1
APPLICATION OF 1 H NMR METABOLOMICS TO A MURINE MODEL OF INFLAMMATORY ARTHRITIS Reza Dowlatabadi 1,2 , B. Joan Miller 3 , Frank R. Jirik 3 , Hans J. Vogel 1 and Aalim M. Weljie 1,4 1 Metabolomics Research Centre, U. of Calgary, Canada; 2 Current Address: Dept. of Med. Chem., Faculty of Pharmacy, Tehran U. of Med. Sci., Iran; 3 Biochem. and Molecular Biology, U. of Calgary; 4 Chenomx Inc., Edmonton, Alberta, Canada. [email protected] Introduction Acknowledgements Rheumatoid arthritis, a debilitating, systemic inflammatory joint disease impacting 1-2% of the population, may be accompanied by alterations in specific metabolites. As an initial approach to investigating this possibility in a well-defined system we selected a murine model of rheuma- toid arthritis, the KBxN mouse. In this transgenic model, a systematic inflammatory response is generated towards the ubiquitously expressed glucose-6-phosphate isomerase enzyme. Here we examine the metabolite profiles of KBxN mice and contextualize them with respect to rheu- matoid arthritis mechanisms (Figure 1). Methods - Metabolite profiling of sera from mice models of human disease is a viable way to understand patho- logical mechanisms, and provide a means for evaluating the function of disease modifying drugs - In the case of the KBxN mouse model, gender has little effect on an appropriately developed multi- variate model. - Some of the metabolic changes are likely related to dietary considerations (e.g. lipid metabolism), al- though a number of specific inflammatory biomarkers are also evident - Our results attest not only to the complexity of systemic inflammatory responses, but also the power of the experimental approach in being able to reveal such a wide variety of biomarkers. choline, acetylcarnatine, threonine , hypoxanthine, uridine Conclusions Lipid metabolism Glycerol, Nucleic acid metabolism Xanthine Reactive oxygen species (ROS) metabolism Taurine, Glycine Methylation Methionine Immune response in macrophages Glycine (via glycine-gated chlorine channels) Figure 2: Paw swelling evident in a) the genetic parent strain and b) the KBxN mouse model of arthritis. Normal Arthritis 2-Hydroxybutryate TMAO Uridine Glycerol Choline O-Acetylcarnitine Xanthine Glycine Creatine Glutamate Hypoxanthine 1-Methylhistidine 2-Oxoglutarate Methionine Taurine Threonine Elevated in Arthritis Decreased from Normal Selected Pathway Implications Sera from arthritic populations of KBxN mice that are genetically-predisposed to arthritis (N=15), as well as healthy parent strain population (N=22) were analyzed using ultrafiltration followed by 1 H NMR spectroscopy. A “Targeted Profiling” approach [1] was used to identify and quantify 54 metabolites using Chenomx NMR Suite (Chenomx Inc, Edmonton Alberta). Metabolite identification was confirmed using 2D methods and sample spiking. Subsequent multivariate analysis was performed using SIMCA-P+ (Umetrics, Sweden) to build an orthongonal par- tial lease squares discriminant analysis (OPLS-DA) model. The model was verified using a cross validation ap- proach, and a number of key metabolites identified as described below. Figure 1: Schematic respresentation of selected mechanisms of rheumatoid arthritis. Stimulation of T cells occurs upon presentation of a self-antigen, which leads to macrophage activation. Cytokine release is implicated in many damage pathways. The autoimmune response leads to joint swelling and eventually degradation of the bone morphology. Results Figure 3: Weight gain in the final six weeks of both arthritc and normal mice. In all cases, the arthritic mice gained less overall weight, and demonstrated a distinct plateau at < 25 g. Weight of arthritc and normal mice over time Scores plot from multivariate statistical modeling using OPLS Loadings plot from multivariate statistical modeling using OPLS Figure 5: Loadings plot from the OPLS analysis of the metabolite concentrations obtained using the targeted profil- ing approach. Metabolites are highlighted either as reduced from the normal case (black) or elevated in the arthritis case (red). Metabolites were chosen if they had a score > 1 in the variable importance plot of the overall model. Summary of important metabolites Potential physiological role of metabolites We thank Glen MacInnis for preparing and acquiring NMR data for several of the serum spectra. This work was supported by Genome Alberta/Canada as part of the Metabolomics Toolbox Project and the Arthritis Society of Canada. AMW was the recipient of an Alberta Ingenuity Industrial Fellowship. -0.3 -0.2 -0.1 -0.0 0.1 0.2 0.3 -0.3 -0.2 -0.1 -0.0 0.1 0.2 p[2]O p[1]P 2-Hy droxybutyrate 2-Ox oglutarate 3-Hy droxybutyrate 3-Methyl-2-oxovalarate Acetat e Adenosine Alanine Allantoi n Arginine Asparagine Aspartate Butyrate Carnitin e Cholin e Citrat e Citrulline Creatin e Cy tidine Formate Fu marate Glucos e Glutam ate Glutam ine Glycerol Glycine Histidine Hypoxanthine Isobutyrate Isoleucin e Lactat e Leucin e Lysine Methionine Nicotinate O-Acet ylcarnitine Phosphocholine-like Ornithine Pheny lalanine Prolin e Py ruvate Serine Suberat e Succinat e Taurin e Threonine TMAO Tr yptophan Ty rosine Uridine Valerate Valin e Xanthine trans-Aconitate 1-Methylhistidine Reduced from Normal Higher in Arthritis Figure 4: Scores plot from the OPLS analysis of the metabolite concentrations obtained using the targeted profiling ap- proach. A) Coloured according to type of mouse and B) coloured according to gender. Note that the first component in OPLS (x-axis) provides information on class separation, and as a result only the first component is considered for interpre- tation of metabolites relevant to arthritis. The first orthogonal component shows preliminary evidence of gender separation. -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 -4 -3 -2 -1 0 1 2 3 4 t[2]O t[1]P -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 -4 -3 -2 -1 0 1 2 3 4 t[2]O t[1]P Normal Female Male A) B) Synovial fluid Cartilage Pannus Bone Collagenase and Metalloproteinase mediated damage SCF Reactive oxygen species (ROS) mediated damage IL-1, IL-6, TNF-α, other cytokines, his- tamines, tryptase, chymase Neutrophil Recruitment Synovial Mast Cells Macrophage Activation Fibroblast activation and hyperplasia T Cell Stimulation Chondrocyte Activation Requires de novo pyramidine synthesis Methylation Cytokine and ROS release is blunted in the glycine-gated choride channed of white blood cells Taurine and Taurine - chloramine are known anti-inflammatory compounds Mainnemare et al., (2004) J. Dent. Res. 83:823 Synovial fluid Reactive oxygen species (ROS) mediated damage Neutrophil Recruitment Macrophage Activation T Cell Stimulation Collagenase and Metalloproteinase mediated damage Wheeler et al., (1999) CMLS, 56:843. Bottiglieri (2002) Am. J. Clin. Nutr., 76(sup):1151S.

