Metabolomics study of metabolic variations in enterotoxigenic Escherichia coli-infected piglets

Abstract

This study aimed to explore the metabolic profiling in the serum of enterotoxigenic Escherichia coli (ETEC) infected piglets. Compared to control piglets, diarrheal piglets had higher contents of m-cresol and urocanic acid, but lower levels of fumaric acid, isoleucine, N-methyl-DL-alanine, 2-hydroxybutanoic acid, L-threose and putrescine, which are mainly associated with amino acid metabolism, energy metabolism and urea cycle. Compared to diarrheal piglets, recovered piglets had higher contents of oxoproline, proline, and ornithine, but lower contents of methyl phosphate, glycolic acid, myristic acid and azelaic acid, which are commonly involved in amino acid metabolism, glutathione synthesis and urea cycle. Collectively, the current study provides insights into metabolic alterations during ETEC infection and the recovery from ETEC induced diarrhea.

---

Introduction

Enterotoxigenic Escherichia coli (ETEC) is a common cause of piglet diarrhea,¹ of childhood diarrhea in resource-limited regions, and of diarrhea in adult travelers to these areas.² ETEC produces several virulence factors, including colonization factors (adhesins), and/or the heat-labile (LT) and heat-stable (ST) toxins, which cause the over-production of cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP), leading to diarrhea.³ We have growing knowledge about the host responses against ETEC infection in the last few years with systemic methods, such as transcriptomic analysis^4,5 and proteomics analysis.⁶ For example, with proteomics analysis, scientists have found that several proteins, including actin beta, heat stress proteins and transferrin, are promising candidate proteins for association with ETEC susceptibility.⁶ However, the host metabolic alterations after ETEC infection are unknown.

Increasing investigations have demonstrated that metabolomics is useful in discovering of molecular biomarkers in many fields,^7–9 and it has also been widely used to study the host metabolic changes during infection.^10–12 For example, with nuclear magnetic resonance (NMR) spectroscopy based metabolomics, the authors found that hepatitis B virus (HBV) infection up-regulates the biosynthesis of hexosamine and phosphatidylcholine, central carbon metabolism and nucleotide synthesis.¹⁰ Besides NMR based metabolomics, methods have been widely used in metabolomics studies include liquid chromatography mass spectrometry (LC-MS),¹³ capillary electrophoresis-mass spectrometry (CE-MS)¹⁴ and gas chromatography-mass spectrometry (GC-MS).¹⁵ As its high sensitivity and reproducibility, GC-MS has become one of the frequently used techniques in metabolomics study.^16–25

In current study, with ETEC infected piglets model, we performed a GC-MS-based metabolomics study to explore the piglet serum metabolic changes during ETEC infection and the recovery from diarrhea. To our knowledge, this is the first systematic analysis of metabolic changes in piglets in the context of ETEC infection.

Materials and methods

Bacterial strains

This study used the Escherichia coli F4-producing strain W25K (O149:K91, K88ac; LT, ST, EAST), which was originally isolated from a diarrheal piglet.²⁶

ETEC infection of piglets

This study was conducted according to the guidelines of the Institute of Subtropical Agriculture, Chinese Academy of Sciences²⁷ 51 Piglets (Landrace × Yorkshire; 18 days old) were purchased from ZhengDa Co., Chongqing, China and orally inoculated for five consecutive days with ETEC W25K at dose of 10¹⁰ CFUs per day. As a control, 10 piglets were orally infected with the same volume of LB medium. At day 6, the blood samples were collected from diarrheal piglets, recovered piglets, resistant piglets and control piglets (n = 6/group). All blood samples were collected through a jugular vein from all of the piglets, and serum was separated by centrifugation at 1500g for 10 min at 4 °C and stored at −20 °C until analysis.²⁸

