Metabolic Regulation for Double Increasing Balofloxacin Concentrations in Vibrio Alginolyticus

Metabolome refers to the complete set of small-molecule metabolites to be found within a biological sample. Metabolomics approach provides the global qualitative and quantitative analysis of the small-molecule metabolites and shows great promise as a mean to identify biomarkers of drug efficacy and understand metabolic mechanisms. In the present study, metabolomics carrying cellular metabolites of Vibrio alginolyticus has been developed to analyze dynamic changes in metabolites induced upon stress with double increasing concentrations of balofloxacin (BLFX).  The objective was twofold.  First, we were interested in identifying metabolites that, being specifically regulated in response to BLFX exposure, might be involved in BLFX-stressed biomarkers.  Second, we wanted to exploit the metabolomics data to identify, among the BLFX-stressed metabolites, compounds capable of being reasonable proofs describing the characteristic feature of V. alginolyticus in response to the double increasing concentrations of BLFX.  We hypothesize that the achievement of the two objectives will tell us a new story on bacterial metabolic response to gradient increasing concentrations of antibiotic.

Effect of double increasing concentrations of BLFX on bacterial growth.  alginolyticus cells were cultured in LB medium separately with 0, 1/64, 1/32, 1/16, 1/8, 1/4 and 1/2 MIC of BLFX.  These cultures were harvested when the bacterial cell density in medium without the drug arrived at approximately 1.0 OD.  The resulting OD was negatively related to the double increasing concentrations of BLFX.  Compared with the BLFX-con0, bacterial survival rates were 98.5 ± 1.34, 97.2 ± 0.78, 95.0 ±0.39, 92.0±0.61, 87.8 ± 0.54, 83.4 ± 0.48 in BLFX-con64, BLFX-con32, BLFX-con16, BLFX-con8, BLFX-con4 and BLFX-2, respectively.  Growth ability of the bacterium was a negative correlation to the double increasing concentrations of the drug in medium.  Significant difference was found at P < 0.01 between them.  These results indicate an association of the inhibitory role in bacterial growth with BLFX concentrations.

 

Characteristics of dynamically metabolic response to double increasing concentrations of BLFX. To investigate a differential metabolome of V. alginolyticus in response to BLFX, GC/MS based metabolomics approach was used to identify metabolic responses to the double increasing concentrations of BLFX. Quantitative data throughout the entire time course were obtained for 89 compounds from 128 peaks after deletion of nine peaks that were below the limit of quantitation or too unstable to yield reliable information. OPLS-DA approach distributed the seven sample groups into different levels with the double increasing concentrations of BLFX in 3-D scatter plot, in which BLFX-con0 and BLFX-con64 were located in the bottom face and in the interface between the bottom and lateral faces, respectively.  The others were clearly separated from BLFX-con0 and the separation was trended with the double increasing BLFX concentrations in the lateral face. Dynamics of the 89 compounds was shown in a clustered heat-map format.  The format showed that BLFX-con0, BLFX-con64 and BLFX-con32, BLFX-con16, BLFX-con8 and BLFX-con4, and single BLFX-con2 were clustered together. In the first two clusters, there was significant distance between BLFX-con0 and the other two, and between BLFX-con4 and the other two, indicating the quantitative/qualitative pattern of metabolic changes in response to the double increasing BLFX concentrations.  These results show that the metabolic profiles have significantly adjusted as a result of the increased antibiotic stress.

Identification of dynamically differential metabolites in response to double increasing concentrations of BLFX. For determining the characteristic compounds in BLFX, the multi-class unpaired method of SAM was used to identify the metabolites whose concentration differed significantly among all samples collected across the double increasing concentrations of BLFX.  Despite these groups being only double-fold elevation of BLFX concentrations, they could be differentiated from SAM analyses. SAM identified 47, 64, 67, 56, 59 and 78 peaks representing 35, 48, 47, 48, 46 and 53 metabolites in BLFX-con64, BLFX-con32, BLFX-con16, BLFX-con8, BLFX-con4 and BLFX-2, respectively.  The 44 positively significant metabolites and 9 negatively significant metabolites from 78 peaks detected in BLFX-2, containing the all changes occurred in the other samples, in order of decreasing significance palmitelaidic acid, phosphoric acid, tetradecanoic acid, adenosine, morphinan, oleic acid as the first six metabolites.  Interestingly, ratios of metabolite areas with BLFX doses were not exactly proportional in every concentration for majority of the metabolites with differential levels, although changing tendency of the differential metabolites was detected.  These results indicate that BLFX-stressed V. alginolyticus showed a quantitative/qualitative change in the response. 

Then, PCA was separately used to identify significant differences in the data set.  The first six metabolites from factor weight were similar in the seven comparisons in order of decreasing significance L-proline, butanedioic acid, silanamine, phosphoric acid, L-alanine, glycine.  