APPLICATION OF 1H NMR METABOLOMICS TO A - ePosters · 2014. 2. 3. · (Chenomx Inc, Edmonton Alberta). Metabolite identification was confirmed using 2D methods and sample spiking

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

Page 1: APPLICATION OF 1H NMR METABOLOMICS TO A - ePosters · 2014. 2. 3. · (Chenomx Inc, Edmonton Alberta). Metabolite identification was confirmed using 2D methods and sample spiking

APPLICATION OF 1H NMR METABOLOMICS TO A MURINE MODEL OF INFLAMMATORY ARTHRITIS Reza Dowlatabadi1,2, B. Joan Miller3, Frank R. Jirik3, Hans J. Vogel1 and Aalim M. Weljie1,4

1Metabolomics Research Centre, U. of Calgary, Canada; 2Current Address: Dept. of Med. Chem., Faculty of Pharmacy, Tehran U. of Med. Sci., Iran; 3Biochem. and Molecular Biology, U. of Calgary; 4Chenomx Inc., Edmonton, Alberta, Canada. [email protected]

Introduction

Acknowledgements

Rheumatoid arthritis, a debilitating, systemic inflammatory joint disease impacting 1-2% of the population, may be accompanied by alterations in specific metabolites. As an initial approach to investigating this possibility in a well-defined system we selected a murine model of rheuma-toid arthritis, the KBxN mouse. In this transgenic model, a systematic inflammatory response is generated towards the ubiquitously expressed glucose-6-phosphate isomerase enzyme. Here we examine the metabolite profiles of KBxN mice and contextualize them with respect to rheu-matoid arthritis mechanisms (Figure 1).

Methods

- Metabolite profiling of sera from mice models of human disease is a viable way to understand patho-logical mechanisms, and provide a means for evaluating the function of disease modifying drugs - In the case of the KBxN mouse model, gender has little effect on an appropriately developed multi-variate model. - Some of the metabolic changes are likely related to dietary considerations (e.g. lipid metabolism), al-though a number of specific inflammatory biomarkers are also evident - Our results attest not only to the complexity of systemic inflammatory responses, but also the power of the experimental approach in being able to reveal such a wide variety of biomarkers.

choline , acetylcarnatine , threonine

, hypoxanthine, uridine

Conclusions

Lipid metabolism – Glycerol,

Nucleic acid metabolism – Xanthine

Reactive oxygen species (ROS) metabolism – Taurine, Glycine

Methylation – Methionine

Immune response in macrophages – Glycine (via glycine - gated chlorine channels)

Figure 2: Paw swelling evident in a) the genetic parent strain and b) the KBxN mouse model of arthritis.

