Lipid metabolic signatures deviate in sepsis survivors compared to non-survivors

Waqas Khaliq, Peter Großmann, Sophie Neugebauer, Anna Kleyman, Roberta Domizi, Sara Calcinaro, David Brealey, Markus Gräler, Michael Kiehntopf, Sascha Schäuble, Mervyn Singer, Gianni Panagiotou, Michael Bauer, Waqas Khaliq, Peter Großmann, Sophie Neugebauer, Anna Kleyman, Roberta Domizi, Sara Calcinaro, David Brealey, Markus Gräler, Michael Kiehntopf, Sascha Schäuble, Mervyn Singer, Gianni Panagiotou, Michael Bauer

Abstract

Sepsis remains a major cause of death despite advances in medical care. Metabolic deregulation is an important component of the survival process. Metabolomic analysis allows profiling of critical metabolic functions with the potential to classify patient outcome. Our prospective longitudinal characterization of 33 septic and non-septic critically ill patients showed that deviations, independent of direction, in plasma levels of lipid metabolites were associated with sepsis mortality. We identified a coupling of metabolic signatures between liver and plasma of a rat sepsis model that allowed us to apply a human kinetic model of mitochondrial beta-oxidation to reveal differing enzyme concentrations for medium/short-chain hydroxyacyl-CoA dehydrogenase (elevated in survivors) and crotonase (elevated in non-survivors). These data suggest a need to monitor cellular energy metabolism beyond the available biomarkers. A loss of metabolic adaptation appears to be reflected by an inability to maintain cellular (fatty acid) metabolism within a "corridor of safety".

Keywords: Beta-oxidation; Energy metabolism; Fatty acid metabolism; Metabolomics; Safety corridor; Sepsis.

Conflict of interest statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

© 2020 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology.

