nature medicine

Article https://doi.org/10.1038/s41591-023-02793-8

A terminal metabolite of niacin promotes

vascular inflammation and contributes to
cardiovascular disease risk
In the format provided by the
authors and unedited
 Supplemental Material

A terminal metabolite of niacin promotes vascular inflammation and contributes

to cardiovascular disease risk
Marc Ferrell1,2, Zeneng Wang1, James T. Anderson1, Xinmin S. Li1, Marco Witkowski1,14,
Joseph A. DiDonato1, James R. Hilser3,4, Jaana A. Hartiala3, Arash Haghikia5,6,7,8,15,
Tomas Cajka,9,10, Oliver Fiehn9, Naseer Sangwan1, Ilja Demuth11,12, Maximilian König11,
Elisabeth Steinhagen-Thiessen11, Ulf Landmesser 5,6,7,8,15, W . H. Wilson Tang1,13,
Hooman Allayee3,4, Stanley L. Hazen 1,13#
1
Department of Cardiovascular & Metabolic Sciences, Lerner Research Institute, Cleveland
Clinic, Cleveland, OH
2
Department of Nutrition, Systems Biology and Bioinformatics Program, Case Western Reserve
University, Cleveland, OH
3
Department of Population and Public Health Sciences, Keck School of Medicine, University of
Southern California, Los Angeles, CA 90033, USA
4
Department of Biochemistry & Molecular Medicine, Keck School of Medicine, University of
Southern California, Los Angeles, CA 90033, USA
5
Department of Cardiology, Angiology and Intensive Care, Deutsches Herzzentrum der Charité,
Campus Benjamin Franklin, Berlin, Germany
6
German Center for Cardiovascular Research (DZHK), Partner Site Berlin, Berlin, Germany
7
Berlin Institute of Health (BIH), Berlin, Germany
8
Friede Springer Cardiovascular Prevention Center at Charité, Berlin, Germany
9
West Coast Metabolomics Center, University of California, Davis, Davis, California, USA
10
Present address: Institute of Physiology of the Czech Academy of Sciences, Prague, Czech
Republic
11
Department of Endocrinology and Metabolism, Charité-Universitätsmedizin Berlin,
Charitéplatz, Berlin, Germany
12
Berlin Institute of Health Center for Regenerative Therapies, Berlin, Germany
13
Department of Cardiovascular Medicine, Heart, Vascular and Thoracic Institute, Cleveland
Clinic, Cleveland, OH
14
Present address: Department of Cardiology, Angiology and Intensive Care, Deutsches
Herzzentrum der Charité, Campus Benjamin Franklin, Berlin, Germany
15
Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and
Humboldt-Universität zu Berlin, Charitéplatz 1, 10117 Berlin, Germany

#Address for Correspondence:

Stanley L. Hazen, MD, PhD
9500 Euclid Avenue, NC-10, Lerner Research Institute, Cleveland Clinic, Cleveland, OH
44124
Phone: (216) 445-9763 / Fax: (216) 444-9404 / E-mail: [email protected]

1
Table of Contents
Supplemental Tables

Supplemental Table 1: Uniform resource locators (URLs) for GWAS datasets used

in meta-analyses of 2PY, 4PY, and sVCAM-1……………………………………….4

Supplemental Table 2: Baseline characteristics of patients in clinical

cohorts…………………………………..……………………………………………..…5

Supplemental Table 3: Previously reported plasma metabolites with known

chemical structures and those associated with prospective residual CVD risk from

untargeted metabolomics studies of the Discovery Cohort..…………………..……6

Supplemental Table 4: Top plasma metabolites with unknown chemical structures

associated with prospective residual CVD risk from untargeted metabolomics

studies of the Discovery Cohort……………………………………………………..…8

Supplemental Table 5: Pathway analysis of untargeted metabolomics comparing

prospective 3-year MACE cases versus controls.………………………....….……..9

Supplemental Table 6: Collision induced dissociation mass spectra of a serum

analyte associated with MACE (m/z = 153.0656 Da) and authentic chemical

standards of 2PY and 4PY ………………………………………………………..…..17

Supplemental Table 7: Phenome-wide associations study results for

rs10496731……………………………………………………………………………..18

Supplemental Table 8: Results of MR analyses with rs10496731 and CVD

outcomes……………………………………………………………………………….19

2
Supplemental Figures

Supplemental Figure 1: 1H-NMR spectrum of

1-methyl-4-oxo-1,4-dihydropyridine-3-carboxamide (4PY)….………………….....20

Supplemental Figure 2: 1H-NMR spectrum of

1-methyl-6-oxo-1,6-dihydropyridine-3-carboxamide (2PY).……….…….………...21

Supplemental Figure 3: Manhattan plots of levels of 2PY and 4PY in the US

Validation Cohort…………………………………………………………...……….....22

Supplemental Figure 4: Representative Western blot images, related to

Figure 4……………………………………………………………………………….…23

Supplemental Figure 5: Plasma sVCAM-1 is associated with increased

prospective MACE risk……..……………………………………………………….…25

Supplemental Figure 6: Impact of intraperitoneal 2PY and 4PY on serum 2PY, 4PY

levels.……..………………………………………………………………………….…26

Supplemental References…………………………………………………………………….27

3
Supplemental Tables
Supplemental Table 1: Uniform resource locators (URLs) for datasets used in

meta-analyses of 2PY, 4PY, and sVCAM-1

Analyte Cohort Ancestry URL

2PY
INTERVAL/EPIC 1 European https://omicscience.org/apps/mgwas

http://ftp.ebi.ac.uk/pub/databases/gwas/sum
CLSA 2 European mary_statistics/GCST90201001-
GCST90202000/GCST90201021/

https://pheweb.org/metsim-
METSIM 3 European
metab/download/C100001468

HCHS/SOL 4 Hispanic https://sites.cscc.unc.edu/hchs/

http://ftp.ebi.ac.uk/pub/databases/gwas/sum
JHS 5 African mary_statistics/GCST90176001-
GCST90177000/GCST90176301

4PY
http://ftp.ebi.ac.uk/pub/databases/gwas/sum
TwinsUK/KORA 6 European mary_statistics/GCST90243001-
GCST90244000/GCST90243701

https://jmorp.megabank.tohoku.ac.jp/gwas-
TMM 7 Japanese
analyses/TGA000005-ea5275c7#

sVCAM-1
https://ega-
INTERVAL 8 European
archive.org/studies/EGAS00001002555

https://www.mrc-
FENLAND 9 European
epid.cam.ac.uk/research/data-sharing/
https://www.ebi.ac.uk/gwas/studies/GCST9
AGES 10 European
0088153

https://download.decode.is/form/folder/prote
ICP/DECODE 11 European
omics

UK Biobank 12 European https://www.ukbiobank.ac.uk/

Meta-analyses were conducted for levels of 2PY, 4PY, and sVCAM-1 with publicly
available GWAS summary statistics from the datasets listed or by carrying out a single
variant analysis in the UK Biobank, as described in Methods.

