatly to the dissemination of antibiotic resistance. Conjugation is one of the major HGT pathways responsible for the spread of antibiotic resistance genes (ARGs). Antidepressant drugs are commonly prescribed antipsychotics for major depressive disorders and are frequently detected in aquatic environments. However, little is known about how antidepressants stress bacteria and whether such effect can promote conjugation. Here, we report that commonly prescribed antidepressants, sertraline, duloxetine, fluoxetine, and bupropion, can promote the conjugative transfer of plasmid‐borne multidrug resistance genes carried by environmentally and clinically relevant plasmids. Noteworthy, the transfer of plasmids across bacterial genera is significantly enhanced by antidepressants at clinically relevant concentrations. We also reveal the underlying mechanisms of enhanced conjugative transfer by employing flow cytometric analysis, genome‐wide RNA sequencing and proteomic analysis. Antidepressants induce the production of reactive oxygen species and the SOS response, increase cell membrane permeability, and upregulate the expression of conjugation relevant genes. Given the contribution of HGT in the dissemination of ARGs, our findings highlight the importance of prudent prescription of antidepressants and to the potential connection between antidepressants and increasing antibiotic resistance.

INTRODUCTION

The emergence and spread of antibiotic resistance pose major threats to public health, killing approximately 4.95 million people worldwide annually. Antibiotic resistance occurs intrinsically in microbial communities regardless of human use of antibiotics (Allen et al., 2010). However, the intensive application of antibiotics for medical, veterinary, and agricultural use has enhanced the problem of antibiotic resistance (Allen et al., 2010). Horizontal gene transfer (HGT) represents a major driver for the spread of antibiotic resistance genes (ARGs) (Davison, 1999; Thomas & Nielsen, 2005). Plasmid conjugation, the dominant HGT pathway in bacteria (von Wintersdorff et al., 2016), relies on direct physical contact between donor and recipient cells and is mediated via a pilus bridge generated by the plasmid‐containing donor cell (Smillie et al., 2010). Conjugative transfer can occur within individual species and also across different bacterial species and genera, leading to increased rates of antibiotic resistance among related organisms.

It is well recognized that the overuse and misuse of antibiotics are a key impetus for the spread of ARGs (Kümmerer, 2001). However, the role of non‐antibiotic pharmaceuticals in this regard has been largely overlooked. Antidepressants are commonly prescribed as long‐term treatments for major depressive disorders (MDD), which is the most common mood disorder (Kessler et al., 2012; Scheuing et al., 2015). The World Health Organization (WHO) estimates that 4.4% of the global population lives with depression (World Health Organization, 2017), and the number of people suffering from such conditions increased by 18.4% from 2005 to 2015 (Vos et al., 2016), with further increases due to COVID‐19 well appreciated (Adu et al., 2021). The increasing need for drugs to treat MDD has stimulated the development and application of antidepressant drugs to an unprecedented level (Pereira & Hiroaki‐Sato, 2018). The number of prescriptions for all antidepressants was 120 million in 1998, a number that tripled in one decade (Pereira & Hiroaki‐Sato, 2018). Antidepressants have a broad range of concentrations in clinical and environmental settings, from nanogram per litre to milligram per litre. Specifically, the concentration range of clinically relevant concentrations is mainly from 0.1 to 100 mg/L after the consumption of antidepressants. The lower clinical concentrations (0.1 and 1.0 mg/L levels) primarily present in the plasma or urine, for example, the peak plasma concentration of bupropion is around 0.3 mg/L (Coles & Kharasch, 2007) and the urine concentration of fluoxetine has a range from 0.52 to 1.46 mg/L (Unceta et al., 2008). The medium concentrations (5.0 and 10.0 mg/L) are not only applied in drug screening (Ejim et al., 2011; Maier et al., 2018) but also can be found in colon, for example, escitalopram arrives in colon with a concentration range of 3.0–51.0 mg/L (McGovern et al., 2019). The high concentrations (50.0 and 100.0 mg/L) are mainly detected in colon, for example, fluoxetine arrives in colon with a concentration range of 25.0–170.0 mg/L (McGovern et al., 2019). In environmental settings, most of the antidepressant concentrations detected are at nanogram per litre levels in surface water, such as in creeks and river (Writer, Antweiler, et al., 2013; Writer, Ferrer, et al., 2013). However, it should be noted that antidepressants concentrations can reach higher than 10 μg/L level in certain conditions, such as in the decentralized wastewater treatment systems or psychiatric hospitals' wastewater (Mole & Brooks, 2019).

Although the effectiveness of antidepressants for MDD treatment could be closely related to their antimicrobial effects on gut bacteria (Macedo et al., 2017; Zhang et al., 2021), it is still unclear whether such antimicrobial stress can promote the spread of ARGs via HGT. Our previous study also demonstrated antidepressant fluoxetine induces mutagenesis conferring multiple antibiotic resistance in a common opportunistic pathogen Escherichia coli (Jin et al., 2018). Based on the above findings, we hypothesized that commonly prescribed antidepressants may promote the conjugative transfer of ARGs between different bacteria.

In this study, we systematically investigated whether antidepressants are capable of promoting cross‐genera conjugative transfer of ARGs harboured by environmentally and clinically sourced plasmids. To this end, we selected six commonly prescribed antidepressants, including sertraline, fluoxetine, and escitalopram (selective serotonin reuptake inhibitors); duloxetine (a serotonin‐norepinephrine reuptake inhibitor); bupropion (a norepinephrine‐dopamine reuptake inhibitor); and agomelatine (an atypical antidepressant). We employed two bacterial mating systems to study the conjugative transfer of an environmentally abundant IncP plasmid RP4 and a clinically sourced superbug IncC plasmid pMS6198A that carries multiple ARGs including bla NDM−1 (Hancock et al., 2021). In addition, our results demonstrated that some of these antidepressants could stimulate the conjugative transfer of both plasmids across different genera. The underlying mechanisms of enhanced conjugation were elucidated, revealing that antidepressants can trigger the overproduction of reactive oxygen species (ROS), the SOS response, increased cell membrane permeability, and the upregulation of essential genes involving conjugation. This is the first study to illustrate how a group of antidepressant drugs can promote the cross‐genera horizontal transfer of environmentally and clinically relevant resistance plasmids. The effect of antidepressants on the spread of antibiotic resistance emphasizes the need for careful oversight for the prescription of such drugs and evaluation of the contribution of these drugs to increasing antibiotic resistance.

EXPERIMENTAL PROCEDURES

Bacterial strains, culture media and conditions

Conjugation assays included two mating systems, which shared the same recipient strain but different donor strains. The donor strains were common opportunistic pathogens in the gut microbiota, E. coli K‐12 LE392 harbouring plasmid RP4 and E. coli K‐12 MG1655 carrying pMS6198A. The plasmid RP4 (60,099 bp) possesses resistance against ampicillin (Amp), kanamycin (Kan), and tetracycline (Tet) (Wang et al., 2019). The clinically sourced plasmid pMS6198A (137,565 bp) was isolated from uropathogenic E. coli (UPEC) strain MS6198 and carries multiple antibiotic resistance genes including bla CMY−6, aacA4, rmtC, bla NDM−1, ble MBL, and sul1 (Hancock et al., 2017). The recipient strain Pseudomonas alloputida KT2440 intrinsically has a chloramphenicol resistance more than 10 times stronger than that of the donor strain (Wang et al., 2019). Appropriate antibiotics were supplemented to Luria‐Bertani (LB) broth for overnight culturing. Details of bacterial resistance and culture conditions are described in Supporting Information (SI) Text S1.

Conjugation assays

Two mating systems were used to investigate whether antidepressants affect the conjugative transfer of different plasmids. For the conjugation assay testing, the transfer of an environmental plasmid, E. coli K‐12 LE392 carrying plasmid RP4 was selected as the donor strain, while in the conjugation assay testing the conjugative transfer of a clinically sourced plasmid, E. coli K‐12 MG1655 harbouring pMS6198A was used as the donor. P. alloputida KT2440 was used as the recipient in both mating systems. Conjugation assays were conducted with the presence of six commonly prescribed antidepressants; sertraline, duloxetine, fluoxetine, bupropion, escitalopram, and agomelatine. Antidepressants often have elimination half‐lives up to days, and more than 10% are excreted in an unmodified form (DeVane, 1999). Typical urinary concentrations of antidepressants are at milligrammes per litre levels (Coles & Kharasch, 2007; Unceta et al., 2008). The final concentrations of antidepressant drugs applied in mating assays were considered clinically relevant concentrations (Coles & Kharasch, 2007; DeVane, 1999; Unceta et al., 2008), as well as sub‐minimum inhibitory concentrations (sub‐MICs) of the donor and recipient strains. Specifically, 0.1, 1.0, 10.0, 50.0, and 100.0 mg/L were employed in the conjugation assays based on plasmid RP4. As for the conjugation assays based on clinical plasmid pMS6198A, drug concentrations consisted of 0.1, 1.0, 5.0, 10.0, 50.0, and 100.0 mg/L, among which 1.0, 5.0, and 10.0 mg/L are clinically relevant concentrations applied for high‐throughput drug screens (Ejim et al., 2011; Maier et al., 2018).

