Purchase this article with an account.
Anna Takaoka, Julia Hogan, Natasha Babar, Mi Jung Kim, Marianne O Price, Francis W Price, Stephen L Trokel, David C Paik; An evaluation of enzymatic collagen cross-linking in keratoconus using LC/MS. Invest. Ophthalmol. Vis. Sci. 2014;55(13):4221.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Riboflavin photochemical corneal cross-linking is gaining widespread clinical acceptance. Recent genetic and immunohistochemical studies suggest that an enzymatic collagen cross-linking defect may underlie keratoconus (KC) pathogenesis. Current literature contains a paucity of information regarding enzymatic collagen cross-linking in the KC cornea. Thus, the present study was undertaken in order to examine levels of enzymatic cross-links in KC tissue by LC/MS, correlating levels with differences in fibril stability as determined by thermal denaturation temperature or melting temperature (Tm).
Surgical KC samples (n=9) and eyebank control (n=4) corneas of age<45y/o were analyzed. The samples were defatted (folch), reduced (NaBH4), hydrolyzed (6N HCl at 110°C 18hrs), and cellulose enriched prior to analysis by C8 HPLC (Agilent1100) equipped with FLD/MSD in SIM mode (20mM heptafluorobutyric acid/MeOH 70:30 isocratic at 1mL/min). Nine different cross-links were measured and included enzymatic collagen di- [DHLNL (m/z=308.2), HLNL (m/z=292.2), LNL (m/z=276.1)] and tri-functional [HHL (m/z=445.2), PYD (m/z=429.2), DPYD (m/z=413)] cross-links, elastin [desmosines (m/z=526.3)] cross-links, and a non-enzymatic age-related glycation [pentosidine (m/z=379.2)] cross-link. Cross-link density (mol/mol) was expressed relative to collagen content (determined colorimetrically). Tm was determined by differential scanning calorimetry (Perkin Elmer DSC-6000).
KC corneas underwent thermal denaturation at Tm=58.90±0.23°C vs. controls=61.4±0.28°C [p<0.001], indicating greater fibril instability. Collagen cross-link levels (mol/mol collagen) were as follows: DHLNL [KC=0.066±0.01 vs. controls=0.033±0.01, (p=0.092)]; HLNL [KC=0.043±0.01 vs. controls=0.041±0.004, (p=0.736)]; LNL [KC: 0.12 ± 0.02 vs. controls: 0.19 ± 0.03, (p=0.121)]; and HHL [KC=0.052±0.014 vs. controls=0.062±0.019 (p=0.693)]. PYD, DPYD, desmosines, and pentosidine were not detected. Collagen content (based on a 14% hydroxyproline calculation) was 48±2% in KC vs. 56±5% in controls (p=0.213).
Alterations in difunctional enzymatic collagen cross-links may contribute to mechanical instability in KC. These studies have relevance to both KC pathogenesis and therapy.
This PDF is available to Subscribers Only