4
Supplemental Table 2: Baseline characteristics of patients in clinical cohorts.
Characteristics Discovery US Validation EU Validation
cohort Cohort Cohort
(n = 1,162) (n = 2,331) (n = 832)
Age (years) 65.0 (55.6, 72.5) 63.4 (55.4, 71.6) 75.0 (66.0, 81.0)
Sex (% male) 63.7 66.8 70.1
Alcohol use
13.7 12.7 54.7
(% >1 drink/day)
Education (% 4-year
28.6 31.8 32.7
degree)
Diabetes (%) 22.1 30.3 27.9
Current Smoking (%) 13.6 12.8 16.8
History of … (%)
CAD 75.7 82.1 69.3
Hypertension 72.1 80.4 80.4
Hyperlipidemia 85.8 84.0 67.0
HDL (mg / dL) 34.3 (28.5, 41.2) 34.9 (28.7, 42.6) 48.0 (39.0, 60.0)
LDL (mg / dL) 96.0 (80.0, 116) 93.0 (76.0, 114) 91.5 (69.0, 122)
C-reactive protein (mg / L) 2.3 (1.0, 5.4) 2.3 (1.0, 5.3) 1.9 (0.8, 5.1)
eGFR (mL / min / 1.73 m2) 81.1 (67.9, 93.7) 91.2 (75.5, 100) 74.0 (60.3, 90.7)
Baseline Medications (%)
Aspirin 76.8 74.3 55.8
ACE inhibitor 49.9 50.3 37.9
Beta blocker 65.3 65.9 58.6
Statin 61.4 63.2 60.0
Event within 3 years n (%)
MACE 116 (10.0) 362* (15.5) 195 (23.4)
Myocardial infarction 44 (3.8) 117 (5.0) 131 (15.7)
Stroke 14 (1.2) 45 (1.9) 12 (1.4)
Death 58 (5.0) 201 (8.6) 52 (6.3)
All three cohorts consist of sequential stable subjects predominantly of northern
European ancestry (>95%) who underwent elective diagnostic coronary angiography
(cardiac catheterization or coronary computed tomography) for evaluation of coronary
artery disease (CAD). There were no nonoverlapping subjects between all three
cohorts. Continuous data are presented as median (interquartile range). Categorical
variables are presented as %; CAD, coronary artery disease; HDL, high density
lipoprotein; LDL, low density lipoprotein; eGFR, estimated glomerular filtration rate;
ACE, angiotensin-converting enzyme. Analyses of CVD risk in these cohorts are shown
in Figures 2 and 3. * One individual suffered a myocardial infarction and a stroke on the
same day; this was counted as one MACE event.
5
Supplemental Table 3: Previously reported plasma metabolites with known
chemical structures and those associated with prospective residual CVD risk
from untargeted metabolomics studies of the Discovery Cohort.
3-year MACE Q4 vs Q1
Proposed Analyte
P
Identity Adjusted Hazard Ratio (95% CI)

trimethylamine-N-oxide* 2.7 (1.5 - 4.8) 0.001

(N6,N6,N6)-trimethyllysine* 2.5 (1.4 - 4.5) 0.003
CE (18:2) 2.1 (1.1 - 3.8) 0.02
phenylacetylglutamine* 2.1 (1.2 – 3.7) 0.01
PE (38:4) 2.0 (1.1 – 3.9) 0.03
acylcarnitine C14:2 2.0 (1.1 - 3.5) 0.02
γ-butyrobetaine* 2.0 (1.1 – 3.7) 0.03
PE (36:2) 2.0 (1.1 - 3.5) 0.02
PC (34:5) 2.0 (1.1 - 3.7) 0.02
zolpidem** 2.0 (1.2 - 3.5) 0.01
TG (50:2) 1.9 (1.0 - 3.5) 0.01
cephalexin** 1.9 (1.2 - 3.0) 0.009
acylcarnitine C2:0 1.9 (1.1 - 3.3) 0.02
lysine 0.55 (0.31 - 0.98) 0.04
tryptophan 0.47 (0.27 - 0.82) 0.008
PC (34:4) 0.44 (0.24 - 0.81) 0.009
LPC (20:5) 0.38 (0.20 - 0.73) 0.0004***
midazolam 0.36 (0.20 - 0.67) 0.001
leucine 0.34 (0.20 - 0.61) 0.0002***
Sequential plasma samples from stable subjects undergoing elective diagnostic cardiac
evaluations were analyzed with untargeted metabolomics, and the relative levels of
analytes were analyzed for their association with incident 3-year MACE risk independent
of traditional CVD risk factors, as described in Methods. Analytes were classified as
Knowns or Unknowns based on whether their putative structures were identified. Shown
are the top putatively identified Knowns associated with residual CVD risk as well as
previously reported Knowns. Adjusted hazard ratios are shown for MACE for the top
versus bottom quartile (adjusted for traditional cardiovascular risk factors (age, sex,
systolic blood pressure, LDL, HDL, triglycerides, diabetes status) and high-sensitivity C-
reactive protein (hsCRP), current (active) smoking status, alcohol use, and education

6
level. P values were determined with two-sided Wald tests with no adjustments for
multiple testing. CI, confidence interval; m/z, mass to charge ratio; Q, quartile; PE,
phosphatidylethanolamine; PC, phosphatidylcholine; CE, cholesterol ester; TG,
triglyceride; SM, sphingomyelin; LPC, lysophosphatidylcholine; UDP, Uridine
diphosphate.
* Structural identification studies and independent stable isotope dilution LC/MS/MS
studies with Validation Cohorts verifying links with incident MACE risks were previously
reported 13,14,15,16,17.
**Untargeted metabolomics data of indicated compounds were previously reported 17.
*** P<0.05 after adjustment for multiple testing by the method of Benjamani and Hochberg

7
Supplemental Table 4: Top plasma metabolites with unknown chemical structures

associated with prospective residual CVD risk from untargeted metabolomics

analysis of the Discovery Cohort.