In both mating systems, the six antidepressants were separately added to the 1 ml mixture (108 cfu/ml), comprising the donor and recipient bacteria at 1:1 ratio. After adding antidepressant drugs, bacteria were vortexed briefly (3 s) to homogenize the mating systems. The mating system mimicking the transfer of environmental and clinical plasmid was established in phosphate‐buffered saline (PBS, pH = 7.2) and LB broth (pH = 7.0), respectively. Both environmental and clinical setups were conducted under static conditions, at 25°C for 8 h and 37°C for 4 h, respectively. At the conclusion of the mating incubation, the mixture was vortexed and then inoculated onto selective LB agar supplemented with appropriate antibiotics, differentiating transconjugants from recipients. In order to eliminate the effect of spontaneous mutation on the conjugative transfer ratio, the donor and recipient were each spread onto the selective plate for transconjugants, and the donor was also spread onto the recipient selection plate as a negative control. Detailed information of selective plates is described in SI Text S2. All the plates were incubated at 30°C for 48 h; colony numbers of the transconjugants and recipients were then counted.

Additionally, two sets of conjugation assays were conducted to validate the contribution of ROS in enhanced conjugation: (i) a set performed in the presence of a ROS scavenger—thiourea that quenching free radicals (Kohanski et al., 2010; Prasad & Mishra, 2017), and (ii) a set conducted in an anaerobic chamber precluding oxygen. Both two sets of conjugation assays were conducted with the presence of each of the six antidepressants, respectively, at designated final concentrations: sertraline, duloxetine, and fluoxetine at 10.0 mg/L; bupropion, escitalopram, and agomelatine at 100.0 mg/L. The transfer ratio is calculated by dividing the transconjugant number by the recipient number. At least three independent biological experiments were performed.

Determining inhibitory concentrations of the donor, recipient and transconjugant against antibiotics

The concentration of antibiotics at which the growth of 90% bacteria was inhibited was recorded as IC90. IC90 of the donor and recipients against antibiotics (ampicillin, kanamycin, tetracycline, and chloramphenicol) was conducted in 96‐well microtiter plates (Corning, USA). At least three independent biological experiments were performed. Details of IC90 assay are described in SI Text S3.

Confirmation of transconjugants

M‐Endo Agar manufactured by Difco™ was applied to confirm the genera of colonies from transconjugant selective plates. The identification is based on the colour of colony, transconjugant colonies grew on the m‐Endo agar plates was confirmed white. Plasmid extraction, electrophoresis, and long‐amplicon PCR on plasmid bla gene were conducted following enrichment culturing of confirmed colonies. Details of transconjugant confirmation are described in SI Text S4.

Quantification of ROS and cell membrane permeability

In order to reveal the underlying mechanisms of enhanced conjugation, to determine whether the elevated bacterial ROS production and increased cell membrane permeability act as the impetus. The ROS production and cell membrane permeability were measured by flow cytometry (CytoFLEX S, Beckman Coulter, USA). The cellular ROS detection assay kit (abcam®, UK) using 2′,7′‐dichlorofluorescin diacetate (DCFDA) was applied following the manufacture's protocol. Cell membrane permeability was assessed by measuring the uptake of 2 mM propidium iodide (PI, Life Technologies, USA) as described in previous studies (Luo et al., 2014; Zhang et al., 2017). Detailed methods of cellular ROS detection and cell membrane permeability measurement are described in SI Text S5. The fold change of ROS production or cell membrane permeability was calculated by dividing the treated sample with the corresponding drug‐free control. At least three independent biological experiments were performed.

Live and dead staining assays

The number of live, damaged, and dead cells in antidepressants treated bacteria was quantified by flow cytometry (CytoFLEX S, Beckman Coulter, USA) using SYTO® 9 and PI from LIVE/DEAD® BacLight™ Bacterial Viability Kit (Invitrogen, USA). Untreated and heat‐treated (80°C for 2 h) bacteria were applied as controls for live and dead bacteria, respectively. Details of live and dead staining assay are described in SI Text S6.

RNA extraction and genome‐wide RNA sequencing

In order to check the transcriptional changes attributed to enhanced conjugation. Whole‐genome RNA sequencing was used to determine gene expression levels during conjugation experiments (plus/minus antidepressants). RNA samples (sample preparation in SI Text S7) were submitted to Novogen Co. (Hong Kong) for cDNA library construction and Illumina paired‐end sequencing (HiSeq 2000, Illumina Inc., USA). Bioinformatic analysis was performed as previously described (Gao et al., 2016). Gene expression was quantified as fragments per kilobase of a gene per million mapped reads (FPKM). Therefore, the fold change of gene expression was calculated by diving the FPKM of the treated sample with that of the control. RNA samples for each mating system were prepared as biological triplicates.

Proteomic analysis

Proteomic analysis was conducted to reveal the relative protein abundance levels during the conjugation process. Digested protein (sample preparation in SI Text S8) was then analysed by liquid chromatography (Ultimate® 3000 RSLCnano system, Thermo Scientific™)‐tandem mass spectrometry (Q‐Exactive™ H‐X Hybrid Quadrupole‐Orbitrap™ mass spectrometer, Thermo Scientific™) (LC–MS/MS). After protein sequencing, raw data were processed by Thermo Proteome Discoverer (version 2.2.0.388) against the E. coli MG1655 and P. alloputida KT2440 database (Uniport, accessed on 12 July 2019). The abundance ratio of each protein was calculated by dividing the abundance of the drug treated sample by that of the drug‐free control sample. Proteins with false discovery rate (FDR), q value, less than 0.01 were considered to have significant expression differences.

Statistical analysis

Data are expressed as mean ± SD (standard deviation). SPSS Statistics 24.0 (SPSS, Chicago, USA) for windows was used for data analysis. Adjusted p value (Benjamini–Hochberg method) was calculated to assess significant differences (Benjamini & Hochberg, 1995). A difference with adjusted p value ≤ 0.05 was considered significant.

RESULTS

Antidepressants promote the transfer of environmental and clinical plasmids across bacterial genera

To investigate the effects of antidepressants on plasmid conjugative transfer, we conducted intergenera conjugation assays involving the plasmids RP4 and pMS6198A under the exposure of six commonly prescribed antidepressants at previously described clinically relevant concentrations (exemplified as Figure 1A) (Mole & Brooks, 2019). The conjugative transfer of plasmid RP4 was significantly enhanced in a dose‐dependent manner by five (sertraline, duloxetine, fluoxetine, bupropion, and escitalopram) out of six antidepressants tested (p adj < 0.05; Figure 1B, raw data in Figure S1). Noteworthy, increased transfer ratio is based on an increased number of transconjugant (Figure S2), rather than an increment only attributed to a reduced recipient number. Sertraline, duloxetine, and fluoxetine exhibited the strongest efficacy promoting conjugation (p adj < 0.01). Lower concentration (1.0 mg/L) of sertraline (p adj < 0.01) and duloxetine (p adj < 0.05) also capable of promoting conjugation. Increased number of dead and damaged cells upon exposure to medium concentration (10.0 mg/L) of duloxetine, fluoxetine, and sertraline was observed in the live and dead staining assays (Figures S3 and S4), while the bactericidal effect of these antidepressants at high concentrations (50.0 and 100.0 mg/L) was directly shown by no viable colony could grow on the selective plates. Particularly, bupropion significantly (p adj < 0.01) stimulated conjugation at a broad concentration range. Also, escitalopram significantly (p adj < 0.01) enhanced the transfer ratio of plasmid RP4. Agomelatine had no effect on plasmid conjugation.

FIGURE 1.

FIGURE 1

Open in a new tab

Intergenera conjugative transfer of plasmid RP4 and pMS6198A induced by different concentrations of antidepressants. (A) Schematic of conjugation assay. (B) Fold changes of plasmid RP4 conjugation ratio upon the exposure to different concentrations of antidepressant drugs, compared to drug‐free solvents. (C) Fold changes of plasmid pMS6198A conjugation ratio upon the exposure to different concentrations of antidepressant drugs, compared to drug‐free solvents. The transfer ratios of plasmid RP4 and pMS6198A under the exposure of antidepressant drugs are presented in Figures S1 and S5, respectively. (D) IC90 verification of transconjugants' antibiotic resistance offered by the donor plasmid (Amp, Kan, Tet) and recipient chromosome (Chl). Significant differences between antidepressant‐treated groups and the control group were calculated using independent‐sample t test, p values were then corrected by the ‘Benjamini–Hochberg’ method as p, *p adj < 0.05, and **p adj < 0.01.

