Extending dynamic and operational range of the biosensor responding to l-carnitine by directed evolution

Abstract

l-carnitine is a quaternary amine compound essential for eukaryotic metabolism. It is mainly involved in the oxidative decomposition of medium-and long-chain fatty acids and provides energy for the body. Therefore, it is widely used in health care and food additives. As a pivotal transcriptional activator of l-carnitine metabolism, CaiF is notably activated by crotonobetainyl-CoA, a key intermediate product in the carnitine metabolic pathway. Capitalizing on this mechanism, a sophisticated biosensor was ingeniously developed. Nevertheless, it is worth mentioning that the biosensor currently exhibits a relatively restricted detection range, which results in some specific limitations in practical application scenarios. In this paper, we constructed a biosensor based on CaiF and developed a strategy for modifying this biosensor. The structural configuration of CaiF was formulated by computer-aided design, and the DNA binding site was simulated, which was verified by alanine scanning. Functional Diversity-Oriented Volume-Conservative Substitution Strategy of the key sites of CaiF was conducted to extend the dynamic range of the biosensor. The biosensor based on CaiF^Y47W/R89A, which exhibited a considerably expanded concentration response range, from 10^-4 mM-10 mM, was obtained. The response range was 1000-fold wider and the output signal intensity was 3.3-fold higher to that of the control biosensor. These variants may have great value in improving the l-carnitine production process.

Keywords: Biosensor; CaiF; Detection range; Transcription factor; l-carnitine.

Fig. 1

Schematic of Endogenous Regulation of l-Carnitine in Escherichia coli. γ-BB, γ-butyrobetaine; l-car, l-Carnitine; Crot, crotonobetaine; γ-BB-CoA, γ-butyrobetainyl-CoA; l-car-CoA, l-carnitinyl-CoA; Crot-CoA, crotonobetainyl-CoA.

Fig. 2

A biosensor based on the transcription factor CaiF has been developed for the detection of l-carnitine. (a) The working principle of the biosensor. There are two reporter genes in the biosensor, sfGFP and mCherry (Down). (b) The biosensors corresponding to two different reporter genes were assessed using varying concentrations of l-carnitine. (c) Dose-response curves to exogenous l-carnitine for Sensor in BW25113 (DE3) ΔcaiA cultivated in glycerol M9 medium. The fluorescence was determined at each indicated effector concentration and normalized to the cell density. The curves were fitted to the Hill equation. The maximum normalized fluorescence signal observed during the cultivation process is shown. Error bars indicate the standard deviation of the mean of three replicates.

Fig. 3

Predicted structure of the CaiF-DNA complex. The three-dimensional structure of CaiF transcription factor bound to its target DNA sequence was modeled using the AlphaFold3 Server [27]. Among them, pink and purple are two reverse complementary sequences in the binding domain, 5′-CAATATTGAAA-3′ and 5′-TTTCAATATTG-3′, respectively, which are also the target sites of CaiF, so each DNA segment may require two CaiF binding. The key binding regions and residues are shown in the magnified view. Hydrogen bond interactions, hydrophobic interactions, and salt bridges are represented by blue solid lines, gray dashed lines, and yellow dashed lines, respectively. The DNA are shown in gray and the binding domain was bright orange. The residues are shown in magenta sticks.

Fig. 4

Interchain mutation analysis of CaiF. (a) The relationship between sensor mutants and concentration response. (b) Comparison of interaction networks before and after R75A mutation. Hydrogen bond and hydrophobic interactions are depicted by blue solid lines and yellow dashed lines, respectively. The DNA are shown in gray and the binding domain was bright orange, while residues are highlighted in magenta. (c) Root Mean Square Fluctuation (RMSF) analysis comparing residue dynamics in chain A before and after R75A mutation.

Fig. 5

Site-directed mutagenesis of CaiF was performed to obtain mutants with expanded response range. (a–f) The response range of the mutants (Y11, V42, N43, T46, Y47, R75) was evaluated after exogenous addition of l-carnitine concentrations of 10⁻⁴ mM, 10⁻² mM, 1 mM and 10² mM. (g) Dose-response curves to exogenous l-carnitine for mutants with expanded response range in E. coli BW25113 (DE3) ΔcaiA cultivated. The fluorescence was determined at each indicated effector concentration and normalized to the cell density. The curves were fitted to the Hill equation. The maximum normalized fluorescence signal observed during the cultivation process is shown. Error bars indicate the standard deviation of the mean of three replicates.

Fig. 6

Comparison of combinatorial mutant Y11L and Y47W. (a) The response values of wild type and mutants towards 10⁻⁴ mM, 10⁻² mM, 10 mM and 10² mM l-carnitine (b) The mutant obtained by superimposing alanine on the basis of Y11L. The response range of the mutant was evaluated after exogenous addition of l-carnitine concentrations of 10⁻⁶ mM–60 mM. (c) The mutant obtained by superimposing alanine on the basis of Y47W. The response range of the mutant was evaluated after exogenous addition of l-carnitine concentrations of 10⁻⁶ mM–60 mM. The maximum normalized fluorescence signal observed during the cultivation process is shown. Error bars indicate the standard deviation of the mean of three replicates.