3-year MACE Q4 vs Q1
Analyte m/z (Da) P
Adjusted Hazard Ratio (95% CI)
Unknown #221
153.0656 2.7 (1.5 - 4.6) 0.0005**
(2PY + 4PY)*
Unknown #043 906.7710 2.1 (1.1 - 4.0) 0.02
Unknown #243 876.6837 2.1 (1.2 - 3.9) 0.01
Unknown #210 185.1278 2.0 (1.1 - 3.6) 0.02
Unknown #244 878.6990 2.0 (1.1 - 3.7) 0.03
Unknown #260 822.6372 1.9 (1.0 – 3.5) 0.04

Sequential plasma samples from stable subjects undergoing elective diagnostic cardiac
evaluations were analyzed by untargeted metabolomics, and the relative levels of
analytes were analyzed for their association with incident 3-year MACE risk
independent of traditional CVD risk factors, as described in Methods. Analytes were
classified as Knowns or Unknowns based on whether their putative structures were
identified. Shown are the top Unknowns associated with residual CVD risk. Adjusted
hazard ratios are shown for incident 3-year MACE for the top versus bottom quartile
(adjusted for traditional cardiovascular risk factors LDL, HDL, triglycerides, diabetes
status) and high-sensitivity C-reactive protein (hsCRP), current (active) smoking status,
alcohol use, and education level). P values were determined with two-sided Wald tests
with no adjustments for multiple testing. CI, confidence interval; m/z, mass to charge
ratio; Q, quartile
*This manuscript focuses on the identification of the top unknown (#221; m/z 153.0656),
and its discovery and characterization as the two terminal metabolites of niacin, 2PY
and 4PY. CVD risk analysis of this top unknown is shown in Figure 2. Structural
elucidation studies are described in Methods and shown in Extended Data Figures 3, 4,
and 5.
** P<0.05 after adjustment for multiple testing by the method of Benjamani and
Hochberg

8
 Supplemental Table 5: Pathway analysis of untargeted metabolomics comparing

prospective 3-year MACE cases versus controls.

Query Matched Matched Mass Matched P

Mass m/z Compound ID Adduct Difference Compound
(Da) (Da)
Valine, leucine and isoleucine degradation, P = 0.004
132.1019 C00123 M+H[1+] 2.35E-05 L-Leucine 0.0037
132.1019 C00407 M+H[1+] 2.35E-05 L-Isoleucine 0.0037
104.1070 C00123 M-CO+H[1+] 1.24E-04 L-Leucine 0.75
104.1070 C00407 M-CO+H[1+] 1.24E-04 L-Isoleucine 0.75
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
76.0757 C03284 M-CO+H[1+] 1.24E-04 L-3-Aminoisobutyrate 0.2
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
100.0756 C00183 M-H2O+H[1+] 7.65E-05 L-Valine 0.093
118.0863 C00183 M+H[1+] 2.35E-05 L-Valine 0.58
Urea cycle/amino group metabolism, P = 0.008
134.0600 C00547 M-H4O2+H[1+] 2.35E-05 Arterenol 0.43
134.0600 C01586 M-HCOOH+H[1+] 7.65E-05 Hippurate 0.43
61.0396 C00086 M+H[1+] 7.65E-05 Urea 0.97
204.123 C00315 M+NaCl[1+] 7.76E-04 Spermidine 0.073
86.0708 C00791 M-CO+H[1+] 3.76E-04 Creatinine 0.59
114.0657 C00791 M+H[1+] 4.76E-04 Creatinine 0.3
114.0657 C00300 M-H2O+H[1+] 4.76E-04 Creatine 0.3
159.0758 C00327 M-NH3+H[1+] 6.76E-04 Citrulline 0.49
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
76.0757 C00334 M-CO+H[1+] 1.24E-04 Gamma-aminobutyric 0.2
acid
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
100.0756 C02946 M-HCOOH+H[1+] 1.76E-04 4-Acetamidobutanoate 0.093
198.0842 C00327 M+Na[1+] 7.76E-04 Citrulline 0.17
132.0768 C00300 M+H[1+] 2.35E-05 Creatine 0.62
132.0768 C00791 M+H2O+H[1+] 2.35E-05 Creatinine 0.62
176.1030 C00327 M+H[1+] 2.35E-05 Citrulline 0.77
118.0863 C02946 M-CO+H[1+] 1.24E-04 4-Acetamidobutanoate 0.58
Butanoate metabolism, P = 0.009
132.1019 C03087 M-CO+H[1+] 1.24E-04 5-Acetamidopentanoate 0.0037
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
76.0757 C00334 M-CO+H[1+] 1.24E-04 Gamma-aminobutyric 0.2
acid
100.0756 C00431 M-H2O+H[1+] 7.65E-05 5-Aminopentanoate 0.093
118.0863 C00431 M+H[1+] 2.35E-05 5-Aminopentanoate 0.58

9
 Matched Mass
Query Mass Matched
Compound Difference Matched Compound P
m/z (Da) Adduct
ID (Da)
Beta-Alanine metabolism, P = 0.01
204.123 C00315 M+NaCl[1+] 7.76E-04 Spermidine 0.073
227.1139 C00386 M+H[1+] 2.35E-05 Carnosine 0.079
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
76.0757 C00334 M-CO+H[1+] 1.24E-04 Gamma-aminobutyric acid 0.2
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Arginine and Proline Metabolism, P = 0.015
61.0396 C00086 M+H[1+] 7.65E-05 Urea 0.97
159.0758 C00327 M-NH3+H[1+] 6.76E-04 Citrulline 0.49
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
148.0604 C05938 M+H[1+] 7.65E-05 L-4-Hydroxyglutamate 0.31
semialdehyde
148.0604 C04281 M+H2O+H[1+] 7.65E-05 L-1-Pyrroline-3-hydroxy-5- 0.31
carboxylate
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
120.0655 C05938 M-CO+H[1+] 2.35E-05 L-4-Hydroxyglutamate 0.14
semialdehyde
100.0756 C02946 M-HCOOH+H[1+] 1.76E-04 4-Acetamidobutanoate 0.093
198.0842 C00327 M+Na[1+] 7.76E-04 Citrulline 0.17
176.1030 C00327 M+H[1+] 2.35E-05 Citrulline 0.77
118.0863 C02946 M-CO+H[1+] 1.24E-04 4-Acetamidobutanoate 0.58
Arachidonic acid metabolism, P = 0.017
279.2309 C06427 M+H[1+] 9.76E-04 Linolenate 0.25
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
876.6837 C00157 M+Na[1+] 1.98E-03 Lecithin 0.022
C21-steroid hormone biosynthesis and metabolism, P = 0.017
279.2309 C05476 M-C3H4O2+H[1+] 9.76E-04 Tetrahydrocorticosterone 0.25
585.2697 C00486 M+H[1+] 1.08E-03 Bilirubin 0.043
584.2612 C00486 M[1+] 2.30E-03 Bilirubin 0.2
468.3079 cholesterols M(S34)+H[1+] 9.68E-04 Cholesterol sulfate 0.1
Lysine metabolism, P = 0.022
146.1176 C05543 M+H[1+] 4.48E-05 0.67
146.1176 C01181 M[1+] 5.00E-04 Butyro-betaine 0.67
189.1598 C03793 M+H[1+] 2.35E-05 N6,N6,N6-Trimethyl-L- 0.15
lysine
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
100.0756 C03239 M-HCOOH+H[1+] 1.76E-04 2-Oxo-6-aminocaproate 0.093
100.0756 C04092 M-CO+H[1+] 2.35E-05 1-Piperideine-2- 0.093
carboxylate