Potential markers of interest were further extracted from S-plots constructed following analysis with OPLS-DA. Fourteen biomarkers were identified from the loading S-plots for metabolic pathway analysis since they showed statistically significant (P < 0.05) difference in abundance. Out of them, the first eight existed only in comparison cross the seven groups (BLFX-con all), in which negative regulation of phosphoric acid, palmitelaidic acid, glutamine, glycine, 5'-adenylic acid and 9H-purin-6-amine and positive regulation of butanedioic acid, L-proline, adenosine were characterized. VIP analysis confirmed the reliability of the first four biomarkers phosphoric acid, butanedioic acid, L-proline and palmitelaidic acid in the total comparison. The other four were validated in comparison between two groups. However, the regulation by each of these compounds did not run through whole responsive process to the double increasing BLFX concentrations. They showed an alternating renewal process of positive/negative regulation. For example, negative and positive regulations were detected in BLFX-con64 vs -con0 (-0.18), BLFX-con16 vs –con32 (-0.19) and BLFX-con32 vs –con64 (0.20), BLFX-con8 vs –con16 (0.53), BLFX-con4 vs –con8 (0.66), BLFX-con2 vs –con4 (0.86) for phosphoric acid, respectively. Generally, an alternating renewal positive/negative regulation was detected in first four comparisons, BLFX-con64 vs -con0, BLFX-con32 vs –con64, BLFX-con16 vs –con32 and BLFX-con8 vs –con16, and then changing tendency kept until the last comparison BLFX-con2 vs –con4 during the stress processes for each of the key metabolites. There were two change points, BLFX-con64 vs -con0, BLFX-con16 vs –con32, in the response process.  These results obtained from S-plot were consistent with the finding obtained from the clustering. These results not only indicate the importance in the structure of advance by degrees for details, but also show the quantitative/qualitative pattern of metabolic changes in response to the double increasing concentrations of BLFX.

Identification of differential metabolic pathways in response to double increasing concentrations of BLFX. Furthermore, ingenuity network analysis was used to perform more detailed analysis of pathways and networks influenced by the double increasing concentrations of BLFX.  Out of the significantly altered 77 metabolites, 66 were used for the analysis in deletion of 11 unclear pathway metabolites. To identify possible pathways that were affected by BLFX, metabolites contributing to the separation of control and groups with the double increasing BLFX concentrations were analyzed in structure of advance by degrees using MetPA.  The analysis revealed metabolites, which were identified together were important for the bacterial response to the drug, were responsible for 27-39 pathways.  When the impact-value threshold was set to 0.10, eight potential targets pathways were filtered out.  They were inositol phosphate metabolism, glycine, serine and threonine metabolism, alanine, aspartate and glutamate metabolism, pantothenate and CoA biosynthesis, arginine and proline metabolism, glyoxylate and dicarboxylate metabolism, tryptophan metabolism, methane metabolism.  However, the Raw p, the original p value calculated from the enrichment analysis, analysis showed an alternating renewal process of significant/insignificant changes in the double increasing concentrations of BLFX. For example, significant difference was detected in BLFX-con64 (0.001), BLFX-con32 (0.03) and BLFX-con8 (0.02), but not in BLFX-con16 (0.09), BLFX-con4 (0.32), BLFX-con2 (0.16) for glycine, serine and threonine metabolism (Table 3), suggesting an efficient strategy for the bacterium in response to BLFX through alternating renewal modification of pathways. 

 

Identification of promoting constructive metabolism in response to double increasing BLFX concentrations. We further detected bacterial dry weight and protein content for understanding whether the bacterium response to BLFX mainly through promoting constructive metabolism or catabolism.  First, bacterial dry weight was measured.  An increasing tendency with the double increasing BLFX concentrations was detected with correlation coefficient (P < 0.05).  Approximately 10 – 20 % dry weight was increased in medium with 1/8 – 1/2 MIC of BLFX.  Then, significantly elevated total protein content was determined at P < 0.01 with a positive correlation between the double increasing BLFX concentrations and protein content. These results indicate that V. alginolyticus enhances protein constructive metabolism in response to BLFX.     

 

In summary, our findings indicate that V. alginolyticus mounts an alternating renewal positive/negative regulation of metabolites and discontinuity modification of pathways for the increasing BLFX concentrations, forming characteristic changes of the quantitative/qualitative pattern.  These findings have significant implication in controlling BLFX-resistant V. alginolyticus by adjusting metabolites and metabolic pathways.  

This work was sponsored by grants from “973” project (2012CB114406) and NSFC projects (30972279, 40976080). Doctoral Fund of Ministry of Education of China (100171110029).