Sample preparation for GC-MS analysis

GC-MS analysis was used to quantify the metabolites concentrations in piglet serum samples. Serum samples were thawed at room temperature and 100 μl of serum was transferred into 2 ml centrifuge tubes. 350 μl of 75% methanol was then added, followed by 50 μl of L-2-chlorophenylalanine (0.1 mg ml⁻¹ stock in dH₂O, CAS#: 103616-89-3, Hengbai biotech., Shanghai, China) as an internal quantitative standard and vortexed for 10 s. The mixture was homogenized for 5 min and subsequently centrifuged at 12 000 × g for 15 min at 4 °C, after which 0.4 ml of the supernatant fraction was transferred to a glass vial and dried in a vacuum concentrator without heating. 80 μl of methoxyamine amine salt (dissolved in pyridine, final concentration of 20 mg ml⁻¹) was then added, vortexed for 30 s, and incubated for 2 h at 37 °C. Finally, 100 μl BSTFA reagent (containing 1% TMCS, REGIS Technologies. Inc. USA) was added and incubated for 1 h at 70 °C.

GC-MS analysis

Samples were analyzed using an Agilent 7890 gas chromatograph system (Agilent 7890A, Agilent, USA) coupled with a Pegasus 4D time-of-flight mass spectrometer (LECO Chroma TOF PEGASUS 4D, LECO, USA). The system utilized a DB-5MS capillary column coated with 5% diphenyl cross-linked with 95% dimethylpolysiloxane (30 m × 250 μm inner diameter, 0.25 μm film thickness; J&W Scientific, Folsom, CA, USA). Helium gas was used as carrier gas with 1 ml min⁻¹ of gas flow rate through the column. Column temperature was initially kept at 80 °C for 2 min, then increased to 180 °C at a rate of 10 °C min⁻¹, and then increased to 295 °C at a rate of 40 °C min⁻¹, where it was held for 8 min. Temperatures for injection, transfer line, and ion source temperatures were 280, 270, and 220 °C, respectively. The mass spectrometry data were acquired in full-scan mode with the m/z range of 20–600 at a rate of 100 spectra per second after a solvent delay of 492 s.

Data analyses

The GC/MS raw were processed by Chroma TOF 4.3X software (LECO Corporation, USA) and LECO-Fiehn Rtx5 database for raw peaks extracting, data baselines filtering and calibration, peak alignment, deconvolution analysis, peak identification and peak area integration. Metabolites from the GC-MS spectra were identified with similarity larger than 700. The normalized data were imported into Simca-P software (version 11.0, Umetrics, Umea, Sweden), which was used for multivariate statistical analyses including a principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA), and pair-wise orthogonal projections to latent structures discriminant analyses (OPLS-DA).

Results

Metabolic profiles analyzed by GC-MS

Typical total ion chromatograms (TICs) of piglet serum samples from the four groups were shown in Fig. 1. 160 metabolites were identified from 565 peaks in the chromatograms, including amino acids, fatty acids, sugars, and organic acids.

Fig. 1 GC/MS TICs of piglet serum samples from diarrheal piglets (A1–A6), recovered piglets (B1–B6), control piglets (C1–C6) and resistant piglets (D1–D6). The ordinate shows the relative mass abundance and the abscissa shows the retention time.

Multivariate statistical analysis of the metabolomics data

In PCA score plot, diarrheal piglets and recovered piglets were separated from control piglets and resistant piglets, but no discernible clustering was observed between diarrheal piglets and recovered piglets (Fig. 2A). A PLS-DA was applied to better understand the different metabolic patterns. Quality of the resulting discriminant models was summarized in Table 1. Most of R² and Q² in pair-wise groups were larger than 0.5, suggesting that all models were robust and had good fitness and prediction. Diarrheal piglets and control piglets were clearly distinguished in PLS-DA plot (Fig. 2B). PLS-DA score plots (Fig. 2C) showed that diarrheal piglets had distinctive metabolic profiles from recovered piglets, and clear separation was also observed between diarrheal piglets and resistant piglets (Fig. 2D).

Fig. 2 PCA score plot derived from the GC-MS analysis of serum from diarrheal piglets (A), recovered piglets (B), control piglets (C) and resistant piglets (D). PLS-DA score plots for pair-wise comparisons between diarrheal piglets and control piglets (B), diarrheal piglets and recovered piglets (C), and diarrheal piglets and resistant piglets (D).