In the conjugation assays involving clinically isolated plasmid pMS6198A, five out of six antidepressants, except for agomelatine, significantly enhanced conjugative transfer across genera (p adj < 0.05; Figure 1C, raw data in Figure S5). Broad concentration range of fluoxetine (p adj < 0.05), sertraline (p adj < 0.01), and bupropion (p adj < 0.01) enhanced pMS6198A transfer. Again, the antibacterial effects of sertraline, fluoxetine, and duloxetine at high concentrations also inhibited the conjugative transfer of plasmid pMS6198A (Figure 1C).

The successful transfer of plasmids into transconjugants was confirmed by m‐Endo agar, which is used for coliform enumeration. Based on the colour of colonies, transconjugant and recipient colonies (white) were readily distinguished from donor colonies (metal luste). Electrophoresis further confirmed transconjugants harbour plasmids with similar size to that of the donor (Figure S6a), as well as PCR amplification of the plasmid carried β‐lactamase gene (Figure S6b). The MICs of the donor, recipient, and transconjugants were determined. As expected, transconjugant cells possessed a resistance profile consistent with the recipient and plasmid combination (Figure 1d). Taken together, these results verify that plasmid RP4 had been successfully transferred from donor to recipient cells. Collectively, five commonly prescribed antidepressants, including sertraline, duloxetine, fluoxetine, bupropion, and escitalopram significantly facilitated the conjugative transfer of both plasmids across strains from different genera. In contrast, agomelatine did not affect the transfer of either plasmid.

Enhanced conjugation ratio and reduced bacteria viability correlate with the increased ROS production

Previous studies have revealed that oxidative stress can promote conjugation (Wang et al., 2019; Zhang et al., 2017), but overstress may affect the viability of bacteria (Zhang et al., 2019). To investigate whether enhanced conjugation and reduced viable cells are associated with ROS overproduction, flow cytometry was employed to measure the production of these highly reactive molecules in donor and recipient cells. ROS production in the donors and recipients significantly increased with the dosage of sertraline, duloxetine, and fluoxetine (p adj < 0.01; Figure 2A). A dose‐dependent trend of ROS response was observed when stimulated by these antidepressants. Nevertheless, extreme ROS overproduction in bacteria triggered by high concentrations of these three antidepressants led to severe cell death and damage; and thus, non‐detectable transfer ratios in conjugation assays. In contrast, escitalopram has relatively milder effect in this regard (p adj < 0.01), which was consistent with the small increment on transfer ratio. Interestingly, bupropion promoted conjugation (p adj < 0.01) without drastic ROS overproduction. Finally, the ROS level in the recipient remained unchanged after agomelatine treatment. Combining the above‐mentioned ROS changes with the elevated conjugation ratio of certain antidepressant‐treated mating groups, it seems the relatively milder ROS response in recipients, compared to that in the donor, could have a major influence on conjugation.

FIGURE 2.

FIGURE 2

Open in a new tab

Effects of antidepressant drugs on reactive oxygen species (ROS) production in the donor (E. coli K‐12 LE392) and recipient (P. alloputida KT2440) strains. (A) Fold changes of fluorescence intensity related to ROS levels upon the exposure to antidepressants under aerobic and anaerobic conditions (broader concentration range in Figure S7). (B) Fold changes of plasmid RP4 transfer ratio under anaerobic conditions or dosed with ROS scavenger thiourea. (C) Fold changes of transcriptional expression of key genes associated with ROS production in the donor and recipient bacteria. (D) Fold changes of translational expression of key genes related to ROS production in the donor and recipient bacteria. Significant differences between antidepressant‐treated groups and the control group were calculated using independent‐sample t test, p values were then corrected by the ‘Benjamini–Hochberg’ method as p, *p adj < 0.05, and **p adj < 0.01.

In the conjugation assays validating the role of ROS, the ratio of conjugative transfer decreased significantly to the control level in ROS eliminated systems when treated by duloxetine, fluoxetine, and sertraline (p adj > 0.05; Figure 2B). Bupropion was the exception, as the conjugation ratio did not drop back to the control level after drug treatment, and the increment was significant (p adj < 0.01). Additionally, the donor and recipient strains treated with antidepressants under anaerobic conditions did not exhibit a significant increase in ROS production (p adj > 0.05), compared with the control group (Figure 2A). Therefore, the antidepressants tested in this study can be divided into three groups: (i) antidepressants that promote conjugation via ROS overproduction, including sertraline, duloxetine, fluoxetine, and escitalopram; (ii) antidepressants that promote conjugation without inducing ROS overproduction, that is, bupropion; and (iii) antidepressant that does not promote conjugation, that is, agomelatine.

Cellular damage attributed to ROS overproduction potentially contributes to enhanced conjugative transfer, live and dead staining assays assessed whether antidepressants caused cell death associated with ROS overproduction. Overall, Group I and II antidepressants significantly increased the proportion of damaged and dead bacteria under aerobic conditions (p adj < 0.05; Figures S3 and S4). Sertraline exhibited the strongest antibacterial effect against P. alloputida with significant (p adj < 0.01) damage and fatality. Generally, the aerobic biocidal efficacy of antidepressants is consistent with their profile on promoting conjugation and ROS overproduction. In contrast, the biocidal effect of antidepressants was considerably retarded by anaerobic conditions. Therefore, the antibiotic‐like role of antidepressants correlates highly with ROS overproduction.

Transcriptional expression of bacteria revealed their response to the oxidative stress posed by antidepressants (Figure 2C). In the donor E. coli, exposure to sertraline, duloxetine, and fluoxetine led to the upregulation of genes that participate in antioxidant defence mechanisms against oxidative stress, including those encoding alkyl hydroperoxide reductase (ahpC, ahpF), hydroperoxidase I (katG), and superoxide dismutase (sodA, sodC). Superoxide dismutase enzymes (SODs) utilize a redox‐active metal to transform superoxide into hydrogen peroxide and oxygen (Broxton & Culotta, 2016). Accumulated hydrogen peroxide is then removed by catalase and peroxidases, such as hydroperoxidase I, encoded by katG (Loewen et al., 1985). In addition, the expression of alkyl hydroperoxide reductase genes (ahpC and ahpF) increased, the product of which acts as a ROS scavenger of endogenous hydrogen peroxide (Seaver & Imlay, 2001). Generally, increased expression of these antioxidant defence genes could protect the bacteria from ROS damage (Jung & Kim, 2003). Indeed, the transcriptional response was consistent with enhanced ROS production observed in sertraline‐, duloxetine‐, and fluoxetine‐treated groups. Interestingly, bupropion stimulated transcriptional expression of antioxidant defence genes; yet decreased ROS production was observed in the donor. Changes in gene expression were also detected in the recipient P. alloputida; antioxidant defence genes were upregulated during the exposure to Group I antidepressants. This included genes encoding alkyl hydroperoxide reductase (ahpC), superoxide dismutase (sodB), redox‐sensitive transcriptional activator (soxR), and thioredoxin reductase (trxB).

The analysis on protein abundance further confirmed that antidepressants promoted translational expression of most genes related to oxidative stress (Figure 2D). Group I and II antidepressants facilitated the expression of antioxidant defence proteins, such as superoxide dismutase, alkyl hydroperoxide reductase, and hydroperoxidase. In addition, increased protein expression was consistent with enhanced transcription of antioxidant defence genes. For example, both duloxetine and fluoxetine stimulated the expression of key response proteins including AhpC, AhpF, KatG, and SodC in the donor bacteria. OxyR is the activator of hydrogen peroxide‐inducible genes, protecting bacteria from oxidative damage (Imlay, 2013; Visick & Clarke, 1997). The expression of OxyR was enhanced by Group I and II antidepressants. The peroxide stress resistance protein YaaA can also protect bacteria from oxidative damage to DNA and proteins (Liu et al., 2011). The expression of YaaA was upregulated by sertraline, duloxetine, and fluoxetine. Similar to the donor, proteins protecting the recipient P. alloputida from oxidative damage were upregulated during the exposure to some antidepressants. These increased proteins included AhpC, AhpD, an AhpC/TSA family protein (NP_743395.1), peroxidase (NP_742403.1), and OxyR. Overall, the donor E. coli exhibited a stronger ROS response compared to the recipient, at both transcriptional (Figure 2C) and translational (Figure 2D) levels upon exposure to Group I and II antidepressants, which corresponds to the phenotypic result of ROS overproduction (Figure 2A).

