Lung endothelium serves as a physical barrier to blood constituents and acts as a gate keeper of organ inflammation (Galley and Webster, 2004; Pries and Kuebler, 2006). Although largely quiescent in the healthy lungs, endothelial cells are an early participant in the pathogenesis of many inflammatory conditions, such as the acute respiratory distress syndrome (ARDS). For instance, endothelial cells produce many of the factors that recruit leukocytes to the lung after injury and critically control immune cell entry through the expression of membrane adhesion molecules (Maniatis et al., 2008; Millar et al., 2016). Thus, tight control over endothelial responses is critical for limiting organ damage to most inflammatory insults, and interventions that attenuate activation of these cells have emerged as a new paradigm in the treatment of inflammatory diseases (Orfanos et al., 2004).

Canonical NF-κB (nuclear factor kappa enhancer binding protein) signaling plays a pivotal role in endothelial cell activation (Baeuerle, 1998). Binding of proinflammatory cytokines to receptors, such as members of interleukin-1 receptor (IL-1R)/toll-like receptor (TLR) superfamily, leads to the activation of the IKK (IκB [NF-κB inhibitor] kinase) complex via recruiting cytoplasmic adaptors, including TNF receptor-associated factors (TRAFs). The activation of the IKK complex causes phosphorylation of IκBα, which leads to polyubiquitination and subsequent proteasomal degradation. The liberated NF-κB dimer then translocates to the nucleus wherein it binds to specific DNA sequences to promote the expression of cytokines and endothelial adhesion molecules (Dunne and O’Neill, 2003; Szmitko et al., 2003). TRAF6 is a RING-domain cognate ubiquitin ligase (E3) associated with the intracellular domain of the IL-1R/TLR that mediates immune responses through lysine 63 (K63)-linked ubiquitination of diverse signaling intermediators (Chen, 2005; Deng et al., 2000). Negative regulators of TRAF6 have been shown to suppress cellular inflammation, suggesting clinical potential in the treatment of inflammatory diseases (Lv et al., 2018; Wang et al., 2006).

As another hallmark of the vascular inflammatory process, leukocyte-endothelial adhesion is mediated by interactions between endothelial adhesion molecules and cognate receptors on immune cells (Szmitko et al., 2003). Intercellular adhesion molecule-1 (ICAM-1; CD54) is an inducible transmembrane protein expressed on the surface of endothelial cells, with known roles in leukocyte adhesion and transmigration (Springer, 1994; Staunton et al., 1988). In addition, ligand-induced ICAM-1 cross-linking acts as a signal transducer to promote proinflammatory effects (Lawson and Wolf, 2009). ICAM-1 is heavily N-linked glycosylated on extracellular Ig-like domains, and previous studies have demonstrated that this post-translational modification (PTM) is indispensable for the proper conformation and biological function of ICAM-1 during inflammatory responses (Chen et al., 2014b). However, the molecular mechanisms underlying the regulation of ICAM-1 glycosylation remain largely unknown.

As a conserved protein repair enzyme, PIMT utilizes S-adenosyl methionine (SAM/AdoMet) to form methyl esters on carboxyl positions of isoaspartyl (isoAsp) residues. IsoAsp is generated from asparagine (Asn) deamidation or aspartate (Asp) dehydration via succinimide intermediate under physiological or stressed conditions and usually disrupts protein structure and/or function (Geiger and Clarke, 1987; Reissner and Aswad, 2003). The methylated intermediate then releases a methyl group to reform succinimide through spontaneous decomposition, which converts isoAsp to Asp in order to restore protein function (Desrosiers and Fanélus, 2011; Mishra and Mahawar, 2019; Reissner and Aswad, 2003). It is increasingly recognized that Asn deamidation is significantly engaged in diverse biological functions such as cell apoptosis and the control of metabolic responses under inflammatory and stress conditions (Deverman et al., 2002; Zhao et al., 2020). PIMT-mediated protein carboxyl methylation (O-methylation) was recently reported to be critically important for cells, indicating it may functionally rival other types of protein PTMs (Sprung et al., 2008). For instance, overexpression of PIMT has been shown to extend longevity of bacteria, fruit flies, and plants (Chavous et al., 2001; Ogé et al., 2008; Pesingi et al., 2017). PIMT deficient mice exhibit increased isoAsp residual deposition and early neurological death due to grand mal seizures (Kim et al., 1997). Furthermore, we and others have shown that PIMT plays an important role in angiogenesis and cardiomyocyte survival under oxidative conditions (Ouanouki and Desrosiers, 2016; Yan et al., 2013).

Here, we demonstrate that PIMT is a potent negative regulator of endothelial activation and vascular inflammation. Mechanistically, we show that PIMT acts by inhibiting LPS-induced NF-κB activation through methylation of TRAF6 in ECs and by physically interacting with ICAM-1 to impact N-glycosylation and its functional interactions with immune cells. In summary, our findings suggest that PIMT-mediated protein O-methylation plays a central role in controlling vascular inflammation in the lung.

PIMT (encoded by Protein-L-Isoaspartate [D-Aspartate] O-Methyltransferase [Pcmt1] gene) is a protein carboxyl methyltransferase that specifically recognizes structurally altered carboxylic acids at aspartyl residues, and its role in monitoring protein homeostasis has been defined as ‘housekeeping’ function in a wide variety of cells and tissues (Desrosiers and Fanélus, 2011; Mishra and Mahawar, 2019; Reissner and Aswad, 2003). In this study, we identified a novel function of PIMT in suppressing endothelial activation and acute inflammatory responses. Specifically, we found that PIMT attenuates LPS-stimulated endothelial activation and acute lung injury by inhibiting NF-κB signaling through methylation of TRAF6 and inhibition of N-glycosylation of the cell surface adhesion molecule ICAM-1.

TRAF6 plays key roles in eliciting IL-1R/TLR-mediated inflammatory signaling (Akira and Takeda, 2004; Chen, 2005; Deng et al., 2000). TRAF6 consists of an N-terminal RING domain, followed by four ZnF domains, a CC domain, and a TRAF-C domain at the C-terminus (Chung et al., 2002). The RING domain, in association with ZnF domains, interacts with Ub-conjugating enzymes (E2) for Ub transfer (Yin et al., 2009). The TRAF-C domain is a binding scaffold mediating TRAF6 interaction with various effectors (Ye et al., 2002). Moreover, the CC domain has also been shown to be indispensable for TRAF6 oligomerization and subsequent activation of TAK1 complex (Wang et al., 2006). Recently, the CC domain has been shown to enhance TRAF6 activation through interacting with E2 Ubc13 (Hu et al., 2017). However, little is known about how the function of TRAF6 CC domain is regulated under inflammatory conditions. Herein, we provide compelling evidence that PIMT-mediated methylation of TRAF6 in the CC-domain represents a novel mechanism in the regulation of TRAF6/NF-κB signaling in vascular ECs. Overexpression of PIMT markedly impeded TRAF6 oligomerization and autoubiquitination, which is known to be a key process in the initiation of NF-κB activation in ECs. Furthermore, we identified Asn residue 350 in the CC domain of TRAF6 as the site of TRAF6 methylation by PIMT in ECs. Whether PIMT affects TRAF6 function through acting on Asn/Asp residues in other domains remains unknown and needs to be further investigated.

Deamidation of Asn is the first step in the formation of PIMT recognizable intermediates, such as isoAsp, succinimide, and methyl-isoAsp, and represents an important mechanism by which cells recognize deranged proteins during development and/or aging of cells (Mishra and Mahawar, 2019; Reissner and Aswad, 2003; Lee et al., 2012). Our data demonstrated that Asn substitution with Ala impaired TRAF6 susceptibility to PIMT-mediated inhibition but did not alter intrinsic functions, strongly suggesting that PIMT catalyzed methylation, rather than deamidated end product repair, is the primary mechanism coordinating TRAF6 homeostasis. Interestingly, we found that the N350 residue in the CC domain is localized in a loop site adjacent to TRAF-C domain, which is exposed to the protein surface and susceptible to deamidation (Figure 4—figure supplement 1C; Ye et al., 2002). Based on these findings, we hypothesize that PIMT-mediated N350 methylation serves as an intrinsic negative regulator of TRAF6 biological activity. It is worth noting that our study primarily focused on functional interactions of PIMT1 and TRAF6 because TRAF6 is critically implicated in LPS-induced endothelial activation and acute lung injury (Chen, 2005; Martin and Wurfel, 2008). Nevertheless, our data also demonstrated a weak interaction between PIMT and TRAF2. Since a conservative deamidation susceptible Asn site is identified in the TRAF-C domain of TRAF2, we cannot rule out the possibility that PIMT1-mediated O-methylation is involved in inhibiting TRAF2 activation. Furthermore, whether PIMT affects TRAF6 function through modification of Asn/Asp residues in other domains remains unknown and warrants further investigation.