Table 1 Summary of the parameters for assessing modeling quality^a

Group	R^2X	R^2Y	Q^2
a R^2X cum and R^2Y cum are the cumulative modeled variation in X and Y matrix, respectively. Q^2 cum is the cumulative predicted variation in Y matrix.
A–B	0.395	0.94	0.687
A–C	0.345	0.963	0.621
A–D	0.528	0.862	0.559

---

Potential metabolite markers for diarrhea and recovery

The variable importance in the projection (VIP) statistic of the first principal component of OPLS-DA model (threshold > 1), together with the p-value of the Student's t-test (threshold < 0.05) were used for selecting significant variables responsible for group separation. The identified potential markers were listed in Table 2. In the evaluation of diarrheal piglets compared to control piglets, eight metabolites were significantly altered: m-cresol and urocanic acid were up-regulated, and fumaric acid, isoleucine, N-methyl-DL-alanine, 2-hydroxybutanoic acid, L-threose and putrescine were down-regulated. Compared to recovered piglets, diarrheal piglets had higher contents of methyl phosphate, glycolic acid, myristic acid and azelaic acid, but lower levels of oxoproline, proline, and ornithine. Compared to resistant piglets, diarrheal piglets had higher contents of urocanic acid and beta-alanine, and lower levels of fumaric acid.

Table 2 List of differential metabolites among different groups

Var id (primary)	Peak	RT	Count	Mass	VIP	p-Value	Fold change^d
a Diarrheal piglets vs. recovered piglets.b Diarrheal piglets vs. control piglets.c Diarrheal piglets vs. resistant piglets.d In fold change value less than 1 means the level of metabolite decreases in diarrheal piglets compared to the corresponding piglets, while more than 1 means the content of metabolite increases in diarrheal piglets compared to the corresponding piglets.
5	Oxoproline^a	9.87	24	156	2.26	0.007	0.45
14	Fumaric acid^b	9.67	24	245	1.89	0.048	0.79
14	Fumaric acid^c	9.67	24	245	1.87	0.001	0.71
27	Urocanic acid^b	9.31	24	180	1.78	0.035	2.04
27	Urocanic acid^c	9.31	24	180	1.45	0.042	1.96
46	Proline^a	8.95	24	142	2.19	0.011	0.52
51	Isoleucine^b	8.82	24	158	1.66	0.014	0.51
110	Beta-alanine^c	7.43	24	102	1.00	0.004	1.92
122	Methyl phosphate^a	7.19	24	241	1.99	0.026	1.68
125	N-Methyl-DL-alanine^b	7.14	24	130	1.31	0.002	0.50
133	m-Cresol^b	6.91	24	165	1.30	0.012	5.98
152	2-Hydroxybutanoic acid^b	6.46	24	131	1.29	0.001	0.21
179	Glycolic acid^a	5.73	24	147	1.86	0.041	1.32
219	L-Threose^b	24.4	24	166	1.07	0.007	0.36
225	Putrescine^b	23.9	24	174	1.07	0.005	0.39
358	Myristic acid^a	16.2	24	132	2.15	0.013	1.68
376	Ornithine^a	15.7	24	142	2.00	0.025	0.52
382	Azelaic acid^a	15.4	20	55	2.04	0.021	1.78

---

Discussions

In this study, we found that ETEC induced diarrhea promotes the serum metabolic perturbations based on the GC-MS-based metabolomics investigation. 16 metabolites were found related to ETEC induced diarrhea and/or recovery from ETEC induced diarrhea.

