LC-MS/MS identification of the O-glycosylation and hydroxylation of amino acid residues of collagen α-1 (II) chain from bovine cartilage

Abstract

O-Glycosylation of collagen is a unique type of posttranslational modifications (PTMs) involving the attachment of galactose (Gal) or glucose-galactose (Glc-Gal) moieties to hydroxylysine (HyK). Also, hydroxyproline (HyP) result from the posttranslational hydroxylation of some proline residues in collagen. Here, LC-MS/MS was effectively employed to identify 23 O-glycosylation sites and a large number of HyP residues associated with bovine type II collagen α-1 chain (CO2A1). The modifications of the 23 O-glycosylation sites varied qualitatively and quantitatively. Both Gal and Glc-Gal moieties occupied 22 of the identified glycosylation sites, while K773 was observed as unmodified. A large number of HyP residues at Yaa positions of Gly-Xaa-Yaa motif were detected. HyP residues at Xaa positions of Gly-HyP-HyP, Gly-HyP-Ala, and Gly-HyP-Val motifs were also observed. Notably, HyP residue of Gly-HyP-Gln motif was detected, which has not been previously reported. Moreover, the deamidation of 8 Asn residues was identified, of which 2 Asp residues were observed at different retention times because of isomerization (Asp vs isoAsp). Partial macroheterogeneities of some CO2A1 glycosylation sites were revealed by LC-MS/MS analysis. ETD experiments revealed partial macroheterogeneities associated with K299-K308, K452-K464, K464-K470, and K857-K884 glycosylation sites. Semiquantitative data suggest that the glycosylation of hydroxylysine residues is site-specific.

Figure 1

Mapping different modifications of CO2A1 protein with sequence coverage of ca. 90%. Signal and propeptides (1~201), and C-telopeptide (1253~1487) are italicized since the sample was already pepsinized by the vendor. Identified amino acids from MASCOT database search are underlined. Hydroxylated proline or lysine residues are bold while unmodified forms are not. If modified and unmodified were detected, additional letters P or K are added in the upper line of the sequence. Glc-Gal or Gal modified lysines are represented.

Figure 2

EICs of TGPpGPAGFAGPpGADGQPGAkGEQGE peptide with modifications of Glc-Gal(A), Gal(B), and HyK(B) and unmodified K(D). At m/z 818.037(+3), two isomers were observed. The peptide eluted at 28.64 min was determined to be the peptide sequence TGPPGPAGFAGPPGADGQPGAKGEQGE with hydroxylations of 3 prolines while the one observed at 30.95 min was hydroxylated on 2 prolines and K848.

Figure 3

CID tandem MS of TGPpGPAGFAGPpGADGQPGAkGEQGE peptide modified with Glc-Gal, Gal, and HyK, and without modification on K848. CID tandem MS of the peptide modified with Glc-Gal moiety (A) and modified with Gal moiety (B). CID tandem MS of m/z 818.037(+3) representing peptide modified with HyP but not HyK(C) and modified with HyP and HyK(D). CID tandem MS of the peptide with no modification of K848(E).

Figure 4

ETD spectra of glycopeptides possessing multiple glycosylation sites of K299 and K308 showing assignments of Gal moiety on K299 with Glc-Gal moiety on K308 (A) and assignments of Glc-Gal moiety on K299 with Gal moiety on K308 (B). Other ETD spectra of glycopeptides possessing multiple glycosylation sites of K452 and K464 with assignment of Glc-Gal moiety on K452 while hydroxylated K464(C) and assignment of Gal moieties on both K452 and K464(D).

Figure 5

Quantitative results of 23 glycosylation sites which have included all the modified peptide sequences of a glycosylation site and been normalized.