10
 Query Mass
Matched Matched
Mass m/z Difference Matched Compound P
Compound ID Adduct
(Da) (Da)
100.0756 C00450 M-CO+H[1+] 2.35E-05 Delta1-Piperideine-6-L- 0.093
carboxylate
100.0756 C04076 M-HCOOH+H[1+] 1.76E-04 Allysine 0.093
172.1331 C03793 M-NH3+H[1+] 1.76E-04 N6,N6,N6-Trimethyl-L- 0.95
lysine
118.0863 C03239 M-CO+H[1+] 1.24E-04 2-Oxo-6-aminocaproate 0.58
118.0863 C04076 M-CO+H[1+] 1.24E-04 Allysine 0.58
Linoleate metabolism, P = 0.022
279.2309 C06426 M+H[1+] 9.76E-04 Gamolenic acid 0.25
279.2309 C14762 M-H2O+H[1+] 8.76E-04 13(S)-HODE 0.25
279.2309 C14826 M-H2O+H[1+] 8.76E-04 12(13)-EpOME 0.25
279.2309 C14825 M-H2O+H[1+] 8.76E-04 9(10)-EpOME 0.25
315.2523 C14762 M+H2O+H[1+] 6.76E-04 13(S)-HODE 0.52
315.2523 C14826 M+H2O+H[1+] 6.76E-04 12(13)-EpOME 0.52
315.2523 C14825 M+H2O+H[1+] 6.76E-04 9(10)-EpOME 0.52
508.3761 C04230 M+H[1+] 5.14E-05 2-Lysolecithin 0.83
876.6837 C00157 M+Na[1+] 1.98E-03 Lecithin 0.022
Histidine metabolism, P = 0.022
83.0599 C05130 M-CO+H[1+] 3.76E-04 Imidazole acetaldehyde 0.34
227.1139 C00386 M+H[1+] 2.35E-05 Carnosine 0.079
159.0758 C05828 M+H2O+H[1+] 6.76E-04 Methylimidazoleacetate 0.49
141.0657 C05828 M+H[1+] 1.76E-04 Methylimidazoleacetate 0.99
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Glycerophospholipid metabolism, P = 0.024
279.2309 C06427 M+H[1+] 9.76E-04 Linolenate 0.25
146.1176 C01996 M[1+] 5.00E-04 Acetylcholine 0.67
104.1070 C00114 M[1+] 5.00E-04 Choline 0.75
550.3845 C04598 M+H[1+] 2.22E-03 Platelet-activating factor 0.45
764.5192 C02737 M-C3H4O2+H[1+] 3.31E-03 Phosphatidylserine 0.62
508.3761 C04230 M+H[1+] 5.14E-05 2-Lysolecithin 0.83
876.6837 C00157 M+Na[1+] 1.98E-03 Lecithin 0.022
572.3688 C04598 M+Na[1+] 8.39E-05 Platelet-activating factor 0.95
Porphyrin metabolism, P = 0.026
585.2697 C00486 M+H[1+] 1.08E-03 Bilirubin 0.043
584.2612 C00500 M(C13)+H[1+] 2.72E-03 Biliverdin 0.2
584.2612 C00486 M[1+] 2.30E-03 Bilirubin 0.2
585.2691 C00486 M+H[1+] 1.68E-03 Bilirubin 0.92
Glycine, serine, alanine and threonine metabolism, P = 0.040
114.0657 C00300 M-H2O+H[1+] 4.76E-04 Creatine 0.3
104.1070 C00114 M[1+] 5.00E-04 Choline 0.75

11
 Query Mass
Matched Matched
Mass m/z Difference Matched Compound P
Compound ID Adduct
(Da) (Da)
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
76.0757 C05519 M-CO2+H[1+] 2.35E-05 L-Allothreonine 0.2
76.0757 C01026 M-CO+H[1+] 1.24E-04 Dimethylglycine 0.2
76.0757 C00188 M-CO2+H[1+] 2.35E-05 L-Threonine 0.2
120.0655 C05519 M+H[1+] 2.35E-05 L-Allothreonine 0.14
120.0655 C00188 M+H[1+] 2.35E-05 L-Threonine 0.14
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
100.0756 C00719 M-H2O+H[1+] 7.65E-05 Betaine 0.093
132.0768 C00300 M+H[1+] 2.35E-05 Creatine 0.62
118.0863 C00719 M+H[1+] 2.35E-05 Betaine 0.58
Glutamate metabolism, P = 0.045
223.0737 C00669 M-CO+H[1+] 8.76E-04 gamma-Glutamylcysteine 0.7
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
76.0757 C00334 M-CO+H[1+] 1.24E-04 Gamma-aminobutyric acid 0.2
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Vitamin B3 (nicotinate and nicotinamide) metabolism, P = 0.047
153.0656 C05842 M+H[1+] 2.76E-04 N1-Methyl-2-pyridone-5- 0.021
carboxamide
153.0656 C05843 M+H[1+] 2.76E-04 1-Methyl-4-pyridone-3- 0.021
carboximide
141.0657 C00153 M+H2O+H[1+] 1.76E-04 Nicotinamide 0.99
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
137.0709 C02918 M[1+] 6.00E-4 N1-methylnicotinamide 0.99
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
136.0388 C05842 M-NH3+H[1+] 5.76E-04 N1-Methyl-2-pyridone-5- 0.017
carboxamide
136.0388 C05843 M-NH3+H[1+] 5.76E-04 1-Methyl-4-pyridone-3- 0.017
carboximide
Vitamin B9 (folate) metabolism, P = 0.048
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
148.0604 C01045 M-CO+H[1+] 2.35E-05 N-Formyl-L-glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Fatty Acid Metabolism, P = 0.048
315.2523 dmnoncrn M-NH3+H[1+] 7.35E-04 NA 0.52
204.1230 C02571 M[1+] 6.00E-04 O-Acetylcarnitine 0.073
227.1139 C02571 M+Na[1+] 1.02E-03 O-Acetylcarnitine 0.079
Aspartate and asparagine metabolism, P = 0.058
144.101 C03078 M[1+] 1.00E-04 4-Guanidinobutanamide 0.46
61.0396 C00086 M+H[1+] 7.65E-05 Urea 0.97
61.0396 C00152 M-C3H4O2+H[1+] 7.65E-05 L-Asparagine 0.97
223.0737 C00669 M-CO+H[1+] 8.76E-04 gamma-Glutamylcysteine 0.7