Figures

Graphical abstract
Graphical abstract
Fig. 1
Fig. 1
Septic-NS (non-survivor) patients have a distinct metabolic signature and are more diverse than other patient groups. (A) Samples are labelled by patient number and day. Biochemical parameters are treated as feature vectors and the pairwise Canberra distance between feature vectors is used as an input for principal component analysis (PCA). Septic-S (survivors) differ markedly from Septic-NS. (non-survivors). Q values were calculated with a balanced bootstrapped PERMANOVA (FDR corrected together with all metabolite groups and the biochemical parameter set) as the mean of 1000 repeats. Arrows are proportionally scaled for esthetic appearance. (B) Metabolite concentrations are treated as feature vectors and the pairwise Canberra distance between feature vectors is used as an input for PCA. The non-septic group is generally not discernible from Septic-S; both differ markedly from Septic-NS. Q values where calculated as above. (C) Beta diversity of septic and non-septic patients as a measure of group spread and variance. Sorted by median spread the groups have the order non-Septic-S < Septic-S < non-Septic-NS < Septic-NS. Significance was assessed by FDR corrected t-tests between all groups. Q values not shown did not reach statistical significance.
Fig. 2
Fig. 2
Statistical and machine learning analysis independently find C4 acylcarnitine and lysoPCs discriminative for survival from sepsis. (A) All metabolites and clinical parameters that differed significantly between Septic-S and Septic-NS, either overall on days 0–3 or at any specific day by ANOVA based on untransformed concentration values after FDR correction. The heatmap shows data between the 5th and 95th percentiles for each measurement. Grey spots mark unmeasured values. Metabolites in bold face differ also between Septic-NS and non-Septic-NS. (B/C) ROC curve and AUC values for test and validation sets after the two best features were selected by Tournament Leave Pair Out-Cross Validation-Recursive Feature Elimination (TLPOCV-RFE) using Random Forests (B) or linear Support Vector Machines (C).
Fig. 3
Fig. 3
Deviations from a corridor of safety are abundant in lipid species even in the absence of statistical differences. Time courses of NS patients at days 0–3 where concentrations are outside the Septic- and non-Septic-NS minimum to maximum range at any day (A) in at least 4 patients and (B) for lysoPC a C 24:0. The scale in (A) is pseudo-logarithmic. Metabolites that differ significantly between the septic groups were excluded.
Fig. 4
Fig. 4
Acylcarnitine concentrations that correlate between plasma and liver predict consistent regulation of beta oxidation. Several acylcarnitines show the same relative change between rat Septic-S and Septic-NS in both liver and plasma; this was used in the fitting of a kinetic model of patient mitochondrial lipid beta-oxidation. (A) The relative concentration change matches for carnitine, short-chain and short/medium-chain acylcarnitines (with the exception of C4 acylcarnitine, which is an intermediate not just in beta oxidation but also in the degradation of branched chain amino acids) at both 6 h and 24 h. Error bars show the 95% confidence interval of the ratio. Only acylcarnitines present in the kinetic model of beta-oxidation are shown. We used all (acyl-)carnitine ratios to fit the model. (B) The kinetic model fitted to patient plasma concentration ratios shows regulation of the enzyme concentrations of CPT2, MCAD, CROT and MSCHAD. (C) Production of Acetyl-CoA, NADH and FADH2 is reduced in Septic-NS vs Septic-S. (D) The regulation of enzymes of mitochondrial beta-oxidation is consistent with their arrangement in the pathway. Red enzymes are downregulated, green upregulated in Septic-NS compared to Septic-S. Red arrows show reduced flux in Septic-NS. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5
Fig. 5
The metabolic safety corridor concept. Metabolite concentrations falling outside a corridor of safety are associated with an increased risk of mortality.
Supplementary figure 1
Supplementary figure 1
Global metabolic signatures in metabolic groups. PCA biplots per metabolite group show the same order as that in Fig. 1B for lysophosphatidylcholins, phosphatydilcholines and sphingolipids. Arrows show correlations of biplot scores with single metabolites. Only the 5 strongest correlating metabolites are shown. Arrows are scaled for legibility. (A) C3-DC (C4-OH): Malonylcarnitine (Hydroxybutyrylcarnitine), C14:2: Tetradecadienoylcarnitine, C16: Palmitoylcarnitine, C18:1: Octadecenoylcarnitine, C18:2: Octadecadienoylcarnitine. (B) Asp: Asparagin, Met: Methionine, Trp: Tryptophan, Pro: Prolin, Val:Valin. (C) DOPA: Dihydroxyphenylalanine. (D) LysoPC a: acyl lysophosphatidylcholine. (E) PC aa: diacyl phosphatidylcholine, PC ae: acyl-alkyl phosphatidylcholine. (F) SM: sphingomyelin, SM (OH): hydroxysphingomyelin.
Supplementary figure 2
Supplementary figure 2
Acylcarnitine difference between Septic-S and Septic-NS increases with decreasing fatty acid chain length. Acylcarnitines show an increasing concentration difference between Septic-S and Septic-NS with decreasing chain length when mapped to their corresponding acyl-CoA in the KEGG pathway map of mitochondrial beta-oxidation (map00071). Left halves correspond to Septic-NS, right halves to Septic-S. Concentrations were z-scored, tanh-transformed to map into the [−1, +1]-interval and averaged by mean for both patient groups.
Supplementary figure 3
Supplementary figure 3
Time courses of individual metabolites showing significant differences between non-Septic-S and non-Septic-NS. Statistical significance is determined by repeated measures ANOVA. Lines indicate the mean. The grey section indicates the 25th and 75th percentile (grey line the mean) of considered metabolites in a healthy French population of 800 volunteers . Only day 0–3 data are shown.
Supplementary figure 4
Supplementary figure 4
Time courses of individual metabolites in Septic-S and Septic-NS showing significant differences between groups. Statistical significance is determined by repeated measures ANOVA. Lines indicate mean. Stars indicate significance in a Tukey HSD post-hoc test for a specific day. The grey section indicates the 25th and 75th percentile (grey line the mean) of considered metabolites in a healthy French population of 800 volunteers . Only day 0–3 data are shown.
Supplementary figure 5
Supplementary figure 5
Time courses of individual metabolites in Septic-NS and non-Septic-NS showing significant differences between groups. Statistical significance is determined by repeated measures ANOVA. Lines indicate mean. The grey section indicates the 25th and 75th percentile (grey line the mean) of considered metabolites in a healthy French population of 800 volunteers . Only days 0–3 are shown.
Supplementary figure 6
Supplementary figure 6
Time courses of individual biochemical parameters showing significant differences between Septic-S and Septic-NS. Statistical significance is determined by repeated measures ANOVA. Lines indicate mean.