ECs are regarded as sentinels for sensing invading pathogens and triggering innate immune system activation; however, hyperactive inflammatory responses by ECs can have detrimental effects on the host. Hence, confining inflammatory reactions is imperative for governing the magnitude and duration of endothelial activation and immune responses. Indeed, several molecules, such as IκBs, deubiquitinase A20, transcriptional co-factor CITED2, and protein chaperone cyclophilin J, have been identified as critical negative regulators of inflammatory signals under various conditions (Boone et al., 2004; Lou et al., 2011; O’Dea and Hoffmann, 2010; Sheng et al., 2018). LPS treatment should cause TRAF6 to become more prone to spontaneous modifications. This is because LPS is known to trigger production of reactive oxygen species (ROS; Hsu and Wen, 2002), and ROS is known to induce isoAsp generation (D’Angelo et al., 2001). In addition to altering the backbone, isoAsp modifications will have effects on protein conformation and PTMs, thus having either positive (e.g. p53) or negative (e.g. tubulin) downstream effects on protein functions (Lee et al., 2012; Biterge et al., 2014). In our study, we found that PIMT knockdown enhances TRAF6/NF-κB activation, suggesting that isoAsp modifications serve to overactivate TRAF6 function. IsoAsp repair is traditionally recognized as a spontaneous process. However, recent studies show that O-methylation of isoAsp occurs in nature, including on proteins such as p53 and histone H4 (Lee et al., 2012; Biterge et al., 2014), and is important for exerting biological functions. Notably, a recent publication demonstrated that O-methylation of p53 is not a transient event and is maintained on this protein for at least 24 hr (Lee et al., 2012). LPS treatment did not alter PIMT levels under either in vitro or in vivo conditions but significantly augmented its interaction with TRAF6 in different EC types, suggesting that the dynamic formation of PIMT/TRAF6 protein complexes constitutes an important negative feedback mechanism in resolving LPS inflammation. It should be noted that in addition to ECs, PIMT is also expressed in other type of cells, such as macrophage and epithelial cells in the lung. Future studies using tissue specific PIMT knockout mice will help to determine the relative contribution of PIMT in different cell types to ALI.

As a hallmark of endothelial activation, cell-surface expression of adhesion molecules mediates the interaction of circulating leukocytes with endothelium in response to proinflammatory signals (Hunt and Jurd, 1998). Adhesion molecules, such as ICAM-1 and VCAM-1, are heavily N-glycosylated, and N-glycosylation defects have been shown to impair expression and membrane localization of adhesion molecules and subsequent intracellular, cell–cell, and cell–matrix recognition events in inflammatory responses (Chen et al., 2014b; Scott and Patel, 2013). N-glycan attaches only to Asn and presents in an N-x-S/T consensus sequence. At this time, little is known about how N-glycosylation of adhesion molecules is regulated under inflammatory conditions in ECs. Here, we show that PIMT affects inflammatory responses through modulating its N-glycosylation of ICAM-1. We found that PIMT differentially regulates the N-glycosylation of ICAM-1 without altering the global endothelial glycocalyx. To further investigate molecular mechanisms underlying inhibition of adhesion molecule N-glycosylation by PIMT, we chose to determine the structure of the N-glycans of ICAM-1 using mass spectrometry. ICAM-1 possesses eight N-glycosylation sites. We found that PIMT inhibits ICAM-1 N-glycosylation at four specific Asn sites (N145, N183, N267, and N385). Furthermore, our mass spectrometric data show that Asn residues at these sites are switched to Asp residues by PIMT, thus preventing N-glycosylation. Our results also suggest that the enzymatic activity of PIMT may be necessary for the O-methylation of isoAsp derived from Asn to form Asp in ICAM-1. In addition to the decreasing ICAM-1 protein stability, as shown in this study, we suspect that deficiency of ICAM-1 N-glycosylation may directly impact the function of ICAM-1, despite our data showing that LMW form of ICAM-1 can still localize to the cell membrane. As shown in Figure 5H, the N145 and N183 sites are localized in the hydrophilic BC and EF loops of ICAM-1 domain 2 (D2) upon structural dissection (Casasnovas et al., 1998). Based on the fact that ICAM-1 D2 sustains the structural and LFA1 binding integrity of D1 (Stanley et al., 2000), we speculate that PIMT-induced ‘site mutation’ at these sites may impair the binding capacity of ICAM-1 to its ligand, and this hypothesis was partially testified in our ICAM-1 antibody cross-linking experiments. Furthermore, lack of ICAM-1 glycosylation at N267 in the D3 domain may also affect interactions with macrophage-1 antigen (Mac-1; Diamond et al., 1993). In this regard, our study provides novel insights into how PIMT-mediated N-glycosylation regulates the functional interaction of ECs and immune cells in vascular inflammation. The reason why PIMT specifically acts on these four specific Asn sites remains elusive. It is attempted to speculate that isoAsp repair and maintenance, just like other protein modifications, may occur in a context and cell type dependent fashion. In our IP experiment, only small fractions of PIMT co-IP’d with LMW ICAM-1. Although we suspect this could be explained by a ‘hit and run’ mechanism. Furthermore, we cannot exclude the possibility that PIMT may affect any one of the many steps along the pathway to protein glycosylation, such as ERQC or glucoside attachment. Elucidation of molecular mechanisms underlying PIMT-induced glycosylation of VCAM-1 in ECs is ongoing.

In summary, our work has identified novel mechanisms in the control of agonist-induced EC activation by PIMT. We demonstrate that PIMT-catalyzed methylation directly restricts TRAF6 activation to maintain TRAF6 in the default static state. Moreover, PIMT provides a ‘double check’ mechanism to avoid endothelial activation by limiting ICAM-1 membrane expression and intracellular signaling through transcriptional and post-translational controls (Figure 6G). Altogether, these results suggest that targeted activation of PIMT may provide a novel therapeutic strategy for treating a wide range of inflammatory vascular disorders.

All data generated or analysed during this study are included in the manuscript and supporting file; Source Data files have been provided for Figures 1-6.

1. 1. Ahn K
   2. Pan S
   3. Beningo K
   4. Hupe D

   (1995) A permanent human cell line (EA.hy926) preserves the characteristics of endothelin converting enzyme from primary human umbilical vein endothelial cells

   Life Sciences 56:2331–2341.

   https://doi.org/10.1016/0024-3205(95)00227-w

   • PubMed
   • Google Scholar
2. 1. Akbar AN
   2. Gilroy DW

   (2020) Aging immunity may exacerbate COVID-19

   Science 369:256–257.

   https://doi.org/10.1126/science.abb0762

   • PubMed
   • Google Scholar
3. 1. Akira S
   2. Takeda K

   (2004) Toll-Like receptor signalling

   Nature Reviews. Immunology 4:499–511.

   https://doi.org/10.1038/nri1391

   • PubMed
   • Google Scholar
4. 1. Baeuerle PA

   (1998) IkappaB-NF-kappab structures: at the interface of inflammation control

   Cell 95:729–731.

   https://doi.org/10.1016/s0092-8674(00)81694-3

   • PubMed
   • Google Scholar
5. 1. Biterge B
   2. Richter F
   3. Mittler G
   4. Schneider R

   (2014) Methylation of histone H4 at aspartate 24 by protein L-isoaspartate O-methyltransferase (PCMT1) links histone modifications with protein homeostasis

   Scientific Reports 4:6674.

   https://doi.org/10.1038/srep06674

   • PubMed
   • Google Scholar
6. 1. Boone DL
   2. Turer EE
   3. Lee EG
   4. Ahmad R-C
   5. Wheeler MT
   6. Tsui C
   7. Hurley P
   8. Chien M
   9. Chai S
   10. Hitotsumatsu O
   11. McNally E
   12. Pickart C
   13. Ma A

   (2004) The ubiquitin-modifying enzyme A20 is required for termination of Toll-like receptor responses

   Nature Immunology 5:1052–1060.

   https://doi.org/10.1038/ni1110

   • PubMed
   • Google Scholar
7. 1. Casasnovas JM
   2. Stehle T
   3. Liu JH
   4. Wang JH
   5. Springer TA

   (1998) A dimeric crystal structure for the N-terminal two domains of intercellular adhesion molecule-1

   PNAS 95:4134–4139.

   https://doi.org/10.1073/pnas.95.8.4134

   • PubMed
   • Google Scholar
8. 1. Chavous DA
   2. Jackson FR
   3. O’Connor CM

   (2001) Extension of the Drosophila lifespan by overexpression of a protein repair methyltransferase

   PNAS 98:14814–14818.

   https://doi.org/10.1073/pnas.251446498

   • PubMed
   • Google Scholar
9. 1. Chen ZJ

   (2005) Ubiquitin signalling in the NF-kappaB pathway

   Nature Cell Biology 7:758–765.

   https://doi.org/10.1038/ncb0805-758

   • PubMed
   • Google Scholar
10. 1. Chen HF
    2. Huang CH
    3. Liu CJ
    4. Hung JJ
    5. Hsu CC
    6. Teng SC
    7. Wu KJ

    (2014a) Twist1 induces endothelial differentiation of tumour cells through the jagged1-KLF4 axis

    Nature Communications 5:4697.

    https://doi.org/10.1038/ncomms5697

    • PubMed
    • Google Scholar
11. 1. Chen M
    2. Yi B
    3. Sun J

    (2014b) Inhibition of cardiomyocyte hypertrophy by protein arginine methyltransferase 5

    The Journal of Biological Chemistry 289:24325–24335.

    https://doi.org/10.1074/jbc.M114.577494

    • PubMed
    • Google Scholar
12. 1. Choo YS
    2. Zhang Z

    (2009) Detection of protein ubiquitination

    Journal of Visualized Experiments 1:1293.

    https://doi.org/10.3791/1293

    • PubMed
    • Google Scholar
13. 1. Chung JY
    2. Park YC
    3. Ye H
    4. Wu H

    (2002) All TRAFs are not created equal: common and distinct molecular mechanisms of TRAF-mediated signal transduction

    Journal of Cell Science 115:679–688.