Antidepressants increase cell membrane permeability

ROS and external chemicals can damage the bacterial cell membrane, thus facilitating the transfer of genetic materials across this barrier (Thomas & Nielsen, 2005). To investigate whether the increased conjugation ratio was associated with changes in cell membrane permeability, the donor and the recipient were separately exposed to antidepressants at concentrations employed in the conjugation assays. Flow cytometry analysis of cell populations revealed that some antidepressant drugs can damage cell membrane integrity, evidenced via DNA staining using PI as a marker of membrane damage (Cevik & Dalkara, 2003). Although cell membrane permeability of both strains was affected by antidepressant drugs, the donor E. coli was more prone to the antidepressant exposure compared with the recipient P. alloputida (Figure 3A, raw data in Figure S8). Increased cell membrane permeability associated with fluoxetine, duloxetine, and sertraline was consistent with their profiles of ROS overproduction. The effect on cell membrane permeability was even observed at low concentrations. Although bupropion did not induce ROS overproduction in the donor, cell membrane permeability increased over a broad concentration range. Typical flow cytometric results of increased cell membrane permeability were exemplified by the exposure of the donor and recipient bacteria to sertraline at 10.0 mg/L (Figure 3B). Still, agomelatine was the only antidepressant failed to enhance cell membrane permeability of bacteria.

FIGURE 3.

FIGURE 3

Open in a new tab

Effects of antidepressant drugs on cell membrane in the donor (E. coli K‐12 LE392) and recipient (P. alloputida KT2440) strains. (A) Fold changes of cell membrane permeability upon the exposure to different concentrations of antidepressants (raw data in Figure S8). (B) Flow cytometric results of sertraline (10.0 mg/L) treated bacteria (stained by PI), cells labelled with blue located at right quadrants were considered with enhanced florescence attributed to increased cell membrane permeability, (1) the donor strain treated by drug‐free solvent (Milli‐Q water), (2) the donor strain exposed to 10.0 mg/L of sertraline, (3) the recipient strain treated by drug‐free solvent (Milli‐Q water), (4) the recipient strain exposed to 10.0 mg/L of sertraline. (C) Fold changes of transcriptional expression of key genes associated with the cell membrane in the donor and recipient bacteria. (D) Fold changes of translational expression of key genes related to the cell membrane in the donor and recipient bacteria. Significant differences between antidepressant‐treated groups and the control group were calculated using independent‐sample t test, p values were then corrected by the ‘Benjamini–Hochberg’ method as p, *p adj < 0.05, and **p adj < 0.01.

Generally, the cell membrane permeability of bacteria in the presence of antidepressants dropped to the control level under anaerobic conditions (Figure S9). The results of cell membrane permeability under anaerobic conditions confirmed that ROS overproduction plays an important role in damaging cell membrane integrity.

Antidepressants altered the transcriptional expression of cell membrane relevant genes, consequently cell membrane permeability. Increased cell membrane permeability contributes to the conjugative transfer of plasmids. Sertraline upregulated the transcription of genes linked to outer membrane integrity in both the donor and recipient cells. Exposure to fluoxetine, duloxetine, escitalopram, and bupropion upregulated porin genes in the recipient. The expression of the outer membrane porin gene ompC, outer membrane protein synthesis regulator gene ompR, transmembrane transporter gene ompW, and outer membrane channel gene tolC was enhanced in the donor. Unlike in the donor bacteria, the expression of porin genes (opdC, opdH, and opdT‐I) in the recipient was enhanced by most of the antidepressant drugs tested. Increased expression of membrane porin genes, outer membrane channel genes, and transmembrane transporter genes was associated with increased cell membrane permeability. Duloxetine, fluoxetine, and bupropion increased cell membrane permeability of the donor, mostly due to drug‐dependent damage on cell membrane integrity. As a defence mechanism, the expression of some membrane associated genes may be downregulated upon the exposure to external stress, particularly antibiotic‐like toxicity (Bystritskaya et al., 2014; Pagès et al., 2008).

The analyses on protein abundance also supported the finding that antidepressants can enhance cell membrane permeability (Figure 3D). Group I antidepressants enhanced the expression of membrane porin proteins in the donor. Bupropion upregulated the expression of porin proteins in the donor and recipient bacteria, and the increment was comparable to that of Group I antidepressants. In the donor, the expression of porin proteins (OmpC, OmpF, OmpR, and LamB) and the outer membrane channel protein TolC increased with the presence of antidepressants. Altered expression of membrane proteins was also observed in the recipient in the presence of antidepressants; the abundance of porin proteins, OprD, OprE, OprF, OprQ, OpdC, OpdP, GlpF, and FecA was upregulated. The expression of porin proteins in the recipient was also upregulated by Group I antidepressants. Sertraline, duloxetine, fluoxetine, and escitalopram increased the abundance of most membrane permeability associated proteins. Bupropion increased the expression of all of these porin proteins in the recipient, suggesting that increased cell membrane permeability could play an important role in bupropion promoted conjugation. Still, agomelatine had limited effect on the expression of membrane proteins.

Overall, based on the results of the cell membrane permeability assay, we found that an enhanced conjugation ratio was accompanied by the increased cell membrane permeability for the sertraline, duloxetine, and fluoxetine treated groups. Although bupropion promoted an increased conjugation ratio, only the cell membrane permeability of the donor increased, and the increment was dose independent. In the escitalopram‐treated group, an increased conjugation ratio occurred without increased cell membrane permeability. Agomelatine neither enhanced conjugation ratio nor increased cell membrane permeability.

Antidepressants trigger SOS and adaptive responses

The stress from antidepressants on bacteria has been demonstrated by ROS overproduction and increased cell membrane permeability. Furthermore, the SOS response was triggered by antidepressants, especially sertraline, duloxetine, fluoxetine, and bupropion (Figure 4A). In the donor bacterium, genes relevant to the SOS response were upregulated, including lexA, recA, recR, sulA, uvrB, yebG. RecA and LexA play important roles in the induction of ROS responses, and the expression of both genes is activated within 5 min of DNA damage (Polosina, 2014). Therefore, protein abundance of LexA and RecA can be an indicator for DNA damage. The extent of LexA cleavage and the expression of RecA are proportional to DNA damage (Polosina, 2014). The analysis of protein abundance (Figure S10) showed that LexA was downregulated while RecA was upregulated when exposed to sertraline, duloxetine, fluoxetine, and bupropion. The yebG gene encodes a DNA‐damage inducible protein (Lomba et al., 1997), the transcription of which was upregulated by sertraline, duloxetine, and fluoxetine. Sertraline and duloxetine also upregulated the transcription of uvrB, which encodes a protein involved in recognition of DNA damage and repair (Verhoeven et al., 2002). In the recipient strain, genes that respond to DNA damage were also upregulated, including genes encoding DNA mismatch repair proteins (mutL, mutS), the gene encoding a DNA repair enzyme (radA), genes relevant to the SOS response (recC, recF, recQ, recR), and the cell division inhibitor gene (sulA). The abundance of proteins associated with these DNA repair genes further confirmed antidepressants trigger the SOS response, in particular sertraline, duloxetine, fluoxetine, and bupropion.

FIGURE 4.

FIGURE 4

Open in a new tab

Effects of antidepressant drugs on SOS response, universal stress response and conjugation relevant genes. (A) Fold changes of transcriptional expression of genes involved in SOS response in the donor and recipient bacteria. (B) Fold changes of transcriptional expression of genes involved in universal stress response in the donor and recipient bacteria. (C) Fold changes of transcriptional expression of regulatory genes related to conjugation carried by plasmid RP4 in the donor bacteria

Apart from SOS responses, antidepressants also triggered universal stress responses in bacteria. E. coli harbours six universal stress protein genes, namely uspA, uspC, uspD, uspE, uspF, and uspG. Briefly, usp genes are crucial for bacterial survival against multiple environmental stresses. Most usp genes in both strains were upregulated by the presence of sertraline, fluoxetine, bupropion, and escitalopram (Figure 4B). On the contrary, agomelatine was the only antidepressant that downregulated the universal stress response in the donor.

Together, these data show that the antidepressant drugs sertraline, duloxetine, fluoxetine, and bupropion promote the conjugative transfer of ARGs (especially sertraline, duloxetine, fluoxetine, and bupropion), impart stress to bacteria, and trigger SOS and universal stress responses. In contrast, agomelatine neither promoted conjugation nor triggered the SOS response.

Antidepressants affect the expression of conjugation relevant genes on plasmid RP4

Conjugation is a complex process coordinated by multiple genes. Among these genes, traC encodes an inner membrane NTPase TraC, which is required for pilus assembly (Virolle et al., 2020); traB, traC, and traE genes are indispensable for the regulation of the pilus assembly in plasmid RP4 (Virolle et al., 2020); the expression of traB and traF regulate the pilus extension (Virolle et al., 2020); traG encodes an inner membrane protein that involved not only in pilus assembly but also mating pair stabilization and exclusion. Transfer gene expression initiates with the production of regulator protein TraJ that activates the transcription of the tra operon (Virolle et al., 2020). The conjugative transfer also requires the relaxosome, TraM is a relaxosome accessory protein that interacts with TraD to prepare the plasmid for conjugation (Penfold et al., 1996) and is crucial for increasing the transfer efficiency (Lessl et al., 1993). In this study, sertraline upregulated the expression of all conjugation relevant genes (Figure 4C). Duloxetine upregulated most of these conjugation genes carried by plasmid RP4, except for traE. Fluoxetine enhanced the expression of most conjugation relevant genes except for traB and traM. On the contrary, escitalopram exerted limited de‐repression effect on the transcription of tra genes. Upon the exposure to sertraline, plasmid replication genes, pilus assembly genes, mate‐pair formation genes, the relaxosome gene, and genes promoting transfer efficiency were all de‐repressed. Duloxetine and fluoxetine also increased the transcription of most plasmid conjugation genes. The effect of bupropion and escitalopram on promoting the transcription of genes carried by plasmid RP4 was less than that of other antidepressants. Interestingly, agomelatine downregulated transcription of the relaxosome gene traM.

