Abstract

   Lysine propionylation modification (Kpr) plays an important role in the
   pathogenesis of several cardiovascular diseases, but the role of Kpr in
   postoperative atrial fibrillation (POAF) is unclear. Here, we
   established an atlas of proteomics and propionylation proteomics in the
   atrial appendage tissues from 28 CABG patients, exploring the role of
   Kpr proteins in the occurrence of POAF. The Kpr of ALDH6A1 was most
   significantly increased on Lys113 (2.25 folds). The activity of ALDH6A1
   increased due to higher binding energy of propionylated ALDH6A1 and
   NAD^+, causing an increase in NADH levels in cells and triggering
   abnormal energy metabolism. Furthermore, the increase in NADH levels
   triggered the accumulation of reactive oxygen species, which may cause
   oxidative stress, resulting in the development of AF. This study
   reveals the important role of ALDH6A1-NADH pathway in POAF, and
   provides new insights for exploring the pathogenesis of POAF in CABG.

   Subject terms: Atrial fibrillation, Translational research
     __________________________________________________________________

   The alterations of lysine propionylation modifications in the human
   right atrium reveal the important mechanism of ALDH6A1-NADH pathway in
   the pathogenesis of new-onset atrial fibrillation following coronary
   artery bypass grafting.

Introduction

   Coronary artery disease is the most common type of cardiovascular
   diseases^[36]1. New-onset postoperative atrial fibrillation (POAF) is a
   common complication after coronary artery bypass grafting (CABG) in
   patients with coronary artery disease^[37]2. The incidence of POAF is
   reported to be about 20–50%^[38]3–[39]5. POAF not only increases the
   risk of long-term atrial fibrillation (AF) in patients, but also leads
   to stroke, heart failure, and even death, seriously affecting the
   health of patients and the effect of surgical treatment of coronary
   artery disease^[40]6–[41]8. The occurrence of AF is affected by many
   factors, including age, gender, race, genetics, and a variety of other
   comorbidities such as hypertension, diabetes, heart failure, coronary
   artery disease, valvular disease, and obstructive sleep apnea-hypopnea
   syndrome^[42]9–[43]11. Studies have shown that the pathogenesis of AF
   is related to atrial electrical remodeling, structural remodeling,
   metabolic regulation, oxidative stress and dysregulation of ion
   channels such as sodium, potassium, and calcium channels^[44]12–[45]19.
   In particular, aging is the major risk factor for the occurrence of
   POAF^[46]1,[47]3,[48]4. In fact, the age for the POAF patients was 65.2
   years old vs. 61.8 years old in patients with normal heart rhythm in a
   large cohort of patients in a recent report^[49]4. Nevertheless, the
   specific pathogenesis of POAF is still unclear.

   Protein post-translational modifications (PTMs) are chemical changes on
   the amino acid side chains of proteins through covalent modification or
   proteolysis^[50]20,[51]21. PTMs are important regulatory mechanisms in
   organisms, which affects the activity, function, and stability of
   proteins by changing their structure, and plays an important role in
   cardiovascular diseases^[52]22–[53]24. With the development of omics
   technology, more and more novel lysine acylation modifications have
   been discovered, including succinylation, crotonylation,
   propionylation, butyrylation, malonylation, glutarylation,
   2-hydroxyisobutyrylation, and lactylation, which provide new evidence
   for the prevention, diagnosis, and treatment of cardiovascular
   disease^[54]25–[55]30.

   Lysine propionylation (Kpr) is a PTM that was first discovered in
   histones in 2007^[56]31. Propionyl-CoA is a key molecule in amino acid
   and odd-chain fatty acid catabolism, and it is the donor of Kpr^[57]32.
   Kpr has been found to exist in both histones and non-histone
   proteins^[58]33. Histone propionylation is a marker of active chromatin
   and provides novel epigenetic regulation for cell
   metabolism^[59]33,[60]34. In a MOZ-TIF2-driven mouse model of leukemia,
   the level of histone H3 propionylation at lysine 23 (H3K23pr) was
   elevated and closely associated with the regulation of gene
   transcriptional activity^[61]35. Propionylation of K198 and K203 of
   phosphate-sensing regulator (PhoP) reduced the transcriptional activity
   of PhoP in Saccharopolyspora erythraea^[62]36. A small-sample study
   revealed that the propionylation at lysine 17 of histone H2B (H2BK17pr)
   regulated proteostasis^[63]37. An analysis of Kpr in the intestinal
   samples of zebrafish showed that superoxide dismutase 2 (Sod2) was
   propionylated at K132, which was related to oxidative stress^[64]38.

   There are few studies on propionylation modifications in cardiovascular
   disease^[65]39. Branched-chain amino acids (BCAAs) metabolism is a
   major source of propionyl-CoA, participating in Kpr to regulate gene
   expression in cardiovascular diseases^[66]40,[67]41. Histone PTMs are
   important in epigenetics and play an important role in regulating gene
   transcriptional activity^[68]42. A recent study^[69]43 showed that
   reducing the concentration of BCAAs may reduce collagen gene expression
   and fibroblast proliferation in cardiac hypertrophy by modulating
   cellular epigenetics. Therefore, reducing BCAAs intake improves cardiac
   stress response in mice by decreasing H3K23pr level in the heart,
   thereby alleviating cardiac hypertrophy and heart failure^[70]43.
   Further, it was reported^[71]44 that BCAAs catabolism was an important
   regulator of platelets activation and was related to arterial
   thrombosis. The specific mechanism was to enhance the K255
   propionylation of tropomodulin-3, thereby promoting platelet activation
   and thrombosis^[72]44. However, the role of propionylation in AF
   (especially POAF) has not been reported yet.