    https://doi.org/10.1242/jcs.115.4.679

    • PubMed
    • Google Scholar
14. 1. Dai S
    2. Ni W
    3. Patananan AN
    4. Clarke SG
    5. Karger BL
    6. Zhou ZS

    (2013) Integrated proteomic analysis of major isoaspartyl-containing proteins in the urine of wild type and protein L-isoaspartate O-methyltransferase-deficient mice

    Analytical Chemistry 85:2423–2430.

    https://doi.org/10.1021/ac303428h

    • PubMed
    • Google Scholar
15. 1. D’Angelo S
    2. Ingrosso D
    3. Perfetto B
    4. Baroni A
    5. Zappia M
    6. Lobianco LL
    7. Tufano MA
    8. Galletti P

    (2001) Uva irradiation induces L-isoaspartyl formation in melanoma cell proteins

    Free Radical Biology & Medicine 31:1–9.

    https://doi.org/10.1016/s0891-5849(01)00518-4

    • PubMed
    • Google Scholar
16. 1. Dauphinee SM
    2. Karsan A

    (2006) Lipopolysaccharide signaling in endothelial cells

    Laboratory Investigation; a Journal of Technical Methods and Pathology 86:9–22.

    https://doi.org/10.1038/labinvest.3700366

    • PubMed
    • Google Scholar
17. 1. Deng L
    2. Wang C
    3. Spencer E
    4. Yang L
    5. Braun A
    6. You J
    7. Slaughter C
    8. Pickart C
    9. Chen ZJ

    (2000) Activation of the IkappaB kinase complex by TRAF6 requires a dimeric ubiquitin-conjugating enzyme complex and a unique polyubiquitin chain

    Cell 103:351–361.

    https://doi.org/10.1016/s0092-8674(00)00126-4

    • PubMed
    • Google Scholar
18. 1. Desrosiers RR
    2. Fanélus I

    (2011)

    Damaged proteins bearing L-isoaspartyl residues and aging: a dynamic equilibrium between generation of isomerized forms and repair by PIMT

    Current Aging Science 4:8–18.

    • PubMed
    • Google Scholar
19. 1. Deverman BE
    2. Cook BL
    3. Manson SR
    4. Niederhoff RA
    5. Langer EM
    6. Rosová I
    7. Kulans LA
    8. Fu X
    9. Weinberg JS
    10. Heinecke JW
    11. Roth KA
    12. Weintraub SJ

    (2002) Bcl-Xl deamidation is a critical switch in the regulation of the response to DNA damage

    Cell 111:51–62.

    https://doi.org/10.1016/s0092-8674(02)00972-8

    • PubMed
    • Google Scholar
20. 1. Diamond MS
    2. Garcia-Aguilar J
    3. Bickford JK
    4. Corbi AL
    5. Springer TA

    (1993) The I domain is a major recognition site on the leukocyte integrin Mac-1 (CD11b/CD18) for four distinct adhesion ligands

    The Journal of Cell Biology 120:1031–1043.

    https://doi.org/10.1083/jcb.120.4.1031

    • PubMed
    • Google Scholar
21. 1. Dunne A
    2. O’Neill LAJ

    (2003) The interleukin-1 receptor/toll-like receptor superfamily: signal transduction during inflammation and host defense

    Science’s STKE 1:re3.

    https://doi.org/10.1126/stke.2003.171.re3

    • PubMed
    • Google Scholar
22. 1. Galley HF
    2. Webster NR

    (2004) Physiology of the endothelium

    British Journal of Anaesthesia 93:105–113.

    https://doi.org/10.1093/bja/aeh163

    • PubMed
    • Google Scholar
23. 1. Geiger T
    2. Clarke S

    (1987)

    Deamidation, isomerization, and racemization at asparaginyl and aspartyl residues in peptides: Succinimide-linked reactions that contribute to protein degradation

    The Journal of Biological Chemistry 262:785–794.

    • PubMed
    • Google Scholar
24. 1. Hsu HY
    2. Wen MH

    (2002) Lipopolysaccharide-Mediated reactive oxygen species and signal transduction in the regulation of interleukin-1 gene expression

    The Journal of Biological Chemistry 277:22131–22139.

    https://doi.org/10.1074/jbc.M111883200

    • PubMed
    • Google Scholar
25. 1. Hu L
    2. Xu J
    3. Xie X
    4. Zhou Y
    5. Tao P
    6. Li H
    7. Han X
    8. Wang C
    9. Liu J
    10. Xu P
    11. Neculai D
    12. Xia Z

    (2017) Oligomerization-primed coiled-coil domain interaction with Ubc13 confers processivity to TRAF6 ubiquitin ligase activity

    Nature Communications 8:814.

    https://doi.org/10.1038/s41467-017-01290-0

    • PubMed
    • Google Scholar
26. 1. Hunt BJ
    2. Jurd KM

    (1998) Endothelial cell activation: A central pathophysiological process

    BMJ 316:1328–1329.

    https://doi.org/10.1136/bmj.316.7141.1328

    • PubMed
    • Google Scholar
27. 1. Irie T
    2. Muta T
    3. Takeshige K

    (2000) Tak1 mediates an activation signal from Toll-like receptor (S) to nuclear factor-kappaB in lipopolysaccharide-stimulated macrophages

    FEBS Letters 467:160–164.

    https://doi.org/10.1016/s0014-5793(00)01146-7

    • PubMed
    • Google Scholar
28. Book

    1. Jørgensen MM
    2. Bross P
    3. Gregersen N

    (2003)

    Protein Quality Control in the Endoplasmic Reticulum

    APMIS.

    • Google Scholar
29. 1. Kagan RM
    2. Clarke S

    (1994) Widespread occurrence of three sequence motifs in diverse S-adenosylmethionine-dependent methyltransferases suggests a common structure for these enzymes

    Archives of Biochemistry and Biophysics 310:417–427.

    https://doi.org/10.1006/abbi.1994.1187

    • PubMed
    • Google Scholar
30. 1. Kang SJ
    2. Jung SI

    (2020) Age-Related morbidity and mortality among patients with COVID-19

    Infection & Chemotherapy 52:154–164.

    https://doi.org/10.3947/ic.2020.52.2.154

    • PubMed
    • Google Scholar
31. 1. Kim E
    2. Lowenson JD
    3. MacLaren DC
    4. Clarke S
    5. Young SG

    (1997) Deficiency of a protein-repair enzyme results in the accumulation of altered proteins, retardation of growth, and fatal seizures in mice

    PNAS 94:6132–6137.

    https://doi.org/10.1073/pnas.94.12.6132

    • PubMed
    • Google Scholar
32. 1. Lawrence T

    (2009) The nuclear factor NF-kappaB pathway in inflammation

    Cold Spring Harbor Perspectives in Biology 1:a001651.

    https://doi.org/10.1101/cshperspect.a001651

    • PubMed
    • Google Scholar
33. 1. Lawson C
    2. Ainsworth M
    3. Yacoub M
    4. Rose M

    (1999)

    Ligation of ICAM-1 on endothelial cells leads to expression of VCAM-1 via a nuclear factor-kappaB-independent mechanism

    Journal of Immunology 162:2990–2996.

    • PubMed
    • Google Scholar
34. 1. Lawson C
    2. Wolf S

    (2009) Icam-1 signaling in endothelial cells

    Pharmacological Reports 61:22–32.

    https://doi.org/10.1016/s1734-1140(09)70004-0

    • PubMed
    • Google Scholar
35. 1. Lee J-C
    2. Kang S-U
    3. Jeon Y
    4. Park JW
    5. You J-S
    6. Ha S-W
    7. Bae N
    8. Lubec G
    9. Kwon SH
    10. Lee J-S
    11. Cho E-J
    12. Han J-W

    (2012) Protein L-isoaspartyl methyltransferase regulates p53 activity

    Nature Communications 3:927.

    https://doi.org/10.1038/ncomms1933

    • PubMed
    • Google Scholar
36. 1. Lou X
    2. Sun S
    3. Chen W
    4. Zhou Y
    5. Huang Y
    6. Liu X
    7. Shan Y
    8. Wang C

    (2011) Negative feedback regulation of NF-κB action by Cited2 in the nucleus

    Journal of Immunology 186:539–548.

    https://doi.org/10.4049/jimmunol.1001650

    • PubMed
    • Google Scholar
37. 1. Lv Y
    2. Kim K
    3. Sheng Y
    4. Cho J
    5. Qian Z
    6. Zhao YY
    7. Hu G
    8. Pan D
    9. Malik AB
    10. Hu G

    (2018) Yap controls endothelial activation and vascular inflammation through TRAF6

    Circulation Research 123:43–56.

    https://doi.org/10.1161/CIRCRESAHA.118.313143

    • PubMed
    • Google Scholar
38. 1. Maniatis NA
    2. Kotanidou A
    3. Catravas JD
    4. Orfanos SE

    (2008) Endothelial pathomechanisms in acute lung injury

    Vascular Pharmacology 49:119–133.

    https://doi.org/10.1016/j.vph.2008.06.009

    • PubMed
    • Google Scholar
39. 1. Martin TR
    2. Wurfel MM

    (2008) A triffic perspective on acute lung injury

    Cell 133:208–210.

    https://doi.org/10.1016/j.cell.2008.04.006

    • PubMed
    • Google Scholar
40. 1. Mellquist JL
    2. Kasturi L
    3. Spitalnik SL
    4. Shakin-Eshleman SH

    (1998) The amino acid following an Asn-X-Ser/Thr sequon is an important determinant of N-linked core glycosylation efficiency

    Biochemistry 37:6833–6837.