Compared to control piglets, diarrheal piglets have higher contents of urocanic acid and m-cresol. Urocanic acid is produced by deamination of histidine with the help of histidine ammonialyase, and is transformed to 4-imidazolone-5-propionic acid and glutamic acids with the help of urocanase in the liver.²⁹ Increased contents of urocanic acid in diarrheal piglets may indicate ETEC infection affects the histidine metabolism in piglets. m-cresol is an isomer of p-cresol and o-cresol. The increase in the contents of m-cresol may suggest the growth of the intestinal microbes and the activation of the intestinal microfloral metabolism because p-cresol is produced with the metabolism of the intestinal microbes.^30,31 Compared to control piglets, fumaric acid, isoleucine, 2-hydroxybutanoic acid, and putrescine were down-regulated in diarrheal piglets. Fumaric acid is an important intermediate in energy generation; decreased concentrations of fumaric acid may indicate energy metabolism is altered in ETEC induced diarrhea. Isoleucine is a branched-chain amino acid, which is supplemented by food uptake. Decrease in isoleucine concentration is possibly due to difficulty in food uptake as a result of small intestine damage³² and increased metabolic utilization during ETEC infection. Indeed, ETEC infection inhibits the intestinal thiamin uptake by secreting heat-labile toxins and decreasing expression of intestinal thiamin transporters.³³ Under oxidative stress, the supply of cysteine for glutathione synthesis becomes limiting, diverting homocysteine from the transmethylation pathway to the transsulfuration pathway to form cysteine, during which 2-hydroxybutyric acid is released as a byproduct.³⁴ Thus, lower levels of 2-hydroxybutanoic acid may suggest that the glutathione synthesis is affected during ETEC induced diarrhea. Lower putrescine contents in diarrheal piglets reflect at least in part a limited capacity of the urea cycle during ETEC infection because putrescine is synthesized from L-ornithine through ornithine decarboxylase.^35,36

Compared to diarrheal piglets, recovered piglets have higher contents of oxoproline, ornithine and proline. Oxoproline is related to oxidative stress because 5-oxoproline is a byproduct of glutathione synthesis and has been proposed as a biomarker of glutathione status.^37,38 Increased levels of oxoproline in recovered piglets indicate the glutathione synthesis increases in recovered piglets. Ornithine is an intermediate of the urea cycle, during which arginine is converted into urea and ornithine by the arginase.³⁹ As mentioned above, diarrheal piglets may have limited capacity of the urea cycle, higher contents of ornithine in recovered piglets may suggest the capacity of the urea cycle is critical for the recovery from diarrhea. Besides the synthesis of polyamines, ornithine is important precursor for proline synthesis,⁴⁰ which may be an explanation for higher contents of proline are found in recovered piglets. Proline is beneficial for the immune response for host against pathogens infection.^41,42 Thus, the increased contents of proline may suggest that proline promotes the recovery from ETEC induced diarrhea.

Recovered piglets have lower contents of glycolic acid, azelaic acid and myristic acid, compared to diarrheal piglets. Glycolic acid is a known product of sugar and of ribose oxidation.^43,44 Azelaic acid is a typical oxidation product from the most abundant unsaturated lipid residues of triglycerides and phospholipids.⁴⁵ The decreased content of glycolic acid, azelaic acid and myristic acid in recovered piglets are suggesting the recovery from diarrhea is associated with the change of metabolism of these substances.

In summary, differentially expressed metabolites between diarrheal piglets and control piglets, and between diarrheal piglets and recovered piglets indicates that the metabolic pathways, like amino acid metabolism, glutathione synthesis and urea cycle, are commonly associated with the pathogenesis of ETEC induced diarrhea, and the recovery from diarrhea.

Potential conflicts of interest

None.

Author contributions

WR, TL and YY designed the experiment. WR, JY, WG and SC performed the experiment. WR, JD, GL and NL analyzed the data. WR, YP and YY wrote the manuscript.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (No. 31330075, 31110103909, 31272463, and 31472106), National Basic Research Program of China (2013CB127301) and National Scientific and Technology Support Project (2013BAD21B04).