DISCUSSION

Antibiotics at sub‐inhibitory concentrations are considered as the stimuli to promote the spread of ARGs (Andersson & Hughes, 2014; Kümmerer, 2001). However, the role of antidepressants in the spread of antibiotic resistance has been largely overlooked. Our results show that several common antidepressants (sertraline, duloxetine, fluoxetine, escitalopram, and bupropion) can promote the dissemination of antibiotic resistance determinants via increased plasmid conjugation. Furthermore, the mating system based on the clinically sourced plasmid pMS6198A further confirmed that commonly prescribed antidepressants at clinically relevant concentrations can promote the spread of ARGs (Ejim et al., 2011; Maier et al., 2018).

In order to reveal the underlying mechanisms of conjugation promoted by commonly prescribed antidepressant drugs, multiple approaches were employed in this study, including flow cytometry to detect ROS production and cell membrane permeability, conjugation with ROS scavenger or under anaerobic conditions, whole‐genome RNA sequencing, and quantitative proteomic analysis. The major mechanisms identified include ROS overproduction, increased cell membrane permeability, and the de‐repression of conjugation relevant genes following exposure to these antidepressant drugs (Figure 5).

FIGURE 5.

FIGURE 5

Open in a new tab

The proposed mechanisms underlying the phenomenon that antidepressant drugs promote the conjugative transfer of plasmid borne ARGs across bacterial genera. (A) Antidepressant drugs induce ROS overproduction in both the donor and recipient bacteria. (B) Antidepressant drugs increase cell membrane permeability in both the donor and recipient bacteria. (C) Increased ROS production and cell membrane permeability further induce SOS response. (D) Transcriptional expression of core genes in charge of conjugation carried by plasmid RP4 is affected by the presence of antidepressants.

ROS overproduction induced by antidepressants was shown to be a major contributor to enhanced conjugative plasmid transfer, particularly in the recipient. The relatively milder response in the recipient, compared to the donor, suggests the response to antidepressants is bacterial species‐specific, which is similar to the bacterial species‐specific response to various antibiotics (Brochado et al., 2018). Following exposure to increasing concentrations of Group I antidepressants (sertraline, duloxetine, fluoxetine, and escitalopram), ROS production in bacteria increased significantly (p adj < 0.05). As a defence mechanism against oxidative stress, bacteria can balance increased ROS by activating antioxidant systems (Bae et al., 2011). Indeed, we detected overexpression of alkyl hydroperoxide reductase (ahpC and ahpF), hydroperoxidase (katG), superoxide dismutase (sodA, sodB, and sodC), and thioredoxin reductase (trxB) following exposure to Group I antidepressants. The role of antidepressants induced ROS overproduction in enhanced conjugation ratios was further consolidated by the conjugation assays conducted under anaerobic conditions or with the addition of ROS scavenger, in which the transfer ratios decreased dramatically toward the spontaneous level. These results further support increased conjugation is driven by the antidepressant‐induced ROS overproduction. Successful transfer of plasmids to transconjugants might not only amplify their advantage for survival but also increase their resistance against antidepressants, protecting the host from ROS induced damages. Antidepressants induced ROS overproduction could lead to DNA damage in bacteria (Kohanski et al., 2010), which potentially affects bacterial viability. Compared to the donor and recipient cells, the transconjugants integrate the resistance from both bacteria and could have plasmid carried genes to repair the ROS‐induced DNA damages (Strike & Lodwick, 1987). Such phenomenon is similar to persistent bacteria under the stress from antibiotics (Cohen et al., 2013). As for the recipient, the bacterium could be lack of plasmid‐encoded proteins to repair the ROS‐induced damages. Consequently, transconjugants are capable of better adapting to the ROS‐induced DNA damage and have advantages in antidepressants' selection during the growth, compared to the plasmid‐free recipient (De Furio et al., 2017). Therefore, transconjugants could survive better and promote the spread of ARGs after acquiring plasmids under the selective pressure from antidepressants and subsequent ROS‐induced damage.

Alterations in bacterial cell membrane permeability co‐occurred with increased conjugation upon the exposure to antidepressants, especially for Group I antidepressants. This was likely attributed to ROS overproduction. Additionally, the transcriptional and translational expression of genes relevant to increased cell membrane permeability was also upregulated, especially in the recipient. The translation of porin proteins was significantly upregulated in bacteria. These porin proteins include OmpC, ‐F, ‐R, LamB, FecA, GlpF, OpdC, ‐D, ‐E, ‐F, ‐P, and ‐Q; with OmpC and OmpF known to control cell membrane permeability (Koebnik et al., 2000; Nikaido & Nakae, 1980). Porins are hydrophilic transmembrane channel proteins, which allow the passive diffusion of salts and nutrients across cell membrane from the high concentration gradient (Sharma et al., 2022). Additionally, antibiotics can enter the cytoplasm of bacteria through porins, consequently exert their antibacterial effects (Liu et al., 2022). Although antidepressants applied in this study are different from antibiotics used in previous studies, increased expression of porins might assist the entering of antidepressants to bacterial cytoplasm. Both increased porin expression and damaged cell membrane could likely lead to elevated cytoplasmic concentration of antidepressants, thus magnifying the antibiotic‐like role of antidepressants, for example, stimulating ROS response. Increased cell membrane permeability can be the consequence of ROS overproduction, while the introduction of ROS eliminated mating systems proved the contribution of ROS generation in stimulating conjugation. Also, bupropion enhanced conjugation showed that antidepressants enhanced conjugation could be accompanied with increased cell membrane permeability without ROS overproduction. Combined with the damage on cell membrane integrity attributed to ROS overproduction, upregulated expression of porins increased cell membrane permeability, which co‐occurred with enhanced conjugative transfer. We suggest that changes in cell membrane integrity and increased cell membrane permeability explain a possible mechanism induced by antidepressants that associates with the conjugative transfer of plasmid borne ARGs. Further investigations are required to reveal the relationship between increased cell membrane permeability and enhanced conjugation.

Antidepressants may promote the conjugative transfer of plasmid RP4 by de‐repressing the mRNA expression of mate pair formation genes and replication relevant genes. Plasmid RP4 is an IncPα group conjugal plasmid with a broad host range (Samuels et al., 2000). The conjugative transfer of which relies on the expression of tra genes which regulated by the host bacteria, cell cycle progression and environmental conditions (Virolle et al., 2020). The mate pair formation system controls the generation of pilus, through which the plasmid DNA is transferred from the donor to the recipient (Smillie et al., 2010). In this study, we detected increased transcription of genes involved in replication (traC) and mate pair formation (traB, ‐E, ‐F, ‐G), with the expression of these conjugation relevant genes most highly upregulated upon exposure to sertraline.

We also found that antidepressants can trigger SOS and universal stress responses in bacteria. Changes in the transcriptional level of SOS response relevant genes, as observed with the increased expression of DNA repair genes, further validated the DNA damage induced by antidepressants, which potentially attributed to elevated ROS levels (Bondarenko et al., 2012). It is known that HGT can be promoted by activation of the SOS response (Beaber et al., 2004; Maiques et al., 2006; Úbeda et al., 2005), and the transfer process, as mediated by ssDNA, may also induce SOS responses (Baharoglu et al., 2010). As a result, the spread of plasmid borne ARGs could be accelerated by the SOS response triggered by antidepressants and the transfer process itself. In addition to SOS response, the presence of antidepressants also triggered universal stress responses. Again, the universal stress response can be linked to ROS overproduction, since the transcription of uspA and uspD was upregulated (Nachin et al., 2005). Our data showed that sertraline, duloxetine, and fluoxetine represent antidepressant drugs that induce bacterial stress. Further validation of antidepressants induced transcriptomic changes by RT‐qPCR could strengthen current findings based on whole‐genome RNA sequencing.

As effective medical interventions to the increasingly common MDD, antidepressant medications commonly last for months to years, even lifetime to avoid the recurrence (Mueller & Leon, 1996). Current COVID pandemic worsens the mental health issues, which has led to increased consumption of antidepressants (Rabeea et al., 2021), consequently expose antibiotic resistance bacteria in the human gut and environments to antidepressants more frequently. Our findings highlight the possibility of antidepressants, in particular sertraline, duloxetine, fluoxetine and bupropion at concentrations above 1 mg/L, facilitating the spread of ARGs, which is of clinical importance. Also, sertraline and bupropion at a concentration as low as 0.1 mg/L, approximating to environmental concentrations, stimulated HGT, which uncovered the environmental significance of antidepressants as micropollutants.