    https://doi.org/10.1021/bi972217k

    • PubMed
    • Google Scholar
41. 1. Millar FR
    2. Summers C
    3. Griffiths MJ
    4. Toshner MR
    5. Proudfoot AG

    (2016) The pulmonary endothelium in acute respiratory distress syndrome: insights and therapeutic opportunities

    Thorax 71:462–473.

    https://doi.org/10.1136/thoraxjnl-2015-207461

    • PubMed
    • Google Scholar
42. 1. Mishra PKK
    2. Mahawar M

    (2019) PIMT-mediated protein repair: mechanism and implications

    Biochemistry. Biokhimiia 84:453–463.

    https://doi.org/10.1134/S0006297919050018

    • PubMed
    • Google Scholar
43. 1. Nottingham LK
    2. Yan CH
    3. Yang X
    4. Si H
    5. Coupar J
    6. Bian Y
    7. Cheng T-F
    8. Allen C
    9. Arun P
    10. Gius D
    11. Dang L
    12. Van Waes C
    13. Chen Z

    (2014) Aberrant IKKα and IKKβ cooperatively activate NF-κB and induce EGFR/AP1 signaling to promote survival and migration of head and neck cancer

    Oncogene 33:1135–1147.

    https://doi.org/10.1038/onc.2013.49

    • PubMed
    • Google Scholar
44. 1. O’Dea E
    2. Hoffmann A

    (2010) The regulatory logic of the NF-kappaB signaling system

    Cold Spring Harbor Perspectives in Biology 2:a000216.

    https://doi.org/10.1101/cshperspect.a000216

    • PubMed
    • Google Scholar
45. 1. Ogé L
    2. Bourdais G
    3. Bove J
    4. Collet B
    5. Godin B
    6. Granier F
    7. Boutin JP
    8. Job D
    9. Jullien M
    10. Grappin P

    (2008) Protein repair L-isoaspartyl methyltransferase 1 is involved in both seed longevity and germination vigor in Arabidopsis

    The Plant Cell 20:3022–3037.

    https://doi.org/10.1105/tpc.108.058479

    • PubMed
    • Google Scholar
46. 1. Orfanos SE
    2. Mavrommati I
    3. Korovesi I
    4. Roussos C

    (2004) Pulmonary endothelium in acute lung injury: from basic science to the critically ill

    Intensive Care Medicine 30:1702–1714.

    https://doi.org/10.1007/s00134-004-2370-x

    • PubMed
    • Google Scholar
47. 1. Ouanouki A
    2. Desrosiers RR

    (2016) The enzyme L-isoaspartyl (D-aspartyl) methyltransferase is required for VEGF-dependent endothelial cell migration and tubulogenesis

    Molecular and Cellular Biochemistry 413:37–46.

    https://doi.org/10.1007/s11010-015-2637-2

    • PubMed
    • Google Scholar
48. 1. Pesingi PK
    2. Kumawat M
    3. Behera P
    4. Dixit SK
    5. Agarwal RK
    6. Goswami TK
    7. Mahawar M

    (2017) Protein-L-isoaspartyl methyltransferase (PIMT) is required for survival of Salmonella typhimurium at 42°C and contributes to the virulence in poultry

    Frontiers in Microbiology 8:361.

    https://doi.org/10.3389/fmicb.2017.00361

    • PubMed
    • Google Scholar
49. 1. Pober JS
    2. Sessa WC

    (2007) Evolving functions of endothelial cells in inflammation

    Nature Reviews. Immunology 7:803–815.

    https://doi.org/10.1038/nri2171

    • PubMed
    • Google Scholar
50. 1. Pries AR
    2. Kuebler WM

    (2006) Normal endothelium

    Experimental Pharmacology 1:1–40.

    https://doi.org/10.1007/3-540-32967-6_1

    • PubMed
    • Google Scholar
51. 1. Reissner KJ
    2. Aswad DW

    (2003) Deamidation and isoaspartate formation in proteins: unwanted alterations or surreptitious signals?

    Cellular and Molecular Life Sciences 60:1281–1295.

    https://doi.org/10.1007/s00018-003-2287-5

    • PubMed
    • Google Scholar
52. 1. Scott DW
    2. Patel RP

    (2013) Endothelial heterogeneity and adhesion molecules N-glycosylation: implications in leukocyte trafficking in inflammation

    Glycobiology 23:622–633.

    https://doi.org/10.1093/glycob/cwt014

    • PubMed
    • Google Scholar
53. 1. Sheng C
    2. Yao C
    3. Wang Z
    4. Chen H
    5. Zhao Y
    6. Xu D
    7. Huang H
    8. Huang W
    9. Chen S

    (2018) Cyclophilin J limits inflammation through the blockage of ubiquitin chain sensing

    Nature Communications 9:4381.

    https://doi.org/10.1038/s41467-018-06756-3

    • PubMed
    • Google Scholar
54. 1. Shimaoka M
    2. Xiao T
    3. Liu J-H
    4. Yang Y
    5. Dong Y
    6. Jun C-D
    7. McCormack A
    8. Zhang R
    9. Joachimiak A
    10. Takagi J
    11. Wang J-H
    12. Springer TA

    (2003) Structures of the alpha L I domain and its complex with ICAM-1 reveal a shape-shifting pathway for integrin regulation

    Cell 112:99–111.

    https://doi.org/10.1016/s0092-8674(02)01257-6

    • PubMed
    • Google Scholar
55. 1. Springer TA

    (1994) Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm

    Cell 76:301–314.

    https://doi.org/10.1016/0092-8674(94)90337-9

    • PubMed
    • Google Scholar
56. 1. Sprung R
    2. Chen Y
    3. Zhang K
    4. Cheng D
    5. Zhang T
    6. Peng J
    7. Zhao Y

    (2008) Identification and validation of eukaryotic aspartate and glutamate methylation in proteins

    Journal of Proteome Research 7:1001–1006.

    https://doi.org/10.1021/pr0705338

    • PubMed
    • Google Scholar
57. 1. Stanley P
    2. McDowall A
    3. Bates PA
    4. Brashaw J
    5. Hogg N

    (2000) The second domain of intercellular adhesion molecule-1 (ICAM-1) maintains the structural integrity of the leucocyte function-associated antigen-1 (LFA-1) ligand-binding site in the first domain

    The Biochemical Journal 351:79–86.

    https://doi.org/10.1042/0264-6021:3510079

    • PubMed
    • Google Scholar
58. 1. Staunton DE
    2. Marlin SD
    3. Stratowa C
    4. Dustin ML
    5. Springer TA

    (1988) Primary structure of ICAM-1 demonstrates interaction between members of the immunoglobulin and integrin supergene families

    Cell 52:925–933.

    https://doi.org/10.1016/0092-8674(88)90434-5

    • PubMed
    • Google Scholar
59. 1. Szmitko PE
    2. Wang C-H
    3. Weisel RD
    4. de Almeida JR
    5. Anderson TJ
    6. Verma S

    (2003) New markers of inflammation and endothelial cell activation: Part I

    Circulation 108:1917–1923.

    https://doi.org/10.1161/01.CIR.0000089190.95415.9F

    • PubMed
    • Google Scholar
60. 1. Wang C
    2. Deng L
    3. Hong M
    4. Akkaraju GR
    5. Inoue J
    6. Chen ZJ

    (2001) Tak1 is a ubiquitin-dependent kinase of MKK and IKK

    Nature 412:346–351.

    https://doi.org/10.1038/35085597

    • PubMed
    • Google Scholar
61. 1. Wang Y
    2. Tang Y
    3. Teng L
    4. Wu Y
    5. Zhao X
    6. Pei G

    (2006) Association of beta-arrestin and TRAF6 negatively regulates Toll-like receptor-interleukin 1 receptor signaling

    Nature Immunology 7:139–147.

    https://doi.org/10.1038/ni1294

    • PubMed
    • Google Scholar
62. 1. Yan G
    2. Qin Q
    3. Yi B
    4. Chuprun K
    5. Sun H
    6. Huang S
    7. Sun J

    (2013) Protein-L-isoaspartate (D-aspartate) O-methyltransferase protects cardiomyocytes against hypoxia induced apoptosis through inhibiting proapoptotic kinase MST1

    International Journal of Cardiology 168:3291–3299.

    https://doi.org/10.1016/j.ijcard.2013.04.045

    • PubMed
    • Google Scholar
63. 1. Yang Y
    2. Jun CD
    3. Liu JH
    4. Zhang R
    5. Joachimiak A
    6. Springer TA
    7. Wang JH

    (2004) Structural basis for dimerization of ICAM-1 on the cell surface

    Molecular Cell 14:269–276.

    https://doi.org/10.1016/s1097-2765(04)00204-7

    • PubMed
    • Google Scholar
64. 1. Ye H
    2. Arron JR
    3. Lamothe B
    4. Cirilli M
    5. Kobayashi T
    6. Shevde NK
    7. Segal D
    8. Dzivenu OK
    9. Vologodskaia M
    10. Yim M
    11. Du K
    12. Singh S
    13. Pike JW
    14. Darnay BG
    15. Choi Y
    16. Wu H

    (2002) Distinct molecular mechanism for initiating TRAF6 signalling

    Nature 418:443–447.

    https://doi.org/10.1038/nature00888

    • PubMed
    • Google Scholar
65. 1. Yin Q
    2. Lin S-C
    3. Lamothe B
    4. Lu M
    5. Lo Y-C
    6. Hura G
    7. Zheng L
    8. Rich RL
    9. Campos AD
    10. Myszka DG
    11. Lenardo MJ
    12. Darnay BG
    13. Wu H