12
 Query Mass
Matched Matched
Mass m/z Difference Matched Compound P
Compound ID Adduct
(Da) (Da)
204.1230 CE5586 M+NaCl[1+] 7.74E-04 NA 0.073
204.1230 C00315 M+NaCl[1+] 7.76E-04 Spermidine 0.073
204.1230 C02571 M[1+] 6.00E-04 O-Acetylcarnitine 0.073
227.1139 C02571 M+Na[1+] 1.02E-03 O-Acetylcarnitine 0.079
159.0758 C00327 M-NH3+H[1+] 6.76E-04 Citrulline 0.49
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
148.0604 C05938 M+H[1+] 7.65E-05 L-4-Hydroxyglutamate 0.31
semialdehyde
148.0604 C01879 M+H2O+H[1+] 7.65E-05 5-Oxoproline 0.31
148.0604 C02238 M+H2O+H[1+] 6.35E-05 NA 0.31
148.0604 C04281 M+H2O+H[1+] 7.65E-05 L-1-Pyrroline-3-hydroxy-5- 0.31
carboxylate
148.0604 C04282 M+H2O+H[1+] 7.65E-05 1-Pyrroline-4-hydroxy-2- 0.31
carboxylate
76.0757 C02356 M-CO+H[1+] 1.24E-04 (S)-2-Aminobutanoate; 0.2
76.0757 C00334 M-CO+H[1+] 1.24E-04 Gamma-aminobutyric acid 0.2
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
120.0655 C05938 M-CO+H[1+] 2.35E-05 L-4-Hydroxyglutamate 0.14
semialdehyde
100.0756 C02946 M-HCOOH+H[1+] 1.76E-04 4-Acetamidobutanoate 0.093
198.0842 C00327 M+Na[1+] 7.76E-04 Citrulline 0.17
176.1030 C00327 M+H[1+] 2.35E-05 Citrulline 0.77
118.0863 C02946 M-CO+H[1+] 1.24E-04 4-Acetamidobutanoate 0.58
Glutathione Metabolism, P = 0.065
223.0737 C00669 M-CO+H[1+] 8.76E-04 gamma-Glutamylcysteine 0.7
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
148.0604 C01879 M+H2O+H[1+] 7.65E-05 5-Oxoproline 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Methionine and cysteine metabolism, P = 0.065
223.0737 C02291 M+H[1+] 9.76E-04 L-Cystathionine 0.7
223.0737 C00542 M+H[1+] 9.76E-04 Cystathionine 0.7
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
148.0604 C00302 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Nitrogen metabolism, P = 0.074
61.0396 C00152 M-C3H4O2+H[1+] 7.65E-05 L-Asparagine 0.97
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Alanine and Aspartate Metabolism, P = 0.084
61.0396 C00152 M-C3H4O2+H[1+] 7.65E-05 L-Asparagine 0.97
159.0758 C00327 M-NH3+H[1+] 6.76E-04 Citrulline 0.49

13
 Query Mass
Matched Matched
Mass m/z Difference Matched Compound P
Compound ID Adduct
(Da) (Da)
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
148.0604 C01042 M-CO+H[1+] 2.35E-05 N-Acetyl-L-aspartate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
198.0842 C00327 M+Na[1+] 7.76E-04 Citrulline 0.17
176.1030 C00327 M+H[1+] 2.35E-05 Citrulline 0.77
Carnitine shuttle, P = 0.088
524.3710 tetpent6crn M+Na[1+] 8.56E-05 Tetracosapentaenoyl 0.073
carnitine
524.3710 tetpent3crn M+Na[1+] 8.56E-05 Tetracosapentaenoyl 0.073
carnitine
496.3407 dcsptn1crn M+Na[1+] 9.30E-04 Docosa-4,7,10,13,16- 0.055
pentaenoyl carnitine
496.3407 clpndcrn M+Na[1+] 9.30E-04 Clupanodonyl carnitine 0.055
608.4875 hexccrn M+HCOONa[1+] 1.49E-03 Hexacosanoyl carnitine 0.55
568.4081 nrvnccrn M+NaCl[1+] 2.19E-03 Nervonyl carnitine 0.92
146.1176 pcrn M-C3H4O2+H[1+] 1.45E-05 Propionyl-carnitine 0.67
468.3079 tmndnccrn M+Na[1+] 5.85E-04 Timnodonyl carnitine 0.1
494.3239 c226crn M+Na[1+] 2.45E-04 Cervonyl carnitine 0.2
172.1331 pcrn M-HCOOH+H[1+] 1.85E-04 Propionyl-carnitine 0.95
218.1387 pcrn M+H[1+] 1.45E-05 Propionyl-carnitine 0.39
Purine metabolism, P = 0.13
61.0396 C00086 M+H[1+] 7.65E-05 Urea 0.97
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Pyrimidine metabolism, P = 0.14
61.0396 C02642 M-C3H4O2+H[1+] 7.65E-05 3-Ureidopropionate 0.97
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
76.0757 C05145 M-CO+H[1+] 1.24E-04 3-Aminoisobutanoate 0.2
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Tryptophan metabolism, P = 0.18
205.0972 C00078 M+H[1+] 2.35E-05 Tryptophan 0.019
205.0972 C00525 M+H[1+] 2.35E-05 D-Tryptophan 0.019
134.0600 C05639 M-CO+H[1+] 2.35E-05 Quinoline-4,6-diol 0.43
134.0600 C05660 M-C3H4O2+H[1+] 7.65E-05 5-Methoxyindoleacetate 0.43
153.0656 C03227 M-C3H4O2+H[1+] 2.76E-04 3-Hydroxy-L-kynurenine 0.021
153.0656 C05651 M-C3H4O2+H[1+] 2.76E-04 5-Hydroxykynurenine 0.021
265.1188 C05642 M+H[1+] 5.24E-04 Formyl-N-acetyl-5- 0.23
methoxykynurenamine
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
177.1022 C00078 M-CO+H[1+] 2.35E-05 Tryptophan 0.6
177.1022 C00525 M-CO+H[1+] 2.35E-05 D-Tryptophan 0.6