References

1. B. Jonach, M. Boye, A. Stockmarr and T. K. Jensen, BMC Vet. Res., 2014, 10, 68.
2. Y. Zboromyrska, J. C. Hurtado, P. Salvador, M. J. Alvarez-Martinez, M. E. Valls, J. Mas, M. A. Marcos, J. Gascon and J. Vila, Clin. Microbiol. Infect., 2014, 20, O753–O759.
3. J. M. Fleckenstein, P. R. Hardwidge, G. P. Munson, D. A. Rasko, H. Sommerfelt and H. Steinsland, Microbes Infect., 2010, 12, 89–98.
4. T. A. Niewold, J. van der Meulen, H. H. Kerstens, M. A. Smits and M. M. Hulst, Vet. Microbiol., 2010, 141, 110–114.
5. C. Zhou, Z. Liu, J. Jiang, Y. Yu and Q. Zhang, BMC Genomics, 2012, 13, 330.
6. Z. Wu, R. Xia, X. Yin, Y. Huo, G. Zhu, S. Wu and W. Bao, PLoS One, 2015, 10, e0127164.
7. J. C. Lindon, E. Holmes and J. K. Nicholson, FEBS J., 2007, 274, 1140–1151.
8. D. G. Robertson, M. D. Reily and J. D. Baker, J. Proteome Res., 2007, 6, 526–539.
9. J. Zhao, Y. H. Jung, C. G. Jang, K. H. Chun, S. W. Kwon and J. Lee, Sci. Rep., 2015, 5, 8890.
10. H. Li, W. Zhu, L. Zhang, H. Lei, X. Wu, L. Guo, X. Chen, Y. Wang and H. Tang, Sci. Rep., 2015, 5, 8421.
11. D. J. Creek, M. Mazet, F. Achcar, J. Anderson, D. H. Kim, R. Kamour, P. Morand, Y. Millerioux, M. Biran, E. J. Kerkhoven, A. Chokkathukalam, S. K. Weidt, K. E. Burgess, R. Breitling, D. G. Watson, F. Bringaud and M. P. Barrett, PLoS Pathog., 2015, 11, e1004689.
12. J. S. Ng Hublin, U. Ryan, R. Trengove and G. Maker, PLoS One, 2013, 8, e77803.
13. T. P. Sangster, J. E. Wingate, L. Burton, F. Teichert and I. D. Wilson, Rapid Commun. Mass Spectrom., 2007, 21, 2965–2970.
14. S. Ullsten, R. Danielsson, D. Backstrom, P. Sjoberg and J. Bergquist, J. Chromatogr. A, 2006, 1117, 87–93.
15. K. Yu, G. Sheng, J. Sheng, Y. Chen, W. Xu, X. Liu, H. Cao, H. Qu, Y. Cheng and L. Li, J. Proteome Res., 2007, 6, 2413–2419.
16. Q. He, X. Kong, G. Wu, P. Ren, H. Tang, F. Hao, R. Huang, T. Li, B. Tan, P. Li, Z. Tang, Y. Yin and Y. Wu, Amino Acids, 2009, 37, 199–208.
17. Q. He, Y. Yin, Z. Feng, X. Kong, G. Wu and P. Ren, Front. Biosci., 2011, 16, 2451–2460.
18. Q. He, P. Ren, X. Kong, W. Xu, H. Tang, Y. Yin and Y. Wang, Mol. BioSyst., 2011, 7, 2147–2155.
19. Q. He, H. Tang, P. Ren, X. Kong, G. Wu, Y. Yin and Y. Wang, J. Proteome Res., 2011, 10, 5214–5221.
20. Q. He, P. Ren, X. Kong, Y. Wu, G. Wu, P. Li, F. Hao, H. Tang and Y. Yin, J. Nutr. Biochem., 2012, 23, 133–139.
21. Z. Ruan, M. Yang, Y. Zhou, Y. Wen, S. Ding, G. Liu, X. Wu, P. Liao, Z. Deng, H. Assaad, G. Wu and Y. Yin, Amino Acids, 2014, 46, 2219–2229.
22. Z. Ruan, M. Yang, Y. Zhou, Y. Wen, S. Ding, G. Liu, X. Wu, P. Liao, Z. Deng, H. Assaad, G. Wu and Y. Yin, Amino Acids, 2013, 45, 877–887.
23. Z. Ruan, M. Yang, Y. Zhou, Y. Wen, S. Ding, G. Liu, X. Wu, P. Liao, Z. Deng, H. Assaad, G. Wu and Y. Yin, Amino Acids, 2014, 46, 2681–2691.
