Assembling the anaerobic gamma-butyrobetaine to TMA metabolic pathway in Escherichia fergusonii and confirming its role in TMA production from dietary L-carnitine in murine models

ABSTRACT Trimethylamine-N-oxide (TMAO) is a major pro-atherogenic and pro-thrombotic metaorganismal molecule produced through the initial conversion of the dietary L-carnitine and other precursors into trimethylamine (TMA). We recently identified a dual-microbe anaerobic pathway for the metabolism of L-carnitine into TMA, in which the widely distributed cai operon in Enterobacteriaceae converts L-carnitine into gamma-butyrobetaine (γBB), followed by the degradation of γBB into TMA by the relatively rare gamma-butyrobetaine utilization (gbu) gene cluster present in Emergencia timonensis and few other related microbes. Studies of this pathway in animal models, however, have been limited by the lack of single microbes harboring the whole L-carnitine→γBB→TMA transformation pathway. Such recombinant microbes would both serve as a tool to further prove the contribution of this pathway to gut microbial TMA production and for future in vivo studies investigating the diet linkage to cardiovascular disease and the involvement of the TMAO pathway in this linkage. Here, we recapitulate the whole pathway in a single microbe by cloning the E. timonensis gbu gene cluster into Escherichia fergusonii, which naturally harbors the cai operon. We then show that the native E. timonensis GroES/GroEL-like proteins are needed for the proper functioning of the gbu cluster at 37°C. Finally, we demonstrate that inoculating germ-free mice with this recombinant E. fergusonii strain is sufficient to raise serum TMAO to pathophysiological levels upon dietary L-carnitine supplementation. The recombinant E. fergusonii strain developed will be a useful tool to facilitate future studies on the role of anaerobic gut microbial L-carnitine metabolism in cardiovascular and metabolic diseases. IMPORTANCE The key atherosclerotic TMAO originates from the initial gut microbial conversion of L-carnitine and other dietary compounds into TMA. Developing therapeutic strategies to block gut microbial TMA production needs a detailed understanding of the different production mechanisms and their relative contributions. Recently, we identified a two-step anaerobic pathway for TMA production from L-carnitine through initial conversion by some microbes into the intermediate γBB which is then metabolized by other microbes into TMA. Investigational studies of this pathway, however, are limited by the lack of single microbes harboring the whole pathway. Here, we engineered E. fergusonii strain to harbor the whole two-step pathway and optimized the expression through cloning a specific chaperone from the original host. Inoculating germ-free mice with this recombinant E. fergusonii is enough to raise serum TMAO to pathophysiological levels upon L-carnitine feeding. This engineered microbe will facilitate future studies investigating the contribution of this pathway to cardiovascular disease.


Chromosomal knock-in of gbu gene cluster in E. fergusonii
Synthetic codon-optimized genes corresponding to gbuA, B, C, and E genes from Emergencia timonensis SN18 were cloned in a suicide plasmid harboring a kanamycinresistant gene cassette, R6K origin, and sacB gene.The two homology recombination arms of the plasmid were designed to insert the operon coding for the gbu genes inbetween the native caiE, and the caiF genes so that the gbu operon was placed immediately downstream of the caiE gene, and in front of a predicated native downstream terminator.This location was chosen to allow for potential enhancement of the gbu operon expression by the upstream cai operon promotor.The suicide plasmid was propagated in E. coli S17-1λpir, then transferred to E. fergusonii through conjugation.The successful transconjugants were obtained through plating the conjugation mixture on LB plates supplemented with kanamycin and ampicillin, as E. fergusonii is naturally resistant to ampicillin.The purified transconjugants were then plated on LSW (1) agar plates supplemented with 10% sucrose to select for the successful knock-in mutants, which were then purified and verified through PCR, Sanger sequencing, and their ability to grow in the presence of ampicillin but not kanamycin.The composition of the LSW agar plates (per liter) was 10 g tryptone, 5 g yeast extract, 5 mL glycerol, 0.4 g NaCl, and 20 g agar.

Mass spectrometry analysis:
Hydrochloric acid at a final concentration of 60 mM was added to the bacterial culture samples immediately after collection to prevent the loss of TMA through evaporation.
1 µl of the prepared samples were injected onto LC column through Shimadzu autosampler (SIL-HTc) and metabolites were resolved on silica column (Luna, 00F-4274-B0, Phenomenx) with LC gradients generated from binary pumps (Shimadzu LC-20AD) connected to two solvents, A: 0.1% propionic acid in water; B: 0.1% acetic acid in methanol; starting from 0% B in the first 2 minutes, followed by linear increasing to 15% over 3 minutes, then to 100% over 5 minute, holding at 100% B for 3 minutes, then finally returning back to 0% B to equilibrate column for 3 minutes.The LC flow rate was 0.2 ml/min.The LC elutes were analyzed on API 5000 Mass spectrometer (AbSciex) with an electrospray ion source.Standards and internal standards were monitored in positive multiple reaction monitoring mode with parent to daughter transitions determined by standards, and MS parameters were optimized by individual standards.Standard curves were generated from serial dilutions of standards undergoing the same procedures as real samples.

Germ-free mouse colonization
All experiments involving mice were performed using protocols approved by the Cleveland Clinic Animal Care and Use Committee.Germfree female C57BL/6 mice were bred at the Cleveland Clinic gnotobiotic animal facilities in a controlled environment in plastic flexible film gnotobiotic isolators (Class Biologically Clean, Ltd., Madison, WI) under a strict 14h light/10h dark cycle, and received sterilized non-acidified water and food (LabDiet, Catalogue # 5010 -Laboratory Autoclavable Rodent Diet) ad libitum.This diet contains choline at 1800 ppm per manufacturer data sheet.The chow was not supplemented with any additional dietary precursors for TMA.For experiments, mice were housed in Allentown Sentry Sealed Positive Pressure cages (Allentown, NJ) and handled according to methods shown previously (4).
Frozen glycerol stocks of E. fergusonii recombinant strains were administered as ~0.2 mL oral gavage into 15-20 week old mice inside a biological safety cabinet.Mice were maintained on a sterilized diet and their drinking water was supplemented with kanamycin sulfate (50 µg/mL) and either L-carnitine (1.3% w/v) or ɣBB (1.6% w/v).Kanamycin was added to ensure the stability of the pGro plasmid in the recombinant E. fergusonii strains.
After 1 week, serum and urine samples were collected to measure TMAO, L-carnitine and ɣBB levels, and then tracer studies were performed.The investigator was not blinded to treatment groups to avoid cross contamination between groups.

In vivo tracer studies
After colonization with the recombinant E. fergusonii strains, mice received a 200 µl oral gavage containing 112 mM each of d3-L-carnitine, d9-γbb, and d6-choline in a biological safety cabinet.Blood was collected from the saphenous vein at the indicated time.Whole blood was centrifuged to isolate serum, which was stored at -80°C until analyzed.Serum levels of endogenous and stable isotope-labeled L-carnitine, γBB, choline, TMA, and TMAO were determined using LC-MS/MS.Laboratory personnel performing MS analyses were blinded to sample group allocation during analysis.