Normal Arthritis

2-Hydroxybutryate TMAO Uridine Glycerol Choline O-Acetylcarnitine

Xanthine Glycine Creatine Glutamate Hypoxanthine 1-Methylhistidine 2-Oxoglutarate Methionine Taurine Threonine

Elevated in Arthritis Decreased from Normal

Selected Pathway Implications

Sera from arthritic populations of KBxN mice that are genetically-predisposed to arthritis (N=15), as well as healthy parent strain population (N=22) were analyzed using ultrafiltration followed by 1H NMR spectroscopy. A “Targeted Profiling” approach [1] was used to identify and quantify 54 metabolites using Chenomx NMR Suite (Chenomx Inc, Edmonton Alberta). Metabolite identification was confirmed using 2D methods and sample spiking. Subsequent multivariate analysis was performed using SIMCA-P+ (Umetrics, Sweden) to build an orthongonal par-tial lease squares discriminant analysis (OPLS-DA) model. The model was verified using a cross validation ap-proach, and a number of key metabolites identified as described below.

Figure 1: Schematic respresentation of selected mechanisms of rheumatoid arthritis. Stimulation of T cells occurs upon presentation of a self-antigen, which leads to macrophage activation. Cytokine release is implicated in many damage pathways. The autoimmune response leads to joint swelling and eventually degradation of the bone morphology.

Results

Figure 3: Weight gain in the final six weeks of both arthritc and normal mice. In all cases, the arthritic mice gained less overall weight, and demonstrated a distinct plateau at < 25 g.

Weight of arthritc and normal mice over time

Scores plot from multivariate statistical modeling using OPLS

Loadings plot from multivariate statistical modeling using OPLS

Figure 5: Loadings plot from the OPLS analysis of the metabolite concentrations obtained using the targeted profil-ing approach. Metabolites are highlighted either as reduced from the normal case (black) or elevated in the arthritis case (red). Metabolites were chosen if they had a score > 1 in the variable importance plot of the overall model.

Summary of important metabolites Potential physiological role of metabolites

We thank Glen MacInnis for preparing and acquiring NMR data for several of the serum spectra. This work was supported by Genome Alberta/Canada as part of the Metabolomics Toolbox Project and the Arthritis Society of Canada. AMW was the recipient of an Alberta Ingenuity Industrial Fellowship.

-0.3

-0.2

-0.1

-0.0

0.1

0.2

0.3

-0.3 -0.2 -0.1 -0.0 0.1 0.2

p[2]O

p[1]P

2-Hydroxybutyrate

2-Ox oglutarate

3-Hydroxybutyrate

3-Methyl-2-oxovalarate

Acetate

Adenosine

Alanine

Allantoin

Arginine

Asparagine

Aspartate

Butyrate

Carnitin e

Choline

Citrate

Citrulline

Creatin e

Cy tidine

Formate

Fumarate

Glucos e

Glutamate

Glutamine

Glycerol

Glycine

Histidine

Hypoxanthine

Isobutyrate

Isoleucin e

Lactat e

Leucin e

LysineMethionine

Nicotinate

O-Acet ylcarnitine

Phosphocholine-like

Ornithine

Pheny lalanineProline

Py ruvate

SerineSuberat e

Succinat e

Taurin eThreonine

TMAO

Tryptophan

Ty rosine Uridine

Valerate

Valin e

Xanthine

trans-Aconitate

1-Methylhistidine

Reduced from Normal

Higher in Arthritis

Figure 4: Scores plot from the OPLS analysis of the metabolite concentrations obtained using the targeted profiling ap-proach. A) Coloured according to type of mouse and B) coloured according to gender. Note that the first component in OPLS (x-axis) provides information on class separation, and as a result only the first component is considered for interpre-tation of metabolites relevant to arthritis. The first orthogonal component shows preliminary evidence of gender separation.

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

6

7

-4 -3 -2 -1 0 1 2 3 4

t[2]O

t[1]P

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

6

7

-4 -3 -2 -1 0 1 2 3 4

t[2]O

t[1]P

Normal

Female

Male

A) B)

Synovial fluid Cartilage

Pannus

Bone

Collagenase and Metalloproteinase mediated damage

SCF

Reactive oxygen species (ROS) mediated damage

IL-1, IL-6, TNF-α, other cytokines, his-tamines, tryptase,

chymase

Neutrophil Recruitment

Synovial Mast Cells

Macrophage Activation

Fibroblast activation and hyperplasia

T Cell Stimulation

Chondrocyte Activation

Requires de novo pyramidine synthesis

Methylation

Cytokine and ROS release is blunted in the glycine-gated choride channed of white blood cells

Taurine and Taurine - chloramine are known anti-inflammatory compounds

Mainnemare et al., (2004) J. Dent. Res. 83:823

Synovial fluid

Reactive oxygen species (ROS) mediated damage

Neutrophil Recruitment

Macrophage Activation

T Cell Stimulation

Collagenase and Metalloproteinase mediated damage

Wheeler et al., (1999) CMLS, 56:843.

Bottiglieri (2002) Am. J. Clin. Nutr., 76(sup):1151S.