In human gut, most of antidepressants travel through the gastrointestinal tract (GI tract) then excreted through faeces and urine. The excremental concentrations of antidepressants have a broad concentration range, from 0.1 to 100.0 mg/L levels, as exemplified by the urine concentration and the colonic concentration of fluoxetine, from low to high (McGovern et al., 2019; Unceta et al., 2008). Absorbed antidepressants can circulate in the human body with half‐life from 1–2 h (agomelatine) to 144 h (fluoxetine) (Marken & Munro, 2000). Therefore, the antidepressant medications expose the microbes in gut and urinary tract to antidepressants at different concentrations. Importantly, antidepressants have been discovered with antibiotic‐like effects on human gut bacteria (McGovern et al., 2019). Furthermore, our study revealed the antibiotic‐like role of antidepressants, which could promote the transfer of antibiotic resistance from gut bacteria to the others.

Our study also implies that antidepressants can promote the conjugal transfer of bla NDM−1 positive plasmid from uropathogenic E. coli to environmental bacteria, such as P. alloputida. Urinary tract infections dominated by uropathogenic E. coli are some of the most common bacterial infections, with a lifetime incidence of 50%–60% in adult women (Wuerstle et al., 2011). Plasmids are prevalent ARGs‐carriers among uropathogenic E. coli, 81% (221/273) of uropathogenic E. coli isolates were plasmid‐positive (Alós, 2005). Therefore, antidepressants in urine potentially facilitate the exchange of plasmids between uropathogenic E. coli, resulting in the emergence or evolution of dangerous multi‐antibiotic‐resistant superbugs and complicated infections hard to be eradicated. Noteworthy, a significant association was found between antidepressant use and lower urinary tract symptoms among men aged 45–69 years (n = 63,579; OR = 1.36, 95% CI: 1.29–1.44) (Tarlton et al., 2019), which further indicate a potential risk that antidepressant medication may affect urinary microbiome resistome via promoting the emergence or dissemination of antibiotic resistance.

Although environmentally relevant concentrations of antidepressants are at nanograms per litre or microgram per litre levels, we still should not overlook the contribution of antidepressants on the spread of antibiotic resistance in the environment. This is because that some microorganisms in the environment might be more sensitive to antidepressants or can accumulate antidepressants (Klünemann et al., 2021). Particularly, it is required to assess the risk of the spread plasmid‐borne ARGs in wastewater treatment plants, where human pathogens and environmental bacteria could co‐exist and high concentrations of antidepressants are detected in influent (Bengtsson‐Palme et al., 2015; Guo et al., 2017; Wilkinson et al., 2022).

In conclusion, our study illustrates that antidepressants significantly promote the horizontal transfer of clinically and environmentally sourced antibiotic resistance plasmids across bacterial genera through conjugation. The findings raise the awareness of antibiotic‐like roles of antidepressants, suggesting the prescription of antidepressants should also consider their effect on the spread of antibiotic resistance and the consequences of depleting our current antibiotic arsenal. Long‐term antidepressant treatment leads to prolonged exposure of bacteria, potentially superbugs, for example, the UPEC strain hosting pMS6198A plasmid, to the antibiotic‐like stress, which may promote the transfer of multidrug resistance from superbugs to other bacteria. In the future, the evaluation of antidepressants should also consider their promotive effect on HGT based on in vivo animal models or cohort studies.

AUTHOR CONTRIBUTIONS

Jianhua Guo and Pengbo Ding conceived and designed the study. Pengbo Ding conducted the conjugation experiments, flow cytometer, ROS and membrane permeability measurement, RNA and protein sequencing, data analysis and interpretation. Ji Lu assisted with conjugation assays, flow cytometric analysis, and mRNA extraction. Yue Wang assisted with mRNA extraction and proteomics analysis. Mark A. Schembri provided Escherichia coli strain harbouring plasmid pMS6198A, and contributed to experimental design and provided critical biological interpretations of the data. Pengbo Ding wrote the paper. Jianhua Guo and Mark A. Schembri supervised this work and edited on the manuscript.

CONFLICT OF INTEREST

The authors report no conflict of interest.

Supporting information

Appendix S1 Supplementary Information

Click here for additional data file. (17.7MB, docx)

ACKNOWLEDGEMENTS

This study was financially supported through Australian Research Council Future Fellowship (FT170100196) awarded to Jianhua Guo. Pengbo Ding also thanks UQ Research Training Scholarship for the financial support. The authors thank M. Nefedov (The University of Queensland) for flow cytometer analysis, A. Nouwens (The University of Queensland) for proteomic MS analysis, L. Mao for assisting with mRNA data analysis. Open access publishing facilitated by The University of Queensland, as part of the Wiley ‐ The University of Queensland agreement via the Council of Australian University Librarians.

Ding, P. , Lu, J. , Wang, Y. , Schembri, M.A. & Guo, J. (2022) Antidepressants promote the spread of antibiotic resistance via horizontally conjugative gene transfer. Environmental Microbiology, 24(11), 5261–5276. Available from: 10.1111/1462-2920.16165

Funding information Australian Research Council Future Fellowship, Grant/Award Number: FT170100196

Data Availability Statement

Genomic information is deposited in public databases as indicated in the Methods section (NCBI Gene Expression Omnibus accession number GSE181900 and PRIDE partner repository PXD028020). The study protocol is available with this publication and a copy of anonymized raw data used for analyses in this study has been included as part of the supporting information.