    (2009) E2 interaction and dimerization in the crystal structure of TRAF6

    Nature Structural & Molecular Biology 16:658–666.

    https://doi.org/10.1038/nsmb.1605

    • PubMed
    • Google Scholar
66. 1. Zhao J
    2. Tian M
    3. Zhang S
    4. Delfarah A
    5. Gao R
    6. Rao Y
    7. Savas AC
    8. Lu A
    9. Bubb L
    10. Lei X
    11. Moshirian R
    12. Zhu W
    13. Peng C
    14. Jiang T
    15. Chen L
    16. Graham NA
    17. Feng P

    (2020) Deamidation shunts RelA from mediating inflammation to aerobic glycolysis

    Cell Metabolism 31:937–955.

    https://doi.org/10.1016/j.cmet.2020.04.006

    • PubMed
    • Google Scholar
67. 1. Zhu N
    2. Zhang GX
    3. Yi B
    4. Guo ZF
    5. Jang S
    6. Yin Y
    7. Yang F
    8. Summer R
    9. Sun J

    (2019) Nur77 limits endothelial barrier disruption to LPS in the mouse lung

    American Journal of Physiology. Lung Cellular and Molecular Physiology 317:L615–L624.

    https://doi.org/10.1152/ajplung.00425.2018

    • PubMed
    • Google Scholar

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting the paper "PIMT: a novel and potent suppressor of endothelial activation" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by a Senior Editor. The reviewers have opted to remain anonymous.

Comments to the Authors:

We are sorry to say that, after consultation with the reviewers, we have decided that this work will not be considered further for publication by eLife in its current form.

The consensus from the review was that although the identification of the role of PIMT in regulating vascular homeostasis is impactful, there were several major issues to be addressed to support the publication. Particularly, the reviewers felt that the premise for the role of isoAsp in pro-inflammatory function needs to be better established, the mass-spec data need to be presented more clearly, and the in vivo relevance of in vitro (over expression) results should be more clearly established (e.g., the effect on ICAM-I).

We, however, remain enthusiastic about the conceptual advancement of the work. Thus, we will be happy to consider an improved manuscript that has addressed these concerns.

Reviewer #1 (Recommendations for the authors):

The authors found that impairing PIMT (KD) exacerbated inflammation and vascular leakage upon LPS challenge. They report here that this is due to the lack of repair on damaged Asn on TRAF6, which resulted in inhibition of its oligomerization and ubiquitination (consequently preventing NF-κB transactivation). The authors also found that PIMT was linked to regulating ICAM-1 expression by inhibiting N-glycosylation. This study reveals a previously unappreciated function of PIMT in regulating innate immune responses- limiting inflammatory responses thus maintaining vascular homeostasis.

I have the following questions/ concerns:

– LC/MS data is very difficult to interpret/evaluate. I have questions as to: Is there any other Asn/Asp on TRAF-6 that gets modified by PIMT at all? What is the peptide coverage? What is the confidence level on distinguishing signal from noise. What is methylated Asn?- Are authors referring to L-isoaspartyl methyl ester? I am unable to identify corresponding modification by the given data.

– I'm having a hard time understanding why TRAF-6 would be stuck at L-isoaspartyl methyl ester. It is my understanding that, after PIMT reaction, succinimide formation and subsequent hydrolysis is rather spontaneous. Also, unlike TRAF-6, why are all Asns in ICAM-I modified into Asp, instead of being just "methylated?"

– Is enzymatic activity necessary for PIMT to regulate ICAM-I glycosylation? Could authors use point mutants at the active site instead of deletion (please ID how much has been deleted) to ensure proper folding and to distinguish whether PIMT binding or catalysis dominates the inhibitory effect?

– Would it be possible that ICAM-I hypo-glycosylation/attenuation simply be the downstream effect since only a small amount of PIMT is co-IP'd even (and only) when overexpressed?

– I'm a bit confused about the sequence of events and targeting mechanism of PIMT. Given that PIMT targets isoAsp, would LPS treatment (activation of TLRs) cause TRAF6 and ICAM-I to become more prone to such spontaneous modifications (how)? Can authors compare the population of modified/unmodified TRAF6 and ICAM-I under normal vs inflammatory conditions? Would the data/model here imply that isoAsp modification somehow over-activates the function of TRAF6? Wouldn't it impair the activity of TRAF6 since it would generate deformed backbone? Also, without PIMT, wouldn't the glycosylation of ICAM-I be impaired anyway since the target Asn residues would be stuck at isoAsp?

Reviewer #2 (Recommendations for the authors):

The authors show a role for PIMT in regulating endothelial inflammatory responses, leading to reduced NFkB activation downstream of LPS and TLR4. They demonstrate that PIMT directly interacts with TRAF6 and ICAM, modifying these targets and regulating their activity. in vivo experiments suggest a role of PIMT in regulating endothelial dysfunction and inflammation downstream of lung injury. Strengths include a thorough analysis of the interactions of PIMT with TRAF6 and ICAM and a demonstration of the modification of these targets by PIMT. Weaknesses include heavy reliance on overexpression of PIMT to observe biological effects and a weak effect of PIMT heterozygosity in vivo. This study should spur further investigation into the role of PIMT not just in repair of damaged proteins, but as a possible signalling regulator of inflammatory responses.

Recommendations

– Since the paper is heavily reliant on overexpression of PIMT, strengthening the data examining endogenous PIMT would greatly strengthen the paper. A second shRNA should be used in Figures 2C, 2F, and 5D. Showing a role for endogenous PIMT in regulating LPS-induced genes as in Figure 2D would also be helpful.

– Since some of the endogenous IPs are fairly dirty and it's not always clear which band is the correct band, it would be helpful to add a knockdown of the co-IPed protein to these IP experiments.

– Knockdown ICAM to show that the lower "hypo-glycosylated" band is specific for ICAM.

– Can the authors show an effect of endogenous PIMT on the ubiquitination of endogenous TRAF6? In Figure 4, the authors IP Flag-TRAF6 and blot for HA-ubiquitin. It would be predicted that ubiquitinated TRAF6 would be detected similarly with HA and Flag antibodies; however, there are no higher molecular weight bands in the Flag blot. Does this suggest that the HA blot is not specific to TRAF6?

– The LCMS data is not easy to understand, particularly in Figure 5. Could the authors give a better explanation of what the labeled peaks mean? Can the authors also show data from the methylated peptide in Figure 4D does not occur in shPIMT cells? Is there evidence of Asn to Asp shift at N350 due to PIMT activity? Can the authors show that Asn to Asp shift in Figure 5 does not occur in Ad-LacZ cells? Is endogenous PIMT enough for this to occur?

– Since the authors don't show an actual analysis of the glycosylation of ICAM, they should refrain from using "hypo-glycosylated" and "full-glycosylated". Their data strongly suggests that PIMT is changing the glycosylated residues of ICAM, but these terms are overstatements unless they directly measure glycosylation of ICAM.

– In Figure 5E, S3B, authors say that PIMT interacts only with hypoglycosylated ICAM. Since they IP only ICAM and their IP antibody does not distinguish from 100kd and 70kd bands, they cannot say this. They need to IP PIMT and show that only one ICAM band comes down in this IP. They should also use a dose of Ad-PIMT at which there is approximately equal higher and lower molecular weight ICAM bands.

– It is a little concerning that no endogenous PIMT comes down with ICAM pulldown in Figure S3B. Can the authors show that PIMT knockdown leads to increased surface localization of ICAM with or without TNF, and that PIMT knockdown leads to increased adhesion of THP1 cells?

Reviewer #3 (Recommendations for the authors):

The authors studied the regulatory role that PIMT plays in endothelial cells' response to innate stimulation. They found that PIMT modulates LPS-induced NF-κB activation by interaction with TRAF6 and methylation of Asn residues in its coiled-coil domain. Therefore, PIMT inhibits TRAF6 oligomerization and autoubiquitination. In addition, the authors found that PIMT inhibits N-glycosylation of ICAM-1 and prevents ICAM-1 expression on endothelial cell surface. In a LPS-induced acute lung injury model, the authors showed that PIMT ameliorates lung inflammation and injury. The study suggests that PIMT limits endothelial cell activation and prevents excessive tissue inflammation.

Using overexpression and knockdown methods, the authors have done elegant work in vitro to show how PIMT regulates endothelial activation. Whether it is the case in vivo needs to be tested.

There are some concerns that need to be addressed or clarified.

1. The authors focused on endothelial PIMT. It seems PIMT is expressed in many kinds of cells, such as macrophages (http://biogps.org/#goto=genereport&id=18537). The authors showed that PIMT is expressed in CD31+ endothelial cells in the lung using IHF. Do other cells in the lung also express PIMT? The authors may use FACS to measure PIMT expression in lung endothelial cells, epithelial cells and immune cells. Alternatively, an endothelial specific KO mice may be used.

2. The authors found that PIMT regulates endothelial activation and ICAM-1 expression in vitro. in vivo evidence is also needed in the acute lung injury model.

3. TRAF6 is an E3 ligase. Does PIMT regulate the polyubiquitination and phosphorylation of its substrate TAK1?

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "PIMT: a novel and potent suppressor of endothelial activation" for further consideration by eLife. Your revised article has been evaluated by Carla Rothlin (Senior Editor) and a Reviewing Editor.