References

    1. Singer M., Deutschman C.S., Seymour C.W., Shankar-Hari M., Annane D., Bauer M. The third international consensus definitions for sepsis and septic shock (Sepsis-3) JAMA. 2016;315(8):801.
    1. Fleischmann C., Scherag A., Adhikari N.K.J., Hartog C.S., Tsaganos T., Schlattmann P. Assessment of global incidence and mortality of hospital-treated sepsis. Current estimates and limitations. Am J Respiratory Crit Care Med. 2016;193(3):259–272.
    1. Prescott H.C., Angus D.C. Enhancing recovery from sepsis: a review. JAMA. 2018;319(1):62–75.
    1. Martin G.S., Mannino D.M., Moss M. The effect of age on the development and outcome of adult sepsis. Crit Care Med. 2006;34(1):15–21.
    1. Marriott I., Huet-Hudson Y.M. Sexual dimorphism in innate immune responses to infectious organisms. Immunol Res. 2006;34(3):177–192.
    1. Sinapidis D., Kosmas V., Vittoros V., Koutelidakis I.M., Pantazi A., Stefos A. Progression into sepsis: an individualized process varying by the interaction of comorbidities with the underlying infection. BMC Infect Dis. 2018;18(1):242.
    1. Rautanen A., Mills T.C., Gordon A.C., Hutton P., Steffens M., Nuamah R. Genome-wide association study of survival from sepsis due to pneumonia: an observational cohort study. Lancet Resp Med. 2015;3(1):53–60.
    1. Schrag SJ, Farley MM, Petit S, Reingold A, Weston EJ, Pondo T, et al. Epidemiology of invasive early-onset neonatal sepsis, 2005 to 2014. Pediatrics. 2016;138(6):e20162013-e.
    1. Angus D.C., van der Poll T. Severe sepsis and septic shock. N Engl J Med. 2013;369(9):840–851.
    1. Antti H., Fahlgren A., Näsström E., Kouremenos K., Sundén-Cullberg J., Guo Y. Metabolic profiling for detection of Staphylococcus aureus infection and antibiotic resistance. PLoS ONE. 2013;8
    1. Lv H., Liu L., Zhang Y., Song T., Lu J., Chen X. Ingenuity pathways analysis of urine metabonomics phenotypes toxicity of gentamicin in multiple organs. Mol BioSyst. 2010;6(10):2056–2067.
    1. Zhu J., Zhang M., Han T., Wu H., Xiao Z., Lin S. Exploring the biomarkers of sepsis-associated encephalopathy (SAE): metabolomics evidence from gas chromatography-mass spectrometry. Biomed Res Int. 2019
    1. Lee J, Banerjee D. Metabolomics and the Microbiome as Biomarkers in Sepsis. Crit Care Clin. 2020 Jan;36(1):105–113.
    1. Fernández-García M, Rojo D, Rey-Stolle F, García A, Barbas C. Metabolomic-Based Methods in Diagnosis and Monitoring Infection Progression. Experientia Supplementum. Springer International Publishing; 2018. pp. 283–315.
    1. Cambiaghi A., Pinto B.B., Brunelli L., Falcetta F., Aletti F., Bendjelid K. Characterization of a metabolomic profile associated with responsiveness to therapy in the acute phase of septic shock. Sci Rep. 2017;7(1):9748.
    1. Ferrario M., Cambiaghi A., Brunelli L., Giordano S., Caironi P., Guatteri L. Mortality prediction in patients with severe septic shock: a pilot study using a target metabolomics approach. Sci Rep. 2016;6(1):20391.
    1. Garcia-Simon M., Morales J.M., Modesto-Alapont V., Gonzalez-Marrachelli V., Vento-Rehues R., Jorda-Miñana A. Prognosis biomarkers of severe sepsis and septic shock by 1H NMR urine metabolomics in the intensive care unit. PLoS ONE. 2015;10(26565633)