REFERENCES

  1. Adu, M.K. , Wallace, L.J. , Lartey, K.F. , Arthur, J. , Oteng, K.F. , Dwomoh, S. et al. (2021) Prevalence and correlates of likely major depressive disorder among the adult population in Ghana during the COVID‐19 pandemic. International Journal of Environmental Research and Public Health, 18, 7106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allen, H.K. , Donato, J. , Wang, H.H. , Cloud‐Hansen, K.A. , Davies, J. & Handelsman, J. (2010) Call of the wild: antibiotic resistance genes in natural environments. Nature Reviews. Microbiology, 8, 251–259. [DOI] [PubMed] [Google Scholar]
  3. Alós, J.I. (2005) Epidemiology and etiology of urinary tract infections in the community. Antimicrobial susceptibility of the main pathogens and clinical significance of resistance. Enfermedades Infecciosas y Microbiología Clínica, 23, 3–8. [DOI] [PubMed] [Google Scholar]
  4. Andersson, D.I. & Hughes, D. (2014) Microbiological effects of sublethal levels of antibiotics. Nature Reviews. Microbiology, 12, 465–478. [DOI] [PubMed] [Google Scholar]
  5. Bae, Y.S. , Oh, H. , Rhee, S.G. & Yoo, Y.D. (2011) Regulation of reactive oxygen species generation in cell signaling. Molecules and Cells, 32, 491–509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Baharoglu, Z. , Bikard, D. & Mazel, D. (2010) Conjugative DNA transfer induces the bacterial SOS response and promotes antibiotic resistance development through integron activation. PLoS Genetics, 6, e1001165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Beaber, J.W. , Hochhut, B. & Waldor, M.K. (2004) SOS response promotes horizontal dissemination of antibiotic resistance genes. Nature, 427, 72–74. [DOI] [PubMed] [Google Scholar]
  8. Bengtsson‐Palme, J. , Angelin, M. , Huss, M. , Kjellqvist, S. , Kristiansson, E. , Palmgren, H. et al. (2015) The human gut microbiome as a transporter of antibiotic resistance genes between continents. Antimicrobial Agents and Chemotherapy, 59, 6551–6560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Benjamini, Y. & Hochberg, Y. (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B: Methodological, 57, 289–300. [Google Scholar]
  10. Bondarenko, O. , Ivask, A. , Käkinen, A. & Kahru, A. (2012) Sub‐toxic effects of CuO nanoparticles on bacteria: kinetics, role of Cu ions and possible mechanisms of action. Environmental Pollution, 169, 81–89. [DOI] [PubMed] [Google Scholar]
  11. Brochado, A.R. , Telzerow, A. , Bobonis, J. , Banzhaf, M. , Mateus, A. , Selkrig, J. et al. (2018) Species‐specific activity of antibacterial drug combinations. Nature, 559, 259–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Broxton, C.N. & Culotta, V.C. (2016) SOD enzymes and microbial pathogens: surviving the oxidative storm of infection. PLoS Pathogens, 12, e1005295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Bystritskaya, E.P. , Stenkova, A.M. , Portnyagina, O.Y. , Rakin, A.V. , Rasskazov, V.A. & Isaeva, M.P. (2014) Regulation of Yersinia pseudotuberculosis major porin expression in response to antibiotic stress. Molecular Genetics, Microbiology and Virology, 29, 63–68. [PubMed] [Google Scholar]
  14. Cevik, I.U. & Dalkara, T. (2003) Intravenously administered propidium iodide labels necrotic cells in the intact mouse brain after injury. Cell Death and Differentiation, 10, 928–929. [DOI] [PubMed] [Google Scholar]
  15. Cohen, N.R. , Lobritz, M.A. & Collins, J.J. (2013) Microbial persistence and the road to drug resistance. Cell Host & Microbe, 13, 632–642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Coles, R. & Kharasch, E.D. (2007) Stereoselective analysis of bupropion and hydroxybupropion in human plasma and urine by LC/MS/MS. Journal of Chromatography B, 857, 67–75. [DOI] [PubMed] [Google Scholar]
  17. Davison, J. (1999) Genetic exchange between bacteria in the environment. Plasmid, 42, 73–91. [DOI] [PubMed] [Google Scholar]
  18. De Furio, M. , Ahn, S.J. , Burne, R.A. & Hagen, S.J. (2017) Oxidative stressors modify the response of Streptococcus mutans to its competence signal peptides. Applied and Environmental Microbiology, 83, e01345–e01317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. DeVane, C.L. (1999) Metabolism and pharmacokinetics of selective serotonin reuptake inhibitors. Cellular and Molecular Neurobiology, 19, 443–466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Ejim, L. , Farha, M.A. , Falconer, S.B. , Wildenhain, J. , Coombes, B.K. , Tyers, M. et al. (2011) Combinations of antibiotics and nonantibiotic drugs enhance antimicrobial efficacy. Nature Chemical Biology, 7, 348–350. [DOI] [PubMed] [Google Scholar]
  21. Gao, S.‐H. , Fan, L. , Peng, L. , Guo, J. , Agulló‐Barceló, M. , Yuan, Z. et al. (2016) Determining multiple responses of Pseudomonas aeruginosa PAO1 to an antimicrobial agent, free nitrous acid. Environmental Science & Technology, 50, 5305–5312. [DOI] [PubMed] [Google Scholar]
  22. Guo, J. , Li, J. , Chen, H. , Bond, P.L. & Yuan, Z. (2017) Metagenomic analysis reveals wastewater treatment plants as hotspots of antibiotic resistance genes and mobile genetic elements. Water Research, 123, 468–478. [DOI] [PubMed] [Google Scholar]
  23. Hancock, S.J. , Phan, M.‐D. , Peters, K.M. , Forde, B.M. , Chong, T.M. , Yin, W.‐F. et al. (2017) Identification of IncA/C plasmid replication and maintenance genes and development of a plasmid multilocus sequence typing scheme. Antimicrobial Agents and Chemotherapy, 61, e01740–e01716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hancock, S.J. , Phan, M.‐D. , Roberts, L.W. , Vu, T.N.M. , Harris, P.N.A. , Beatson, S.A. et al. (2021) Characterization of DtrJ as an IncC plasmid conjugative DNA transfer component. Molecular Microbiology, 116, 154–167. [DOI] [PubMed] [Google Scholar]
  25. Imlay, J.A. (2013) The molecular mechanisms and physiological consequences of oxidative stress: lessons from a model bacterium. Nature Reviews. Microbiology, 11, 443–454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jin, M. , Lu, J. , Chen, Z. , Nguyen, S.H. , Mao, L. , Li, J. et al. (2018) Antidepressant fluoxetine induces multiple antibiotics resistance in Escherichia coli via ROS‐mediated mutagenesis. Environment International, 120, 421–430. [DOI] [PubMed] [Google Scholar]
  27. Jung, I.L. & Kim, I.G. (2003) Transcription of ahpC, katG, and katE genes in Escherichia coli is regulated by polyamines: polyamine‐deficient mutant sensitive to H2O2‐induced oxidative damage. Biochemical and Biophysical Research Communications, 301, 915–922. [DOI] [PubMed] [Google Scholar]
  28. Kessler, R.C. , Petukhova, M. , Sampson, N.A. , Zaslavsky, A.M. & Wittchen, H.U. (2012) Twelve‐month and lifetime prevalence and lifetime morbid risk of anxiety and mood disorders in the United States. International Journal of Methods in Psychiatric Research, 21, 169–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Klünemann, M. , Andrejev, S. , Blasche, S. , Mateus, A. , Phapale, P. , Devendran, S. et al. (2021) Bioaccumulation of therapeutic drugs by human gut bacteria. Nature, 597, 533–538. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Koebnik, R. , Locher, K.P. & Van Gelder, P. (2000) Structure and function of bacterial outer membrane proteins: barrels in a nutshell. Molecular Microbiology, 37, 239–253. [DOI] [PubMed] [Google Scholar]
  31. Kohanski, M.A. , DePristo, M.A. & Collins, J.J. (2010) Sublethal antibiotic treatment leads to multidrug resistance via radical‐induced mutagenesis. Molecular Cell, 37, 311–320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kümmerer, K. (2001) Drugs in the environment: emission of drugs, diagnostic aids and disinfectants into wastewater by hospitals in relation to other sources–a review. Chemosphere, 45, 957–969. [DOI] [PubMed] [Google Scholar]
  33. Lessl, M. , Balzer, D. , Weyrauch, K. & Lanka, E. (1993) The mating pair formation system of plasmid RP4 defined by RSF1010 mobilization and donor‐specific phage propagation. Journal of Bacteriology, 175, 6415–6425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Liu, Y. , Bauer, S.C. & Imlay, J.A. (2011) The YaaA protein of the Escherichia coli OxyR regulon lessens hydrogen peroxide toxicity by diminishing the amount of intracellular unincorporated iron. Journal of Bacteriology, 193, 2186–2196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Liu, Y. , Yang, F. , Wang, S. , Chi, W. , Ding, L. , Liu, T. et al. (2022) HopE and HopD porin‐mediated drug influx contributes to intrinsic antimicrobial susceptibility and inhibits streptomycin resistance acquisition by natural transformation in Helicobacter pylori . Microbiology Spectrum, 10, e0198721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Loewen, P.C. , Switala, J. & Triggs‐Raine, B.L. (1985) Catalases HPI and HPII in Escherichia coli are induced independently. Archives of Biochemistry and Biophysics, 243, 144–149. [DOI] [PubMed] [Google Scholar]
  37. Lomba, M.R. , Vasconcelos, A.T. , Pacheco, A.B.F. & de Almeida, D.F. (1997) Identification of yebG as a DNA damage‐inducible Escherichia coli gene. FEMS Microbiology Letters, 156, 119–122. [DOI] [PubMed] [Google Scholar]
  38. Luo, Y. , Wang, Q. , Lu, Q. , Mu, Q. & Mao, D. (2014) An ionic liquid facilitates the proliferation of antibiotic resistance genes mediated by class I integrons. Environmental Science & Technology Letters, 1, 266–270. [Google Scholar]
  39. Macedo, D. , Filho, A. , Soares de Sousa, C.N. , Quevedo, J. , Barichello, T. , Junior, H.V.N. et al. (2017) Antidepressants, antimicrobials or both? Gut microbiota dysbiosis in depression and possible implications of the antimicrobial effects of antidepressant drugs for antidepressant effectiveness. Journal of Affective Disorders, 208, 22–32. [DOI] [PubMed] [Google Scholar]