The manuscript has been improved. However, we find the LC/MS data still a bit incomplete and noisy. Can authors show the degree of modification of ICAM by endogenous PIMT? Similarly, could authors show the ratio between unmodified and modified peptides by (ectopically expressed) PIMT? These data will be crucial for establishing the physiological relevance of ICAM modification by PIMT.

Reviewer #1 (Recommendations for the authors):

The authors have largely addressed my concerns; however several issues remain:

I remain concerned about the lack of evidence for endogenous PIMT in the regulation of ICAM1. The Western blot that the authors include for shPIMT cells shows no effect on the lower molecular weight band (cropped from Figure 5C, but included in the full image). If the authors treated those cells with TNFa, IL1B, or LPS, would the increased amount of total ICAM be sufficient to see differences in low molecular weight ICAM1? The authors show that low molecular weight ICAM1 is present on the membrane of AD-LacZ cells. Is this small amount of low molecular weight ICAM reduced by knockdown of PIMT? It's worth mentioning that the ICAM1 bands in Figures 5C and 6C don't line up with the ladder in the same way as they do in other figures. Was a different ladder used for these blots?

The LC/MS data remains difficult to evaluate. Presumably, the TRAF6 analyzed is a mixture of modified and non-modified peptides. I would assume that not every single TRAF6 in the cell contains methylated asparagines and that the ~50% knockdown of PIMT would not completely abolish all of these methylations. Is there any way to represent the data that shows the relative amount of PIMT-modified to unmodified TRAF6 present in cells? This would presumably be more difficult in ICAM given the issues with glycosylation.

The authors' change of hypo- and full- glycosylated to partially and highly glycosylated misses the point. They have not analyzed glycosylation of ICAM at all. It is possible (even highly likely) that the two different bands represent different glycosylation states of ICAM, but without experimentally demonstrating this, the authors should stick to high molecular weight and low molecular weight ICAM throughout. It is fine to suggest that the effects of PIMT are likely due to differential glycosylation, but statements like "we uncovered an unexpected role of PIMT in modulating N-glycosylation of adhesion molecules", "PIMT specially inhibits the N-glycosylation of ICAM1 and VCAM1", and "PIMT blocked ICAM1 expression by inhibiting N-glycosylation" go beyond what the data show.

Reviewer #2 (Recommendations for the authors):

The authors have addressed my concerns.

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Essential revisions:.

Reviewer #1 (Recommendations for the authors):

The authors found that impairing PIMT (KD) exacerbated inflammation and vascular leakage upon LPS challenge. They report here that this is due to the lack of repair on damaged Asn on TRAF6, which resulted in inhibition of its oligomerization and ubiquitination (consequently preventing NF-κB transactivation). The authors also found that PIMT was linked to regulating ICAM-1 expression by inhibiting N-glycosylation. This study reveals a previously unappreciated function of PIMT in regulating innate immune responses- limiting inflammatory responses thus maintaining vascular homeostasis.

Thank you for your comments and helping us to improve the manuscript.

I have the following questions/ concerns:

– LC/MS data is very difficult to interpret/evaluate. I have questions as to: Is there any other Asn/Asp on TRAF-6 that gets modified by PIMT at all?

Great question. We focused on the CC domain because our domain mapping studies demonstrated this domain to be responsible for the binding of PIMT on TRAF6. This led us to hypothesize that PIMT may modify Asn/Asp residues in this domain. Ideally, this could be confirmed by overexpressing the CC domain but the small size (60 aa) of this region makes such a strategy almost impossible. As such, we choose to overexpresss a larger TRAF6 fragment containing the entire c-terminal domain (aa: 289-522). Doing this, we uncovered by LC/MS that both the D339 and N350 residues were modified in PIMT expressing HEK293 cells, but not in PIMT knockdown cells.

Functional studies also showed that the N350 modification is critically important in the regulation of TRAF6 by PIMT. Although we suspect that PIMT may also modify Asn/Asp residues outside of the cterminal domain we feel these experiments are beyond the scope of this manuscript.

What is the peptide coverage? What is the confidence level on distinguishing signal from noise.

We have a high degree of confidence in our peptide identification. In mass spec studies, we overexpressed TRAF6 fragment in HEK293 cells and extracted immunoprecipitated TRAF6 from gels for LC/MS analysis. We then followed standard guidelines established by the proteomic research societies (https://www.mcponline.org/mass-spec-guidelines) for next steps. Our false discovery rate threshold for peptide identification was less than 1%. Moreover, since we used high-resolution mass spectrometry, the peptide mass tolerance was 10 ppm for identification. More importantly, the identification of peptides was further validated by manual inspection of MS/MS data to ensure accurate peptide and PTM site assignments. Using this approach, we identified clear fragment ions corresponding to methylated Asn in our MS/MS spectra. Altogether, we believe our approach utilized gold standard methods for performing PTM localization and have a high degree of confidence in our ability to distinguish signal from noise.

What is methylated Asn?- Are authors referring to L-isoaspartyl methyl ester? I am unable to identify corresponding modification by the given data.

Sorry for the confusion. Methylated Asn is actually L-isoaspartyl methyl ester. In the revised figure 4D, the formation of L-isoaspartyl methyl ester from Asn is now clearly depicted. Peptide mass was matched with 1.5 ppm accuracy to the theoretical mass of L-isoaspartyl methyl ester (881.4211, z=2) instead of 874.4120 for the unmodified form. Also, diagnostic fragmentation ions such as 147.08681 (N-me), 307.11740 (CN-me), 435.17598 (QCN-me) corresponding to the Asn-derived Lisoaspartyl methyl ester were observed.

– I'm having a hard time understanding why TRAF-6 would be stuck at L-isoaspartyl methyl ester. It is my understanding that, after PIMT reaction, succinimide formation and subsequent hydrolysis is rather spontaneous.

We agree that isoAsp repair is traditionally recognized as a spontaneous process. However, recent studies show that O-methylation of IsoAsp occurs in nature, including on proteins such as p53 (Nat Commun. 2012;3:927) and histone H4 (Sci Rep. 2014;4:6674), and exerts important biological functions. Notably, a recent publication demonstrated that O-methylation of p53 is not just a transient event and can be maintained for at least 24 hours (Nat Commun. 2012;3:927). We have added these comments into the Discussion section of the revised manuscript (page 26, from bottom, line 5-9).

Also, unlike TRAF-6, why are all Asns in ICAM-I modified into Asp, instead of being just "methylated?"

Very thoughtful but challenging question. Our data show that among 8 Asns that were identified to be glycosylated in ICAM-1, only 4 Asns were modified to Asp in the presence of PIMT. Although we do not have a great answer regarding the specificity, it is likely that isoAsp repair and maintenance, just like other protein modifications, occur in a context and cell type dependent fashion. We have now modified the Discussion to address this criticism (page 28, 1^st paragraph, line 8-11).

– Is enzymatic activity necessary for PIMT to regulate ICAM-I glycosylation? Could authors use point mutants at the active site instead of deletion (please ID how much has been deleted) to ensure proper folding and to distinguish whether PIMT binding or catalysis dominates the inhibitory effect?

Thank you for your suggestion. Yes enzymatic activity of PIMT is necessary to regulate ICAM-I glycosylation since we found that modification of 4 Asns in ICAM-I needs the interaction of PIMT with ICAM1 and a subsequent PIMT catalyzed o-methylation of IsoAsp. In our preliminary studies, we generated PIMT catalytic null mutations including G88A (Nat Commun. 2012;3:927) and D83V (Front Genet. 2021 Jan 21;11:612343) and found that they still maintain some enzymatic activity as compared with methyltransferase motif I (81-90) deleted mutant (Arch Biochem Biophys. 1994;310:417). Thus, we used an enzymatic inactive PIMT deletion mutant (deletion of aa 81-89) in our study. It seems that this small deletion (9 aa) will not significantly affect the proper protein folding, based on a previous report (Arch Biochem Biophys. 1994;310:417). Some of the comments have been added into the Discussion section of the revised manuscript (page 27, from bottom, line 5-6 ).

– Would it be possible that ICAM-I hypo-glycosylation/attenuation simply be the downstream effect since only a small amount of PIMT is co-IP'd even (and only) when overexpressed?

Our data show that PIMT associates with the low MW ICAM-1 band. However, in our IP experiment, only small fractions of PIMT co-IP’d with LMW ICAM-1. Although we suspect this could be explained by a “hit and run” mechanism, we cannot exclude the possibility that PIMT affects any one of the many steps along the pathway to protein glycosylation, such as ER quality control or glucoside attachment. In future investigations we hope to further answer these questions and have added new text to the Discussion to address this concern in the revised manuscript (page 28, 1st paragraph, line 10-14).

– I'm a bit confused about the sequence of events and targeting mechanism of PIMT. Given that PIMT targets isoAsp, would LPS treatment (activation of TLRs) cause TRAF6 and ICAM-I to become more prone to such spontaneous modifications (how)?

Yes, we believe that LPS should cause TRAF6 and ICAM-I to undergo spontaneous modifications. This is because LPS is known to trigger production of reactive oxygen species (ROS) (J Biol Chem. 2002;277(25):22131-9.), and ROS is known to induce isoAsp generation (Free Radic Biol Med. 2001 Jul 1;31(1):1-9.). This comment has been added to the Discussion section of the revised manuscript (page 26, 2^nd paragraph, line 8-9).

Can authors compare the population of modified/unmodified TRAF6 and ICAM-I under normal vs inflammatory conditions?

Unfortunately, measuring O-methylation of TRAF6 by LC/MS/MS is not possible at this point. Due to the fact that antibodies against O-methylation of TRAF6 do not exist, this also limits our ability to evaluate this modification using western blot or ELISA. Finally, O-methylation of ICAM-1 seems spontaneous and rapid, which does not permit its quantification by LC/MS.