14
 Matched Mass
Query Mass Matched
Compound Difference Matched Compound P
m/z (Da) Adduct
ID (Da)
Tyrosine metabolism, P = 0.19
134.0600 CE5536 M-HCOOH+H[1+] 1.19E-04 Adrenochrome 0.43
134.0600 C00547 M-H4O2+H[1+] 2.35E-05 Artereno 0.43
61.0396 C09642 M(C13)+3H[3+] 2.76E-04 Salsolinol 0.97
265.1188 C04148 M+H[1+] 5.24E-04 Phenylacetylglutamine 0.23
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Sialic acid metabolism, P = 0.26
608.4875 C01190 M-H4O2+H[1+] 8.95E-04 Glucosylceramide 0.55
764.5192 C02737 M-C3H4O2+H[1+] 3.31E-03 Phosphatidylserine 0.62
Bile acid biosynthesis, P = 0.34
455.2904 C05457 M+K[1+] 1.42E-03 NA 0.62
455.2904 C17339 M+K[1+] 1.38E-03 4-Cholesten- 0.62
7alpha,12alpha-diol-3-one
455.2904 C17336 M+K[1+] 1.38E-03 7alpha,26-Dihydroxy-4- 0.62
cholesten-3-one
455.2904 C17331 M+K[1+] 1.38E-03 7alpha,24-Dihydroxy-4- 0.62
cholesten-3-one
455.2904 C17333 M+K[1+] 1.38E-03 3beta-Hydroxy-5- 0.62
cholestenoate
455.2904 C17332 M+K[1+] 1.38E-03 7alpha,25-Dihydroxy-4- 0.62
cholesten-3-one
454.3000 C05463 M-HCOOH+H[1+] 1.32E-03 Taurodeoxycholate 0.86
454.3000 C05465 M-HCOOH+H[1+] 1.32E-03 Taurochenodeoxycholate 0.86
Glycosphingolipid metabolism, P = 0.34
608.4875 C01190 M-H4O2+H[1+] 8.95E-04 Glucosylceramide 0.55
790.5727 C00550 M+NaCl[1+] 1.35E-04 Sphingomyelin 0.2
731.6058 C00550 M[1+] 8.89E-04 Sphingomyelin 0.24
Vitamin D3 (cholecalciferol) metabolism, P > 0.99
455.2904 C01673 M+K[1+] 1.38E-03 Calcitriol 0.62
455.2904 CE2201 M+K[1+] 1.42E-03 NA 0.62
455.2904 CE1337 M+K[1+] 1.42E-03 NA 0.62
Putative anti-Inflammatory metabolites formation from EPA, P > 0.99
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Alkaloid biosynthesis II, P > 0.99
172.1331 egme M-CO+H[1+] 2.65E-05 Ecgonine methyl ester 0.95
218.1387 egme M+H2O+H[1+] 2.65E-05 Ecgonine methyl ester 0.39
Lipoate metabolism, P > 0.99
357.1274 CE2102 M+Na[1+] 3.65E-04 Lipoyllysine 0.62
Drug metabolism – cytochrome P450, P > 0.99
312.1583 C16608 M(C13)+H[1+] 4.76E-04 Demethylcitalopram 0.35

15
 Mass
Query Mass Matched Matched
Difference Matched Compound P
m/z (Da) Compound ID Adduct
(Da)
De novo fatty acid biosynthesis, P > 0.99
279.2309 C06426 M+H[1+] 9.76E-04 Gamolenic acid 0.25
279.2309 C06427 M+H[1+] 9.76E-04 Linolenate 0.25
Omega-3 fatty acid metabolism, P > 0.99
279.2309 C06427 M+H[1+] 9.76E-04 Linolenate 0.25
Sphingolipid metabolism, P > 0.99
790.5727 C00550 M+NaCl[1+] 1.35E-04 Sphingomyelin 0.2
731.6058 C00550 M[1+] 8.89E-04 Sphingomyelin 0.24
Vitamin B6 (pyridoxine) metabolism, P > 0.99
134.0600 C00314 M-H4O2+H[1+] 2.35E-05 Pyridoxol 0.43
Leukotriene metabolism, P > 0.99
148.0604 C00302 M+H[1+] 7.65E-05 Glutamate 0.31
Fatty acid activation, P > 0.99
279.2309 C06426 M+H[1+] 9.76E-04 Gamolenic acid 0.25
279.2309 C06427 M+H[1+] 9.76E-04 Linolenate 0.25
Aminosugars metabolism, P > 0.99
148.0604 C00025 M+H[1+] 7.65E-05 Glutamate 0.31
120.0655 C00025 M-CO+H[1+] 2.35E-05 Glutamate 0.14
Phytanic acid peroxisomal oxidation, P > 0.99
360.1397 C07343 M+3H[3+] 1.71E-03 2-Hydroxyphytanoyl-CoA 0.092

High resolution mass spectra acquired from patient serum were analyzed with
mummichog (v1.0.10) pathway analysis using the Metaboanalyst v4.0 release of the
Human MFN genome-scale metabolic model comparing prospective 3-year MACE
cases vs controls (Methods). Human MFN contains pathway information derived from
KEGG release 81.0, SMPDB v2.0, HMDB v4.0, ChEBI release 131, and Biocyc release
17.0. For each pathway (bold), the enrichment P value is shown for pathway enrichment
with MACE-associated metabolites. Pathway enrichment P values were determined
using the mummichog method 18, which includes correction for multiple testing. Below
each pathway name, the component metabolites (Matched Compound) matched to a
detected metabolite (Query Mass) are listed. For each detected metabolite (Query
Mass), the matched compound or compounds are shown with the predicted adduct,
mass difference from the predicted m/z for the reference compound, and the two-sided
student t test P value (not adjusted for multiple testing) comparing metabolite
abundance in prospective 3-year MACE cases vs controls. MACE-associated pathways
containing MACE-associated metabolites are shown in Extended Data Figure 1. The
“vitamin B3 [niacin/NAD] metabolism” pathway is shown in Extended Data Figure 2.
Compounds lacking a common name are shown as “NA”. HODE, hydroxyoctadecanoic
acid; EpOME, Epoxyoctadecanoic acid.

16
 Supplemental Table 6: Collision induced dissociation mass spectra of a serum

analyte associated with MACE (m/z = 153.0656 Da) and authentic chemical

standards of 2PY and 4PY

Elemental Predicted Measured m/z (Da) Measured m/z (Da) Difference

Analyte
composition m/z (Da) (Authentic standard) (Unknown metabolite) (ppm)
Metabolite 1 / 2PY
Molecular ion C7H9N2O2 153.0664 153.0658 153.0665 +0.7 (+4.6)
Fragment 1 C6H8NO 110.0606 110.0605 110.0607 +0.9 (+1.8)
Fragment 2 C6H6NO 108.0449 108.0450 108.0451 +1.9 (+0.9)
Fragment 3 C5H6NO 96.0449 96.0451 96.0452 +3.1 (+1.0)
Fragment 4 C5H6N 80.0500 80.0502 80.0503 +3.8 (+1.2)
Metabolite 2 / 4PY
Molecular ion C7H9N2O2 153.0664 153.0660 153.0662 -1.3 (+1.3)
Fragment 1 C7H6NO2 136.0399 136.0395 136.0397 -1.5 (+1.5)
Fragment 2 C6H6NO 108.0449 108.0450 108.0452 +2.8 (+1.9)
Fragment 3 C6H6N 92.0500 92.0501 92.0502 +1.1 (+1.1)

Untargeted metabolomics revealed a serum analyte with calculated elemental formula

