Hypothetical Protein gene1038 Contributes to Colistin Resistance in Aeromonas hydrophila.

LETTER

Inhibition of vital respiratory enzymes, such as NADH:ubiquinone oxidoreductase (complex I), type II NADH-quinone oxidoreductases (NDH-2), and malate:quinone oxidoreductase, in the inner membrane is a secondary antibacterial mechanism of colistin (1–3). However, colistin resistance mechanisms associated with this secondary mode of action of colistin have rarely been reported. Herein, we confirm that the hypothetical protein gene1038 was associated with colistin resistance in Aeromonas hydrophila by reducing antibiotic function in the inner membrane, providing novel knowledge on the generation of colistin resistance.

The expression of gene1038 was significantly increased in the colistin-resistant strain 23-C-23 compared to that of the colistin-susceptible strain WCX23 (Fig. 1A) via quantitative reverse transcription-PCR (RT-PCR) using the primers gene1038-F (5′-GCTGCTTCGGCTTCCTCTAT-3′) and gene1038-R (5′-GGTCTCGCCGAACATGAGAT-3′) as previously described (4). This suggests that gene1038 might be associated with colistin resistance in A. hydrophila. To test this hypothesis, gene1038 was knocked out in a 23-C-23 background (23-C-23:Δgene1038) and subsequently restored in complemented strain 23-C-23:CΔgene1038, as previously described (4) (see Table S1 in the supplemental material). The results showed that the colistin MIC for 23-C-23:Δgene1038 was 8-fold lower than for the parent strain and that the complementation of gene1038 in 23-C-23:CΔgene1038 restored resistance to colistin (Table 1). Furthermore, antimicrobial susceptibility testing indicated that 23-C-23, 23-C-23:Δgene1038, and 23-C-23:CΔgene1038 showed similar resistance profiles toward other antimicrobial drugs (Table S2). These findings confirmed that gene1038 is exclusively involved in colistin resistance.

FIG 1

mRNA expression and protein modeling of gene1038. (A) Expression of gene1038 in a colistin-susceptible strain (WCX23) and colistin-resistant strain (23-C-23). Relative expression levels of genes were determined using the 2–ΔΔCT method (where CT is threshold cycle). Error bars represent the standard deviations from three biological replicates. Statistical analysis was performed using a Student two-tailed t test. ***, P value < 0.001. (B) TMHMM prediction results for gene1038. (C) SMART prediction results for gene1038. (D) I-Tasser homology modeling analysis of gene1038.

TABLE 1

Colistin MICs of A. hydrophila strains

Strain	Description	Colistin MIC (mg/liter)
WCX23	A. hydrophila strain isolated from a snake with diarrhea (5)	1
23-C-23	WCX23 after 23 passages with colistin (4)	256
23-C-23:Δgene1038	23-C-23 with the gene1038 knocked out	32
23-C-23:CΔgene1038	23-C-23:Δgene1038 with gene1038 complementation	256

gene1038 had >97.36% similarity with hypothetical proteins in A. hydrophila and at least 86.81% similarity with other Aeromonas spp. via BLASTP analysis (https://blast.ncbi.nlm.nih.gov/Blast.cgi). In contrast, the maximum similarity with other species was only 66.23%. These results suggest that gene1038 is exclusive to Aeromonas spp. According to the predictions of the TMHMM server (http://www.cbs.dtu.dk/services/TMHMM/) and SMART (http://smart.embl.de/), the protein product of gene1038, as expected, forms eight transmembrane regions (Fig. 1B and C). Structural models of gene1038 constructed using the i-TASSER server (https://zhanglab.ccmb.med.umich.edu/I-TASSER/) showed that the protein was structurally close to the membrane domain of respiratory complex I (Protein Data Bank accession no. 3RKO) (Fig. 1D), which was the target of colistin in the inner membrane (1). Accordingly, we speculated that upregulation of gene1038 in A. hydrophila might offset the inhibition of complex I by colistin, leading to the impaired bactericidal action of colistin in the inner membrane.

We observed that gene1038 was involved in NAD+/NADH ratio regulation using the NAD(H) level detection kit (Solarbio, Beijing, China) as reported previously (6). The deletion of gene1038 was accompanied by a decrease in NADH concentration and an increase in the NAD+/NADH ratio (Fig. 2). The complementation of gene1038 led to an increasing NADH concentration and a decreasing NAD+/NADH ratio (Fig. 2). These data indicated that gene1038 participated in the upregulation of NADH, resulting in the downregulation of the NAD+/NADH ratio. We speculated that cells might utilize gene1038 to regulate the NAD+/NADH ratio in response to the inhibition of complex I by colistin.

FIG 2

NAD+ and NADH levels and the NAD+/NADH ratio in bacterial strains. Changes in the NAD+ content (A), the NADH content (B), and the NAD+/NADH ratio (C) in 23-C-23, 23-C-23:Δgene1038, and 23-C-23:CΔgene1038. Error bars represent standard deviations from three biological replicates. Statistical analysis was performed using a Student two-tailed t test. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant.

Conclusively, the hypothetical protein gene1038 is a protein exclusive to Aeromonas spp., it possesses 8 transmembrane regions, it is structurally close to the membrane domain of respiratory complex I, and it participates in the modulation of the NAD+/NADH ratio. We revealed a novel (to our knowledge) colistin resistance mechanism mediated by the upregulation of gene1038 that might weaken colistin’s antibacterial effect through antagonizing the inhibition of respiratory complex I in the inner membrane.

Data were statistically analyzed using GraphPad Prism version 7.0 (GraphPad Software Inc., San Diego, CA, USA). The differences were analyzed using Student’s two-tailed unpaired t tests and are expressed as means ± standard deviations (SD), unless otherwise noted. Statistical significance was set at a P of <0.05.

Data availability.

The nucleotide sequences of gene1038 have been submitted to GenBank with the accession number MN862665.