    1. Liu Y., Hou J.-h., Li Q., Chen K.-j., Wang S.-N., Wang J.-m. Biomarkers for diagnosis of sepsis in patients with systemic inflammatory response syndrome: a systematic review and meta-analysis. SpringerPlus. 2016;5(1):2091.
    1. Seymour C.W., Yende S., Scott M.J., Pribis J., Mohney R.P., Bell L.N. Metabolomics in pneumonia and sepsis: an analysis of the GenIMS cohort study. Intensive Care Med. 2013;39:1423–1434.
    1. Langley R.J., Tsalik E.L., van Velkinburgh J.C., Glickman S.W., Rice B.J., Wang C. An integrated clinico-metabolomic model improves prediction of death in sepsis. Sci Transl Med. 2013;5(195) 195ra95-ra95.
    1. Bowden J.A., Heckert A., Ulmer C.Z., Jones C.M., Koelmel J.P., Abdullah L. Harmonizing lipidomics: NIST interlaboratory comparison exercise for lipidomics using SRM 1950 metabolites in frozen human plasma. J Lipid Res. 2017;58(12):2275–2288.
    1. Simón-Manso Y., Lowenthal M.S., Kilpatrick L.E., Sampson M.L., Telu K.H., Rudnick P.A. Metabolite profiling of a NIST standard reference material for human plasma (SRM 1950): GC-MS, LC-MS, NMR, and clinical laboratory analyses, libraries, and web-based resources. Anal Chem. 2013;85(24):11725–11731.
    1. Brealey D., Karyampudi S., Jacques T.S., Novelli M., Stidwill R., Taylor V. Mitochondrial dysfunction in a long-term rodent model of sepsis and organ failure. Am J Physiol-Regulatory, Integr Comp Physiol. 2004;286(3):R491–R497.
    1. Rudiger A., Dyson A., Felsmann K., CJ E., Taylor V., Hughes S. Early functional and transcriptomic changes in the myocardium predict outcome in a long-term rat model of sepsis. Clin Sci. 2013;124(6):391–401.
    1. Zolfaghari P.S., Pinto B.B., Dyson A., Singer M. The metabolic phenotype of rodent sepsis: cause for concern? Intensive Care Med Exp. 2013;1(1):6.
    1. van Maldegem B.T., Wanders R.J.A., Wijburg F.A. Clinical aspects of short-chain acyl-CoA dehydrogenase deficiency. J Inherit Metab Dis. 2010;33(5):507–511.
    1. Breiman L. Random forests. Machine Learn. 2001;45(1):5–32.
    1. Wright M.N., Ziegler A. ranger: a fast implementation of random forests for high dimensional data in C++ and R. J Stat Softw. 2017;77(1):1–17.
    1. Cortes C., Vapnik V. Support-vector networks. Machine Learn. 1995;20(3):273–297.
    1. Meier M.A., Ottiger M., Vögeli A., Steuer C., Bernasconi L., Thomann R. Activation of the tryptophan/serotonin pathway is associated with severity and predicts outcomes in pneumonia: results of a long-term cohort study. Clin Chem Lab Med. 2017;55:1060.
    1. Perez I.M., Airola A., Boström P.J., Jambor I., Pahikkala T. Tournament leave-pair-out cross-validation for receiver operating characteristic analysis. Stat Methods Med Res. 2019;28(10–11):2975–2991.
    1. Bagshaw S.M., Bellomo R., Jacka M.J., Egi M., Hart G.K., George C. The impact of early hypoglycemia and blood glucose variability on outcome in critical illness. Crit Care. 2009;13(3):R91.
    1. Tamhane A.C., Logan B.R. Finding the maximum safe dose level for heteroscedastic data. J Biopharm Stat. 2004;14(4):843–856.
    1. Pochini L., Scalise M., Galluccio M., Indiveri C. OCTN cation transporters in health and disease. J Biomol Screen. 2013;18(8):851–867.
    1. Feingold K.R., Moser A., Patzek S.M., Shigenaga J.K., Grunfeld C. Infection decreases fatty acid oxidation and nuclear hormone receptors in the diaphragm. J Lipid Res. 2009;50:2055–2063.