C7H9N2O2 (m/z = 153.0656 Da). This analyte was separated into two analytes with
calculated elemental formula C7H9N2O2 using HPLC (Metabolite 1 and Metabolite 2).
Stable isotope-labeled 2PY and 4PY were diluted in human serum, and Metabolite 1 co-
eluted with d3-2PY while Metabolite 2 co-eluted with d3-4PY. High resolution collision
induced dissociation (CID) experiments were performed on Metabolite 1 and Metabolite
2 in serum as well as authentic chemical standards of 2PY and 4PY dissolved in
deionized water. For the molecular ion and fragments, Metabolite 1 is compared to 2PY
and Metabolite 2 is compared to 4PY. The measured m/z ratios for authentic standards
and calculated elemental compositions are shown. Also shown are the m/z ratios
predicted from the elemental composition based on atomic weights and the m/z ratios
measured in serum for Metabolite 1 and Metabolite 2. The expected experimental error
in m/z ratios was 5 ppm. The difference between measured m/z ratios for unknown
metabolites (Metabolite 1 and Metabolite 2) and authentic chemical standards is shown
in parentheses, and the difference between measured m/z ratios for unknown
metabolites and predicted m/z ratios is shown without parentheses. HPLC
chromatograms showing full separation are shown in Extended Data Figure 4, and CID
mass spectra are shown in Extended Data Figure 5.

17
Supplemental Table 7: Phenome-wide associations study results for rs10496731.

Trait Type P

1,5-anhydroglucitol Metabolite 7.4 x 10-16

Quinolinate Metabolite 2.5 x 10-13

1-methylnicotinamide Metabolite 1.7 x 10-12

**X – 25790 Metabolite 7.3 x 10-9

3,5-dichloro-2,6-dihydroxybenzoic acid Metabolite 3.9 x 10-6

Indolepropionate Metabolite 2.0 x 10-5

Lactase-phlorizin hydrolase Protein 2.5 x 10-54

Histone acetyltransferase (KAT6A) Protein 3.6 x 10-5

Vascular cell adhesion molecule 1 (VCAM-1) Protein 4.5 x 10-5

Fractalkine (CX3CL1) Protein 7.2 x 10-5

Proteomic 8 and metabolomic 3 GWAS datasets were searched for associations with
rs10496731. Only associations with P < 1 x 10-4 are shown. Rs10496731 was
significantly associated with several proteins and metabolites, such as lactase, 1,5-
anhydroglucitol, quinolinate, and 1-methylnicotinamide, and suggestively associated
with levels of lysine acetyltransferase 6A (P=3.6x10-5), sVCAM-1 (P=4.5x10-5), and
fractalkine (P=7.2x10-5). We ruled out genetically regulated 2PY/4PY levels being
related to lactose metabolism and 1,5-anhydroglucitol levels since rs10496731 is in
weak linkage disequilibrium (LD; r2=0.23) with another variant (rs4988235) that is one of
the strongest known determinants of 1,5-anhydroglucitol (P=4.9x10-44) 3 and lactase
(P=1.5x10-222) 8 levels, and is the genetic basis for lactose intolerance in humans 19.
Therefore, we focused on replicating the suggestive associations of rs10496731 with
the remaining three proteomic markers. Based on the results of an independent
proteomics GWAS in a larger Icelandic cohort 10, consistent evidence of association
was only observed between rs10496731 and sVCAM-1 (P=1.7x10-4), but not with lysine
acetyltransferase 6A (P=0.47) or fractalkine (P=0.22). Association of rs10496731 with
increased quinolinate and 1-methylnicotinamide levels is also consistent with the T
allele being associated with reduced ACMSD functional capacity. P values were
determined with two-sided Wald tests with no adjustment for multiple testing.
** Unknown metabolite

18
Supplemental Table 8: Results of MR analyses with rs10496731 and CVD

outcomes.

2PY 4PY

Outcome OR(95% CI) P-value OR(95% CI) P-value

MACE 0.97(0.85-1.10) 0.61 0.96(0.83-1.11) 0.61

CAD 1.00(0.91-1.09) 0.93 1.00(0.90-1.10) 0.93

MI 1.09(0.94-1.27) 0.26 1.10(0.93-1.29) 0.26

Stroke 0.93(0.83-1.05) 0.25 0.93(0.82-1.05) 0.25

Data are shown as odds ratios (95% CIs) for association of genetically increased 2PY

and 4PY levels on various CVD outcomes. P values were determined with two-sided

Wald ratio method with no adjustment for multiple testing, as implemented in the Two-

Sample MR package in R v4.0.3 20. The genetic instrument for the exposures were

based on the effect sizes of rs10496731 on 2PY and 4PY levels obtained from the

meta-analyses with the various cohorts (Supplemental Table 1 and Extended Data

Table 1). For the outcome, effect estimates were derived from our analysis of

rs10496731 with risk of incident MACE in the UK Biobank as well as those from publicly

available summary statistics for CAD 21, MI 22, and stroke 23.

19
Supplemental Figures

Supplemental Figure 1: 1H-NMR spectrum of 1-methyl-4-oxo-1,4-dihydropyridine-

3-carboxamide (4PY).

4PY was synthesized from 4-chloronicotinic acid as described in Methods. 1H-NMR of

the resulting product was performed at 400 MHz on the resulting product dissolved in

DMSO-d6, and the resulting spectrum matched previously reported spectra24.

20
Supplemental Figure 2: 1H-NMR spectrum of 1-methyl-6-oxo-1,6-dihydropyridine-

3-carboxamide (2PY).

2PY was synthesized from 6-hydroxynicotinic acid as acid as described in Methods. 1H-

NMR was performed at 400 MHz on the resulting product dissolved in DMSO-d6, and the

resulting spectrum matched previously reported spectra 25.

21
Supplemental Figure 3: Manhattan plots of levels of 2PY and 4PY in the US
Validation Cohort

(a) Manhattan plot of GWAS conducted in the US Validation Cohort for levels of 4PY.
(b) Manhattan plot of GWAS conducted in the US Validation Cohort for levels of 2PY.
Red and blue lines indicate the genome-wide thresholds for significant and suggestive
association at P = 5.0 x 10-8 and P = 5.0 x 10-6, respectively. P values were determined
with two-sided Wald tests using linear regression analysis assuming an additive model
with adjustment for age, sex, and genotyping array. P values are not adjusted for
multiple testing.