Would the data/model here imply that isoAsp modification somehow over-activates the function of TRAF6? Wouldn't it impair the activity of TRAF6 since it would generate deformed backbone? Also, without PIMT, wouldn't the glycosylation of ICAM-I be impaired anyway since the target Asn residues would be stuck at isoAsp?

In addition to altering the backbone, IsoAsp modifications will likely alter conformation and post-translational modifications, which could have positive (e.g. p53) or negative (e.g. tubulin) effects on protein function. In our study, we found that PIMT knockdown enhances TRAF6/NF-κB activation (Figure 2F), suggesting that IsoAsp modifications act to enhance TRAF6 function.

Under inflammatory conditions Asn resides in ICAM are stable and resistant to deamination (Figure 5G). Thus, without PIMT we expect elevated expression and full glycosylation of ICAM. Consistent with this, our studies indicate that targeted activation of PIMT in ECs reduces ICAM-1 expression and glycosylation and mitigates endothelial activation in ALI. These proof of concept studies support using PIMT activators in the treatment of inflammatory diseases. Comments related to this have been added into the Discussion section of the revised manuscript (page 26, 2nd paragraph, line 9-13).

Reviewer #2 (Recommendations for the authors):

The authors show a role for PIMT in regulating endothelial inflammatory responses, leading to reduced NFkB activation downstream of LPS and TLR4. They demonstrate that PIMT directly interacts with TRAF6 and ICAM, modifying these targets and regulating their activity. in vivo experiments suggest a role of PIMT in regulating endothelial dysfunction and inflammation downstream of lung injury. Strengths include a thorough analysis of the interactions of PIMT with TRAF6 and ICAM and a demonstration of the modification of these targets by PIMT. Weaknesses include heavy reliance on overexpression of PIMT to observe biological effects and a weak effect of PIMT heterozygosity in vivo. This study should spur further investigation into the role of PIMT not just in repair of damaged proteins, but as a possible signalling regulator of inflammatory responses.

Recommendations

– Since the paper is heavily reliant on overexpression of PIMT, strengthening the data examining endogenous PIMT would greatly strengthen the paper. A second shRNA should be used in Figures 2C, 2F, and 5D. Showing a role for endogenous PIMT in regulating LPS-induced genes as in Figure 2D would also be helpful.

Thank you for your suggestion. We have performed additional experiments using a second shRNA to deplete PIMT in ECs. Our new results are added into the revised figure 2C, 2F, and 5C. qPCR detection of inflammatory molecules in response to LPS in control and PIMT depleted ECs was also performed and is now shown in the revised figure 2D.

– Since some of the endogenous IPs are fairly dirty and it's not always clear which band is the correct band, it would be helpful to add a knockdown of the co-IPed protein to these IP experiments.

Thank you for your suggestion. As shown in Figure 3E, the expression of TRAF6 in ECs is relatively low, which may lead to weak interactions in IP. To fully verify that TRAF6 indeed interacts with PIMT, we performed knockdown studies using TRAF6 shRNA. As shown in Figure 3E, TRAF6 with a MW of 60 kDa in control shRNA transfected cells, but not in TRAF6 shRNA transfected cells, interacts with PIMT in ECs. This interaction is further enhanced by LPS stimulation.

– Knockdown ICAM to show that the lower "hypo-glycosylated" band is specific for ICAM.

Thank you for your comments. Although we did not perform ICAM-1 knockdown studies, we cut the lower "partially-glycosylated" band from the gel (Figure 5F) and submitted it for mass spectrometric analysis. Our results clearly identified peptide fragments of ICAM-1 (data not shown) in the band, supporting the notion that our lower band is derived from ICAM-1.

– Can the authors show an effect of endogenous PIMT on the ubiquitination of endogenous TRAF6? In Figure 4, the authors IP Flag-TRAF6 and blot for HA-ubiquitin. It would be predicted that ubiquitinated TRAF6 would be detected similarly with HA and Flag antibodies; however, there are no higher molecular weight bands in the Flag blot. Does this suggest that the HA blot is not specific to TRAF6?

As shown in many other studies, detection of endogenous ubiquitination of TRAF6 in ECs is very challenging. We speculate this relates, in large part, to the lack of high quality TRAF6 antibodies as well as the relatively low expression of TRAF6 in ECs. As shown in Figure 4A, by using protein ectopic expression, the poly-ubiquitination chain could be specifically and effectively recognized by HA antibody. Since ubiquitin is HA tagged, western blot using HA antibody only detects ubiquitinated TRAF6. After IP, western blot using Flag antibody detects the total TRAF6, which includes both ubiquitinated and non-ubiquitinated TRAF6. Since the level of ubiquitinated TRAF6 is much lower compared with non-ubiquitinated TRAF6 in the immunoprecipitates we believe this explains why we normally observe a strong total TRAF6 band in western blot using anti-Flag antibody. Through increasing exposure time, we may observe higher molecular weight bands, but the quality of image would be much lower.

– The LCMS data is not easy to understand, particularly in Figure 5. Could the authors give a better explanation of what the labeled peaks mean?

We apologize for not presenting our data clearly. We now provide a better explanation of what the labelled peaks mean in the revised Figure 5G and Figure 5-supplement 2. In Figure 5G and Figure 5-supplement 2, the mass spectrometry spectra demonstrate that DHHGANFSCR is deamidated (DHHGADFSCR). Peptide mass was matched with 1.5 ppm accuracy to the theoretical mass of the Asn-methylated form (m/z 401.1649, double charge) instead of m/z 400.8370 (double charge) for the unmodified form. In MS/MS, diagnostic fragmentation ions such as

633.3015(DHHGAD), 755.3333(DFSCR),794.31586 (ADFSCR), 812.3377 (GADFSCR) corresponding to the deamidation were observed.

Can the authors also show data from the methylated peptide in Figure 4D does not occur in shPIMT cells? Is there evidence of Asn to Asp shift at N350 due to PIMT activity? Can the authors show that Asn to Asp shift in Figure 5 does not occur in Ad-LacZ cells? Is endogenous PIMT enough for this to occur?

Unmethylated TRAF6 peptide from shPIMT cells was detected and shown in the revised Figure 4—figure supplement 1F. Asn to Asp shift, known as the repaired isoAsp, requires PIMT catalyzed O-methylation. As shown in the revised Figure 4D, O-methylation was only detected in PIMT expressing cells, but not in PIMT knockdown cells, strongly indicating that PIMT is required to methylate isoAsp derived from Asn or Asp.

In Ad-lacZ cells, ICAM-1 is fully glycosylated and after trypsin digestion, identification of N-linked glycopeptides (glycans plus peptides) using LC/MS/MS is very challenging. In our study, we only detected a peptide spectrum of ICAM-1 with an undeamidated N-glycosite in Ad-LacZ cells as shown in Author response image 1, indicating this site is neither glycosylated nor deamidated.

Author response image 1

Download asset Open asset

After knocking down endogenous PIMT, we observed increased expression of ICAM-1 at MW 100 kDa (Figure 5C), mainly due to a rapid induction of ICAM-1 expression by NF-ΚB activation.Proinflammatory stimuli cause a rapid upregulation in adhesion molecules and ER stress markers. In this case, some neo-synthesized ICAM1 proteins without proper folding might escape from ER quality check. Thus, we sometimes observe weak lower MW ICAM1 bands in western blot, indicating endogenous PIMT might be functional here. Nevertheless, our data show that overexpression of PIMT markedly suppresses ICAM-1 glycosylation and provides proof of concept that targeted activation of PIMT in ECs may be beneficial for inhibition of ICAM-1 function under inflammatory conditions.

– Since the authors don't show an actual analysis of the glycosylation of ICAM, they should refrain from using "hypo-glycosylated" and "full-glycosylated". Their data strongly suggests that PIMT is changing the glycosylated residues of ICAM, but these terms are overstatements unless they directly measure glycosylation of ICAM.

Thank you for your comments. We now use “partially-glycosylated” and “highly-glycosylated” instead of "hypo-glycosylated" and "full-glycosylated", respectively, in the revised manuscript.

– In Figure 5E, S3B, authors say that PIMT interacts only with hypoglycosylated ICAM. Since they IP only ICAM and their IP antibody does not distinguish from 100kd and 70kd bands, they cannot say this. They need to IP PIMT and show that only one ICAM band comes down in this IP. They should also use a dose of Ad-PIMT at which there is approximately equal higher and lower molecular weight ICAM bands.

Thank you for the suggestion. We have now performed IP using anti-PIMT antibody and results are shown in the revised Figure 5E. Although results show an equal input of high and low molecular weight ICAM bands only partially glycosylated ICAM-1 interacted with PIMT.

– It is a little concerning that no endogenous PIMT comes down with ICAM pulldown in Figure S3B.

We understand your concern; however, our data indicate that PIMT mostly associates with partially-glycosylated ICAM-1. In Ad-LacZ cells, since ICAM-1 is known to be highly glycosylated, we are unable to see endogenous PIMT1 in ICAM-1 immunocomplexes (highly glycosylated). Hopefully this explanation alleviates some of this reviewer’s concerns.

Can the authors show that PIMT knockdown leads to increased surface localization of ICAM with or without TNF, and that PIMT knockdown leads to increased adhesion of THP1 cells?