    1. Maitra U., Chang S., Singh N., Li L. Molecular mechanism underlying the suppression of lipid oxidation during endotoxemia. Mol Immunol. 2009;47:420–425.
    1. Carré J.E., Orban J.-C., Re L., Felsmann K., Iffert W., Bauer M. Survival in critical illness is associated with early activation of mitochondrial biogenesis. Am J Respir Crit Care Med. 2010;182(6):745–751.
    1. Weber M., Lambeck S., Ding N., Henken S., Kohl M., Deigner H.P. Hepatic induction of cholesterol biosynthesis reflects a remote adaptive response to pneumococcal pneumonia. FASEB J. 2012;26(6):2424–2436.
    1. Seymour C.W., Kennedy J.N., Wang S., Chang C.-C.H., Elliott C.F., Xu Z. Derivation, validation, and potential treatment implications of novel clinical phenotypes for sepsis. JAMA. 2019;321(20):2003–2017.
    1. Kiehntopf M., Nin N., Bauer M. Metabolism, metabolome, and metabolomics in intensive care: is it time to move beyond monitoring of glucose and lactate? Am J Respir Crit Care Med. 2013;187(9):906–907.
    1. Rivers E., Nguyen B., Havstad S., Ressler J., Muzzin A., Knoblich B. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–1377.
    1. PRISM Investigators Early, goal-directed therapy for septic shock. A patient-level meta-analysis. N Engl J Med. 2017;376(23):2223–2234.
    1. Casaer M.P., den Berghe G.V. Nutrition in the acute phase of critical illness. N Engl J Med. 2014;370(13):1227–1236.
    1. Schmerler D., Neugebauer S., Ludewig K., Bremer-Streck S., Brunkhorst F.M., Kiehntopf M. Targeted metabolomics for discrimination of systemic inflammatory disorders in critically ill patients. J Lipid Res. 2012;53:1369–1375.
    1. Dixon P.M., Wu L., Widrlechner M.P., Wurtele E.S. Iowa State University; IA: 2009. Weighted Distance Measures for Metabolomic Data.
    1. Fox J., Weisberg S. 2nd ed. Sage; Thousand Oaks CA: 2011. An R companion to applied regression.
    1. Hothorn T., Bretz F., Westfall P. Simultaneous inference in general parametric models. Biometrical J. 2008;50(3):346–363.
    1. Meyer D, Dimitriadou E, Hornik K, Weingessel A, Leisch F. e1071: Misc Functions of the Department of Statistics, Probability Theory Group (Formerly: E1071), TU Wien. 2018.
    1. Kanehisa M., Furumichi M., Tanabe M., Sato Y., Morishima K. KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res. 2016;45(D1):D353–D361.
    1. Luo W., Brouwer C. Pathview: an R/Bioconductor package for pathway-based data integration and visualization. Bioinformatics. 2013;29(14):1830–1831.
    1. Hoops S., Sahle S., Gauges R., Lee C., Pahle J., Simus N. COPASI – a COmplex PAthway SImulator. Bioinformatics. 2006;22(24):3067–3074.
    1. van Eunen K., Volker-Touw C.M.L., Gerding A., Bleeker A., Wolters J.C., van Rijt W.J. Living on the edge: substrate competition explains loss of robustness in mitochondrial fatty-acid oxidation disorders. BMC Biol. 2016;14(1):107.
    1. Trabado S., Al-Salameh A., Croixmarie V., Masson P., Corruble E., Fève B. The human plasma-metabolome: Reference values in 800 French healthy volunteers; impact of cholesterol, gender and age. PLoS ONE. 2017;12

Source: PubMed

Patrocinadores y Colaboradores

Condiciones médicas

Intervenciones de drogas

3
Suscribir