22
Supplemental Figure 4: Representative Western blot images, related to Figures 4
and 5
Representative Western blot images used to generate protein quantifications shown in
Figures 4 and 5. (a) Representative Western blot image used to generate protein
quantifications shown in Figure 4f Mice were transduced with AAV expressing either
scrambled shRNA (Control) or shRNAs targeting Acmsd transcripts (Knockdown), as
indicated. Tissues or cells were lysed in RIPA lysis buffer containing a
protease/phosphatase inhibitor cocktail. Total protein was quantified with a bicinchoninic

23
acid (BCA) protein assay kit. Membranes were probed with primary antibodies, washed,
and then probed with fluorophore-conjugated secondary antibodies as described in
Methods. Membranes were scanned using the Odyssey Imaging System (LiCor Inc).
Liver homogenates were collected four weeks after transfection, and ACMSD protein was
analyzed with Western blots as described in Methods. Quantifications are shown in Figure
4f. (b) Membranes were stripped and reprobed for loading control tubulin beta. This
experiment was performed once. (c) Representative Western blot image used to generate
protein quantifications shown in Figure 5d. HUVECs were treated with vehicle (water),
2PY (100 M), 4PY (100 M), or LPS (30 ng/mL), as indicated. Cell lysates were collected
after 6 hours of exposure, and VCAM-1 protein was analyzed with Western blots as
described in Methods. Membranes were cut, and the upper half were probed for VCAM-
1. (d) The lower half of membranes were probed for loading control GAPDH. This
experiment was performed three times, and a representative blot is shown.

24
Supplemental Figure 5: Plasma sVCAM-1 is associated with increased

prospective MACE risk.

(a) sVCAM-1 was quantitated in the US Validation Cohort. Kaplan-Meier estimates are

shown for the risk of incident 3-year MACE by quartiles of serum sVCAM-1 levels. (b)

Forest plot of Cox proportional hazard ratios for MACE for quartiles of plasma levels of

sVCAM-1 in human subjects. Cox models are shown both before (black) and after (red)

adjustments for traditional cardiovascular risk factors (age, sex, systolic blood pressure,

LDL, HDL, triglycerides, diabetes status) and hsCRP, current (active) smoking status,

alcohol use, and education level. Symbols represent hazard ratios and error bars

represent 95% confidence intervals. Ref, reference group.

25
Supplemental Figure 6: Impact of intraperitoneal 2PY and 4PY on serum 2PY, 4PY
levels.
Mice were treated intraperitoneally with 10 mg/kg of (a) 2PY or (b) 4PY in sterile saline,

and serial blood collections were performed. Serum 2PY and 4PY were measured with

LC/MS/MS. Serum concentrations are shown of 2PY (red) and 4PY (blue). Points and

error bars show the mean ± standard error. These results confirm that circulating levels

of 2PY and 4PY achieved are well within the ranges observed in the 4th quartile for both

2PY and 4PY of subjects in the US Validation Cohort. Further, the results show there is

no observed cross metabolism between the two structural isomers (i.e., injection of 4PY

only leads to systemic 4PY (and not 2PY) elevation, and conversely, injection of 2PY

only increases 2PY and not 4PY).

26
Supplemental References
1. Surendran, P. et al. Rare and common genetic determinants of metabolic
individuality and their effects on human health. Nat Med. 28, 2321–2332 (2022).
2. Chen, Y. et al. Genomic atlas of the plasma metabolome prioritizes metabolites
implicated in human diseases. Nat Genet. 55, 44–53 (2023).
3. Yin, X. et al. Genome-wide association studies of metabolites in Finnish men
identify disease-relevant loci. Nat Comm. 13 (1), 1–14 (2022).
4. Feofanova, E. et al. A genome-wide association study discovers 46 loci of the
human metabolome in the Hispanic Community Health Study/Study of Latinos.
Am J Human Genet. 107 (5), 849–863 (2020).
5. Tahir, U. et al. Whole genome association study of the plasma metabolome
identifies metabolites linked to cardiometabolic disease in Black individuals. Nat
Comm. 13 (1), 4923 (2022).
6. Shin, S. et al. An atlas of genetic influences on human blood metabolites. Nat
Genet. 46 (6), 543–550 (2014).
7. Sakaue, S. et al. A cross-population atlas of genetic associations for 220 human
phenotypes. Nat Genet. 53 (10), 1415–1424 (2021).
8. Sun, B. et al. Genomic atlas of the human plasma proteome. Nature. 558 (7708),
73–79 (2018).
9. Pietzner, M. et al. Mapping the proteo-genomic convergence of human diseases.
Science. 374 (6569), (2021).
10. Gudjonsson, G. et al. A genome-wide association study of serum proteins reveals
shared loci with common diseases. Nat Comm. 13 (1), 1–13 (2022).
11. Ferkingstad, E. et al. Large-scale integration of the plasma proteome with
genetics and disease. Nat Genet. 53 (12), 1712–1721 (2021).
12. Sun, B. et al. Plasma proteomic associations with genetics and health in the UK
Biobank. Nature. 622 (7982), 329-338 (2023).
13. Li, X. S. et al. Untargeted metabolomics identifies trimethyllysine, a TMAO-
producing nutrient precursor, as a predictor of incident cardiovascular disease
risk. JCI Insight (2018) doi:10.1172/jci.insight.99096.
14. Nemet, I. et al. A cardiovascular disease-linked gut microbial metabolite acts via
adrenergic receptors. Cell 180, 862-877.e22 (2020).
15. Wang, Z. et al. Gut flora metabolism of phosphatidylcholine promotes
cardiovascular disease. Nature 472, 57–65 (2011).
16. Buffa, J. A. et al. The microbial gbu gene cluster links cardiovascular disease risk
associated with red meat consumption to microbiota l-carnitine catabolism. Nat.
Microbiol. 7, 73–86 (2022).
17. Witkowski, M. et al. The artificial sweetener erythritol and cardiovascular event
risk. Nat. Med. (2023). DOI: 10.1038/s41591-023-02223-9

27
18. Li, S. et al. Predicting network activity from high throughput metabolomics. PloS
Comput. Biol. 9, (2013).
19. Enattah, NS. et al. Identification of a variant associated with adult-type
hypolactasia. Nat Genet. 30 (2), 233–237 (2021).
20. Hemani, G. et al. The MR-Base platform supports systematic causal inference
across the human phenome. Elife. 7 (2018) DOI: 10.7554/eLife.34408
21. Aragam, K. et al. Discovery and systematic characterization of risk variants and
genes for coronary artery disease in over a million participants. 54 (12), 1803-
1815 (2022)
22. Hartiala, J. et al. Genome-wide analysis identifies novel susceptibility loci for
myocardial infarction. Eur. Heart J. 42 (9), 919-933 (2021)
23. Mishra, A. et al. Stroke genetics informs drug discovery and risk prediction across
ancestries. 611 (7934), 115-123 (2022).
24. Lang, R. et al. Development of a hydrophilic liquid interaction chromatography-
high- performance liquid chromatography-tandem mass spectrometry based
stable isotope dilution analysis and pharmacokinetic studies on bioactive pyridines
in human plasma and urine after coffee. Anal. Chem. 82, 1486 – 1497 (2010).
25. Ruf, S. et al. Novel nicotinamide analog as inhibitor of nicotinamide N-
methyltransferase. Bioorganic Med. Chem. Lett. 28, 922–925 (2018).

28