   In this study, we constructed the proteomics and propionylation
   proteomics atlas of right atrial appendage tissues discarded during
   CABG surgery in both patients who later remained in sinus rhythm (SR)
   or developed POAF after CABG, and explored the correlation and
   mechanism by comparing the Kpr protein between the patients who
   developed POAF and postoperative sinus rhythm (POSR) after CABG.

Results

Association atlas of proteomics and propionylation proteomics in the right
atrial appendage tissues

   The discarded right atrial appendage tissues were collected during
   surgery for quantitative proteomics and propionylation proteomics
   analysis using the label-free 4D method, and the results of the two
   omics were analyzed by correlation (Fig. [73]1, Fig. [74]S1). All
   patients selected for this study were in SR before CABG. After CABG,
   the patients were allocated into the POSR group (SR, n = 5) or into
   POAF group (AF, n = 5), depending on their heart rhythm.

Fig. 1. Association atlas of proteomics and propionylation proteomics in
intraoperative right atrial appendage tissues of patients with POSR and POAF.

   [75]Fig. 1
   [76]Open in a new tab

   a The intersection of all proteins identified by the two omics (n = 5
   per group). b Distribution of all protein abundance quantified by the
   two omics. c Nine-quadrant diagram of comparative analysis of
   differentially expressed proteins between the two omics. Different
   colors represented different expression patterns. d Distribution
   statistics of differential proteins and differential sites. The left
   side was the proteomics results, and the right side was the
   propionylation proteomics results. e Venn diagram for comparative
   analysis of differentially expressed proteins between the two omics.
   The numbers in the figure represented the number of proteins contained
   in the intersection. f Heat map of functional enrichment differences
   between the two omics of differently classified proteins on KEGG
   pathway. The redder the color, the more significant the enrichment.

   A total of 5002 proteins were identified in the proteomics and 800
   proteins were identified in the propionylation proteomics, of which 740
   proteins were overlapped (Fig. [77]1a). The histogram showed the
   distribution of the intensity values of all quantified proteins in the
   proteomics and propionylation proteomics (Fig. [78]1b). There were 108
   up-regulated proteins and 25 down-regulated proteins in the proteomics.
   In the propionylation proteomics, 16 proteins were up-regulated and 73
   proteins were down-regulated. The nine-quadrant diagram, bar chart and
   venn diagram all showed the distribution of all differentially
   expressed proteins in the two omics (Fig. [79]1c–e). In addition, we
   used heat maps to visualize the connections between different
   classified proteins in the two omics in terms of GO functions
   (Fig. [80]S1a–c), KEGG pathways (Fig. [81]1f) and domain binding
   (Fig. [82]S1d). Detailed data of the proteomics results are shown in
   Fig. [83]S2.

Analysis of lysine-propionylated proteins

   The propionylation proteomics data were normalized based on the
   proteomics data of the same sample, thus eliminating the effect of
   protein expression on modification level. There were 3018 modification
   sites on 800 proteins identified, of which 2129 modification sites on
   557 proteins were available for quantitative comparison. Compared with
   the SR group, 19 Kpr sites of 16 proteins were significantly
   up-regulated (AF/SR > 1.5, P < 0.05) and 118 Kpr sites of 73 proteins
   were significantly down-regulated (AF/SR < 1/1.5, P < 0.05) in the AF
   group. The volcano plot visualized the distribution of each
   differential Kpr sites between the two groups, where the top 5
   differentially expressed Kpr sites in up- and down-regulation were
   labeled (Fig. [84]2a). The heatmap demonstrated the relative expression
   of all differential Kpr sites in different samples, with red color
   representing high expression and blue color representing low expression
   (Fig. [85]2b). Peptide sequences of 10 amino acids up-stream and
   down-stream of all Kpr sites were identified using MoMo analysis based
   on the motify-x algorithm (−10 to +10), and the frequency of occurrence
   of 20 amino acids near the Kpr sites was presented in the form of a
   heat map (Fig. [86]2c). In addition, a total of 9 Kpr motifs were
   identified in 1677 modified peptides (Fig. [87]2d).

Fig. 2. Proteomics analysis of Kpr in intraoperative right atrial appendage
tissues from patients with POSR and POAF.

   [88]Fig. 2
   [89]Open in a new tab

   a The volcano plot of differential Kpr sites. The red points indicated
   significant up-regulation, and the blue points indicated significant
   down-regulation. The information on the differentially modified sites
   for Top5 up- and down-regulation were also marked in the figure,
   respectively. b The heat map of differential Kpr sites. Red represented
   high expression and blue represented low expression. c Motify analysis
   heat map of 10 amino acids upstream and downstream of the modification
   sites. Yellow or green indicated upstream or downstream, respectively.
   d The diagram of motif logo showed the flanking sequence preferences