As suggested, we have performed PIMT knockdown and examined the surface localization of ICAM by flow cytometry. As shown in the revised figure 5—figure supplement 1C, we found that PIMT knockdown significantly increased cell surface ICAM-1 expression. Consistent with this, we found that PIMT knockdown also increased the adhesion of THP1 cells to activated ECs. These data are now shown in the revised figure 5—figure supplement 1D.

Reviewer #3 (Recommendations for the authors):

The authors studied the regulatory role that PIMT plays in endothelial cells' response to innate stimulation. They found that PIMT modulates LPS-induced NF-κB activation by interaction with TRAF6 and methylation of Asn residues in its coiled-coil domain. Therefore, PIMT inhibits TRAF6 oligomerization and autoubiquitination. In addition, the authors found that PIMT inhibits N-glycosylation of ICAM-1 and prevents ICAM-1 expression on endothelial cell surface. In a LPS-induced acute lung injury model, the authors showed that PIMT ameliorates lung inflammation and injury. The study suggests that PIMT limits endothelial cell activation and prevents excessive tissue inflammation.

Using overexpression and knockdown methods, the authors have done elegant work in vitro to show how PIMT regulates endothelial activation. Whether it is the case in vivo needs to be tested.

Thank you for your comments and helping us to improve the manuscript.

There are some concerns that need to be addressed or clarified.

1. The authors focused on endothelial PIMT. It seems PIMT is expressed in many kinds of cells, such as macrophages (http://biogps.org/#goto=genereport&id=18537). The authors showed that PIMT is expressed in CD31+ endothelial cells in the lung using IHF. Do other cells in the lung also express PIMT? The authors may use FACS to measure PIMT expression in lung endothelial cells, epithelial cells and immune cells. Alternatively, an endothelial specific KO mice may be used.

Thank you for your suggestion. Yes, PIMT is also expressed in macrophages and epithelial cells in the lung. In this study, we did not perform FACS to determine the relative expression of PIMT in different types of cells in the lung. Data documented in the public domain of The Human Protein Atlas (https://www.proteinatlas.org/ENSG00000120265-PCMT1/tissue/lung) provides a single cell transcriptome analysis in human lung tissues demonstrating that PIMT is highly expressed in lung endothelial cells, as shown in Author response image 2. Although we did not use endothelial specific KO mice, our in vitro experiments using gain- and loss-of function strategies clearly demonstrated that PIMT is important for lung endothelial activation. Determination of relative contribution of PIMT in different cell types to ALI requires generation of various tissue specific knockout mice, which is extremely time-consuming, will need to be the subject of future studies. The comments have been added into the Discussion section of the revised manuscript (page 27, 1^st paragraph, line 3-6).

Author response image 2

Download asset Open asset

2. The authors found that PIMT regulates endothelial activation and ICAM-1 expression in vitro. in vivo evidence is also needed in the acute lung injury model.

Thank you for your suggestion. As shown in Figure 1B and E, we now demonstrate that neutrophil infiltration and ICAM-1 expression (both are hallmarks of endothelial activation) are increased in the lungs of PIMT knockout mice after LPS when compared to wild-type littermates.

3. TRAF6 is an E3 ligase. Does PIMT regulate the polyubiquitination and phosphorylation of its substrate TAK1?

Excellent point. We have now performed an ubiquitination assay and found that TAK1 polyubiquitination was effectively inhibited by active PIMT, but not by its catalytic null mutant.

Furthermore, we found that TAK1 phosphorylation (S412) was negatively regulated by PIMT in ECs. These results have now been incorporated into the revised figure 4—figure supplement 1A and 1B.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Reviewer #1 (Recommendations for the authors):

The authors have largely addressed my concerns; however several issues remain:

I remain concerned about the lack of evidence for endogenous PIMT in the regulation of ICAM1. The Western blot that the authors include for shPIMT cells shows no effect on the lower molecular weight band (cropped from Figure 5C, but included in the full image). If the authors treated those cells with TNFa, IL1B, or LPS, would the increased amount of total ICAM be sufficient to see differences in low molecular weight ICAM1? The authors show that low molecular weight ICAM1 is present on the membrane of AD-LacZ cells. Is this small amount of low molecular weight ICAM reduced by knockdown of PIMT? It's worth mentioning that the ICAM1 bands in Figures 5C and 6C don't line up with the ladder in the same way as they do in other figures. Was a different ladder used for these blots?

Thank you for the comments. Our results indicate that PIMT regulates ICAM-1 expression by affecting both NF-ΚB activation (transcriptional mechanism) and modifying protein glycosylation (posttranslational mechanism). Our results indicate that transcriptional mechanisms are more significant in PIMT knockdown cells since this mainly causes an increase in NF-ΚB activation without affecting N-linked glycosylation of ICAM-1. Consistent with this, we did not observe a change in the 8 ICAM-1 glycosylated sites (Asn) in shPIMT cells and we would not expect to find changes in the lower molecular weight band in Figure 5C. We also anticipate that treatment of shPIMT cells with other NF-ΚB activators, such as TNFa, IL1B, or LPS, would not significantly alter the amount of low molecular weight ICAM on gels because ICAM-1 should be highly glycosylated in this setting. However, this needs to be confirmed in future studies.

As shown in Figure 6C, the size of low molecular weight (LMW) ICAM-1 is different on the membrane of AD-LacZ and Ad-PIMT cells. Although we do not have an explanation for these findings we speculate this relates to different types of N-glycosylation or to other types of protein modifications. Further, our data indicate that ICAM-1 is highly glycosylated in PIMT knockdown cells. Based on this, we speculate that smaller amounts of LMW ICAM-1 present on the membrane of AD-LacZ cells will be shifted to HMW ICAM-1 in PIMT knockdown cells.

Yes, we used 2 types of protein ladders in this study, which were described in the Key Resources Table.

The LC/MS data remains difficult to evaluate. Presumably, the TRAF6 analyzed is a mixture of modified and non-modified peptides. I would assume that not every single TRAF6 in the cell contains methylated asparagines and that the ~50% knockdown of PIMT would not completely abolish all of these methylations. Is there any way to represent the data that shows the relative amount of PIMT-modified to unmodified TRAF6 present in cells? This would presumably be more difficult in ICAM given the issues with glycosylation.

We appreciate the valuable advice provided and completely agree with your comments. In order to investigate the degree of modification, we revisited the mass spectrometry spectra. We found that two forms of protein co-eluted in our HPLC and that both forms had a similar intensity in our analysis. As a result, the comment that "the ~50% knockdown of PIMT would not completely abolish all of these methylations " is absolutely correct. The relative amount of modified to unmodified protein at N350 in TRAF6-CC domain is now presented in Figure 4—figure supplement 1F and the revised manuscript (page 21, 2^nd paragraph, line 9-10).

Indeed, it is more difficult to show the modification of ICAM-1 in PIMT knockdown cells as a significant amount of ICAM-1 will be highly glycosylated.

The authors' change of hypo- and full- glycosylated to partially and highly glycosylated misses the point. They have not analyzed glycosylation of ICAM at all. It is possible (even highly likely) that the two different bands represent different glycosylation states of ICAM, but without experimentally demonstrating this, the authors should stick to high molecular weight and low molecular weight ICAM throughout. It is fine to suggest that the effects of PIMT are likely due to differential glycosylation, but statements like "we uncovered an unexpected role of PIMT in modulating N-glycosylation of adhesion molecules", "PIMT specially inhibits the N-glycosylation of ICAM1 and VCAM1", and "PIMT blocked ICAM1 expression by inhibiting N-glycosylation" go beyond what the data show.

As suggested, we now use the terms high molecular weight (HMW) and low molecular weight (LMW) to refer to different forms of ICAM-1 and have modified other statements in the manuscript to more clearly represent our data.

Author details

1. Chen Zhang

   Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8198-3090

2. Zhi-Fu Guo

   Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

   Contribution

   Data curation, Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

3. Wennan Liu

   Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

   Contribution

   Data curation, Visualization, Methodology

   Competing interests

   No competing interests declared

4. Kyosuke Kazama

   Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

   Contribution

   Visualization, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6794-6982

5. Louis Hu

   Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

   Contribution

   Visualization, Methodology

   Competing interests

   No competing interests declared

6. Xiaobo Sun

   Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

   Contribution

   Validation, Methodology

   Competing interests

   No competing interests declared

7. Lu Wang

   Department of Cell and Developmental Biology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, United States

   Contribution

   Visualization, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8896-0115

8. Hyoungjoo Lee

   Quantitative Proteomics Resource Center, University of Pennsylvania, Philadelphia, United States

   Contribution

   Data curation, Methodology

   Competing interests

   No competing interests declared

9. Lin Lu

   Department of Cardiovascular Medicine, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

   Contribution

   Conceptualization, Visualization, Methodology, We add Dr. Lin Lu as a coauthor because of his contribution to the experimental design and data interpretation of mass spectrometric analysis of ICAM-1 glycosylation

   Competing interests

   No competing interests declared

10. Xiao-Feng Yang

    Center for Metabolic Disease Research, Lewis Katz School of Medicine, Temple University, Philadelphia, United States

    Contribution

    Conceptualization, Visualization, Writing – review and editing

    Competing interests

    No competing interests declared

11. Ross Summer

    Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

    Contribution

    Conceptualization, Visualization, Writing – review and editing

    Competing interests

    No competing interests declared

12. Jianxin Sun

    Center for Translational Medicine, Thomas Jefferson University, Philadelphia, United States

    Contribution

    Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing – review and editing

    For correspondence

    jianxin.sun@jefferson.edu

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0001-5319-912X

• 225

  Page views

• 39

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.