  40. Maier, L. , Pruteanu, M. , Kuhn, M. , Zeller, G. , Telzerow, A. , Anderson, E.E. et al. (2018) Extensive impact of non‐antibiotic drugs on human gut bacteria. Nature, 555, 623–628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Maiques, E. , Úbeda, C. , Campoy, S. , Salvador, N. , Lasa, Í. , Novick, R.P. et al. (2006) β‐Lactam antibiotics induce the SOS response and horizontal transfer of virulence factors in Staphylococcus aureus . Journal of Bacteriology, 188, 2726–2729. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Marken, P.A. & Munro, J.S. (2000) Selecting a selective serotonin reuptake inhibitor: clinically important distinguishing features. Primary Care Companion to the Journal of Clinical Psychiatry, 2, 205–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. McGovern, A.A.‐O. , Hamlin, A.S. & Winter, G.A.‐O.X. (2019) A review of the antimicrobial side of antidepressants and its putative implications on the gut microbiome. The Australian and New Zealand Journal of Psychiatry, 53, 1151–1166. [DOI] [PubMed] [Google Scholar]
  44. Mole, R.A. & Brooks, B.W. (2019) Global scanning of selective serotonin reuptake inhibitors: occurrence, wastewater treatment and hazards in aquatic systems. Environmental Pollution, 250, 1019–1031. [DOI] [PubMed] [Google Scholar]
  45. Mueller, T.I. & Leon, A.C. (1996) Time to recovery, chronicity, and levels of psychopathology in major depression. The Psychiatric Clinics of North America, 19, 85–102. [DOI] [PubMed] [Google Scholar]
  46. Nachin, L. , Nannmark, U. & Nyström, T. (2005) Differential roles of the universal stress proteins of Escherichia coli in oxidative stress resistance, adhesion, and motility. Journal of Bacteriology, 187, 6265–6272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Nikaido, H. & Nakae, T. (1980) The outer membrane of Gram‐negative bacteria. Advances in Microbial Physiology, 20, 163–250. [DOI] [PubMed] [Google Scholar]
  48. Pagès, J.‐M. , James, C.E. & Winterhalter, M. (2008) The porin and the permeating antibiotic: a selective diffusion barrier in Gram‐negative bacteria. Nature Reviews. Microbiology, 6, 893–903. [DOI] [PubMed] [Google Scholar]
  49. Penfold, S.S. , Simon, J. & Frost, L.S. (1996) Regulation of the expression of the traM gene of the F sex factor of Escherichia coli . Molecular Microbiology, 20, 549–558. [DOI] [PubMed] [Google Scholar]
  50. Pereira, V.S. & Hiroaki‐Sato, V.A. (2018) A brief history of antidepressant drug development: from tricyclics to beyond ketamine. Acta Neuropsychiatrica, 30, 307–322. [DOI] [PubMed] [Google Scholar]
  51. Polosina, Y. (2014) DNA Repair. In: Caplan, M. (Ed.) Reference module in biomedical sciences. Amsterdam, Netherlands: Elsevier. [Google Scholar]
  52. Prasad, A.K. & Mishra, P.C. (2017) Scavenging of superoxide radical anion and hydroxyl radical by urea, thiourea, selenourea and their derivatives without any catalyst: a theoretical study. Chemical Physics Letters, 684, 197–204. [Google Scholar]
  53. Rabeea, S.A. , Merchant, H.A.‐O. , Khan, M.U. , Kow, C.S. & Hasan, S.A.‐O. (2021) Surging trends in prescriptions and costs of antidepressants in England amid COVID‐19. Daru, 29, 217–221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Samuels, A.L. , Lanka, E. & Davies, J.E. (2000) Conjugative junctions in RP4‐mediated mating of Escherichia coli . Journal of Bacteriology, 182, 2709–2715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Scheuing, L. , Chiu, C.‐T. , Liao, H.‐M. & Chuang, D.‐M. (2015) Antidepressant mechanism of ketamine: perspective from preclinical studies. Frontiers in Neuroscience, 9, 249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Seaver, L.C. & Imlay, J.A. (2001) Alkyl hydroperoxide reductase is the primary scavenger of endogenous hydrogen peroxide in Escherichia coli . Journal of Bacteriology, 183, 7173–7181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Sharma, A. , Yadav, S.P. , Sarma, D. & Mukhopadhaya, A. (2022) Chapter two ‐ modulation of host cellular responses by Gram‐negative bacterial porins. Advances in Protein Chemistry and Struct Biology, 128, 35–77. [DOI] [PubMed] [Google Scholar]
  58. Smillie, C. , Garcillán‐Barcia, M.P. , Francia, M.V. , Rocha, E.P.C. & de la Cruz, F. (2010) Mobility of plasmids. Microbiology and Molecular Biology Reviews, 74, 434–452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Strike, P. & Lodwick, D. (1987) Plasmid genes affecting DNA repair and mutation. Journal of Cell Science. Supplement, 6, 303–321. [DOI] [PubMed] [Google Scholar]
  60. Tarlton, N.J. , Moritz, C. , Adams‐Sapper, S. & Riley, L.W. (2019) Genotypic analysis of uropathogenic Escherichia coli to understand factors that impact the prevalence of β‐lactam‐resistant urinary tract infections in a community. Journal of Global Antimicrobial Resistance, 19, 173–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Thomas, C.M. & Nielsen, K.M. (2005) Mechanisms of, and barriers to, horizontal gene transfer between bacteria. Nature Reviews. Microbiology, 3, 711–721. [DOI] [PubMed] [Google Scholar]
  62. Úbeda, C. , Maiques, E. , Knecht, E. , Lasa, Í. , Novick, R.P. & Penadés, J.R. (2005) Antibiotic‐induced SOS response promotes horizontal dissemination of pathogenicity Island‐encoded virulence factors in staphylococci. Molecular Microbiology, 56, 836–844. [DOI] [PubMed] [Google Scholar]
  63. Unceta, N. , Gómez‐Caballero, A. , Sánchez, A. , Millán, S. , Sampedro, M.C. , Goicolea, M.A. et al. (2008) Simultaneous determination of citalopram, fluoxetine and their main metabolites in human urine samples by solid‐phase microextraction coupled with high‐performance liquid chromatography. Journal of Pharmaceutical and Biomedical Analysis, 46, 763–770. [DOI] [PubMed] [Google Scholar]
  64. Verhoeven, E.E.A. , Wyman, C. , Moolenaar, G.F. & Goosen, N. (2002) The presence of two UvrB subunits in the UvrAB complex ensures damage detection in both DNA strands. The EMBO Journal, 21, 4196–4205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Virolle, C. , Goldlust, K. , Djermoun, S. , Bigot, S. & Lesterlin, C. (2020) Plasmid transfer by conjugation in Gram‐negative bacteria: from the cellular to the community level. Genes, 11, 1239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Visick, J.E. & Clarke, S. (1997) RpoS‐ and OxyR‐independent induction of HPI catalase at stationary phase in Escherichia coli and identification of rpoS mutations in common laboratory strains. Journal of Bacteriology, 179, 4158–4163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. von Wintersdorff, C.J.H. , Penders, J. , van Niekerk, J.M. , Mills, N.D. , Majumder, S. , van Alphen, L.B. et al. (2016) Dissemination of antimicrobial resistance in microbial ecosystems through horizontal gene transfer. Frontiers in Microbiology, 7, 173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Vos, T. , Allen, C. , Arora, M. , Barber, R.M. , Bhutta, Z.A. , Brown, A. et al. (2016) Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet, 388, 1545–1602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Wang, Y. , Lu, J. , Mao, L. , Li, J. , Yuan, Z. , Bond, P.L. et al. (2019) Antiepileptic drug carbamazepine promotes horizontal transfer of plasmid‐borne multi‐antibiotic resistance genes within and across bacterial genera. The ISME Journal, 13, 509–522. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Wilkinson, J.L. , Boxall, A.B.A. , Kolpin, D.W. , Leung, K.M.Y. , Lai, R.W.S. , Galbán‐Malagón, C. et al. (2022) Pharmaceutical pollution of the world's rivers. Proceedings of the National Academy of Sciences of the United States of America, 119, e2113947119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. World Health Organization . (2017) Depression and other common mental disorders: global health estimates. Geneva, Switzerland: World Health Organization. [Google Scholar]
  72. Writer, J.H. , Antweiler, R.C. , Ferrer, I. , Ryan, J.N. & Thurman, E.M. (2013) In‐stream attenuation of neuro‐active pharmaceuticals and their metabolites. Environmental Science & Technology, 47, 9781–9790. [DOI] [PubMed] [Google Scholar]
  73. Writer, J.H. , Ferrer, I. , Barber, L.B. & Thurman, E.M. (2013) Widespread occurrence of neuro‐active pharmaceuticals and metabolites in 24 Minnesota rivers and wastewaters. Science of the Total Environment, 461, 519–527. [DOI] [PubMed] [Google Scholar]
  74. Wuerstle, M.C. , Van Den Eeden, S.K. , Poon, K.T. , Quinn, V.P. , Hollingsworth, J.M. , Loo, R.K. et al. (2011) Contribution of common medications to lower urinary tract symptoms in men. Archives of Internal Medicine, 171, 1680–1682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Zhang, S. , Wang, Y. , Song, H. , Lu, J. , Yuan, Z. & Guo, J. (2019) Copper nanoparticles and copper ions promote horizontal transfer of plasmid‐mediated multi‐antibiotic resistance genes across bacterial genera. Environment International, 129, 478–487. [DOI] [PubMed] [Google Scholar]
  76. Zhang, W. , Qu, W. , Wang, H. & Yan, H. (2021) Antidepressants fluoxetine and amitriptyline induce alterations in intestinal microbiota and gut microbiome function in rats exposed to chronic unpredictable mild stress. Translational Psychiatry, 11, 131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Zhang, Y. , Gu, A.Z. , He, M. , Li, D. & Chen, J. (2017) Subinhibitory concentrations of disinfectants promote the horizontal transfer of multidrug resistance genes within and across genera. Environmental Science & Technology, 51, 570–580. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix S1 Supplementary Information

Click here for additional data file. (17.7MB, docx)

Data Availability Statement

Genomic information is deposited in public databases as indicated in the Methods section (NCBI Gene Expression Omnibus accession number GSE181900 and PRIDE partner repository PXD028020). The study protocol is available with this publication and a copy of anonymized raw data used for analyses in this study has been included as part of the supporting information.

Articles from Environmental Microbiology are provided here courtesy of Wiley

RESOURCES