24. M. Wu, X. Xiao, W. Ren and Y. Yin, PLoS One, 2014, 9, e0113687.
25. H. Xiao, M. Wu, F. Shao, B. Tan and Y. Yin, J. Anim. Sci., 2015, 54, 8229–8239.
26. W. Ren, G. Liu, J. Yin, S. Chen, T. Li, X. Kong, Y. Peng, Y. Yin and P. R. Hardwidge, Genome Announc., 2014, 2(3), e00593.
27. F. Yin, Z. Zhang, J. Huang and Y. Yin, Br. J. Nutr., 2010, 103, 1404–1412.
28. J. Duan, J. Yin, M. Wu, P. Liao, D. Deng, G. Liu, Q. Wen, Y. Wang, W. Qiu, Y. Liu, X. Wu, W. Ren, B. Tan, M. Chen, H. Xiao, L. Wu, T. Li, C. M. Nyachoti, O. Adeolaand and Y. Yin, PLoS One, 2014, 9, e112357.
29. A. H. Mehler and H. Tabor, J. Biol. Chem., 1953, 201, 775–784.
30. T. A. Clayton, D. Baker, J. C. Lindon, J. R. Everett and J. K. Nicholson, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 14728–14733.
31. K. Akira, H. Hichiya, M. Morita, A. Shimizu and H. Mitome, J. Pharm. Biomed. Anal., 2013, 85, 155–161.
32. W. Ren, J. Yin, J. Duan, G. Liu, X. Zhu, S. Chen, T. Li, S. Wang, Y. Tang and P. R. Hardwidge, Microbes Infect., 2014, 16, 954–961.
33. A. Ghosal, N. S. Chatterjee, T. Chou and H. M. Said, Am. J. Physiol.: Cell Physiol., 2013, 305, C1185–C1191.
34. A. Pechlivanis, A. C. Chatziioannou, A. S. Veskoukis, D. Kouretas, V. Mougios and G. A. Theodoridis, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2014, 966, 127–131.
35. J. Hadrevi, B. Ghafouri, A. Sjors, H. Antti, B. Larsson, A. G. Crenshaw, B. Gerdle and F. Hellstrom, Eur. J. Appl. Physiol., 2013, 113, 2977–2989.
36. H. Kotarsky, M. Keller, M. Davoudi, P. Leveen, R. Karikoski, D. P. Enot and V. Fellman, PLoS One, 2012, 7, e41156.
37. A. Z. Fenves, H. M. Kirkpatrick 3rd, V. V. Patel, L. Sweetman and M. Emmett, Clin. J. Am. Soc. Nephrol., 2006, 1, 441–447.
38. S. Geenen, F. B. du Preez, J. L. Snoep, A. J. Foster, S. Sarda, J. G. Kenna, I. D. Wilson and H. V. Westerhoff, Biochim. Biophys. Acta, 2013, 1830, 4943–4959.
39. K. A. Wijnands, T. M. Castermans, M. P. Hommen, D. M. Meesters and M. Poeze, Nutrients, 2015, 7, 1426–1463.
40. G. Wu, F. W. Bazer, M. C. Satterfield, X. Li, X. Wang, G. A. Johnson, R. C. Burghardt, Z. Dai, J. Wang and Z. Wu, Amino Acids, 2013, 45, 241–256.
41. W. Ren, M. Wu, W. Luo, R. Huang, Y. Yin, Y. Li, T. Li and X. Yu, Br. J. Nutr., 2013, 110, 1492–1499.
42. W. Ren, L. Zou, Z. Ruan, N. Li, Y. Wang, Y. Peng, G. Liu, Y. Yin, T. Li, Y. Hou and G. Wu, Amino Acids, 2013, 45, 555–561.
43. K. D. Sugden and K. E. Wetterhahn, Chem. Res. Toxicol., 1997, 10, 1397–1406.
44. W. von Schonfels, O. von Kampen, E. Patsenker, F. Stickel, B. Schniewind, S. Hinz, M. Ahrens, K. Balschun, J. H. Egberts, K. Richter, A. Landrock, B. Sipos, O. Will, P. Huebbe, S. Schreiber, M. Nothnagel, C. Rocken, G. Rimbach, T. Becker, J. Hampe and C. Schafmayer, PLoS One, 2013, 8, e72022.
45. E. N. Frankel, Prog. Lipid Res., 1984, 23, 19.

---