Cyclophilin B Deficiency Causes Abnormal Dentin Collagen Matrix

Abstract

Cyclophilin B (CypB) is an endoplasmic reticulum-resident protein that regulates collagen folding, and also contributes to prolyl 3-hydroxylation (P3H) and lysine (Lys) hydroxylation of collagen. In this study, we characterized dentin type I collagen in CypB null (KO) mice, a model of recessive osteogenesis imperfecta type IX, and compared to those of wild-type (WT) and heterozygous (Het) mice. Mass spectrometric analysis demonstrated that the extent of P3H in KO collagen was significantly diminished compared to WT/Het. Lys hydroxylation in KO was significantly diminished at the helical cross-linking sites, α1/α2(I) Lys-87 and α1(I) Lys-930, leading to a significant increase in the under-hydroxylated cross-links and a decrease in fully hydroxylated cross-links. The extent of glycosylation of hydroxylysine residues was, except α1(I) Lys-87, generally higher in KO than WT/Het. Some of these molecular phenotypes were distinct from other KO tissues reported previously, indicating the dentin-specific control mechanism through CypB. Histological analysis revealed that the width of predentin was greater and irregular, and collagen fibrils were sparse and significantly smaller in KO than WT/Het. These results indicate a critical role of CypB in dentin matrix formation, suggesting a possible association between recessive osteogenesis imperfecta and dentin defects that have not been clinically detected.

Cyclophilin B control of lysine post-translational modifications of skin type I collagen

“…We next analyzed Lys hydroxylation in the telopeptides of type I collagen. We previously analyzed telopeptidyl Lys hydroxylation at N-telopeptide (Lys-9 N ) and C-telopeptide (Lys-16 C ) of the α1(I) chain after sequential digestion by Grimontia collagenase and pepsin [23]. In the present study, we further identified Lys-5 N -containing peptide (pQYSDKGVSSGPGPM; pQ indicates pyroglutamic acid) from N-telopeptide of α2(I) chain (S2 Fig).…”

“…Type III collagen content in the skin did not differ significantly among three genotypes (14.2 ± 2.6% for WT, 15.0 ± 1.0% for Het, and 13.9 ± 1.0% for KO) (S1B Table). The majority of skin collagen was type I collagen (>85%), which is similar to tail tendon [24] and dentin [23].…”

“…The distributions of Lys hydroxylation/glycosylation at specific sites within the triple helical region of type I collagen were semiquantitatively estimated by LC-QTOF-MS using tryptic digests of skin samples as described in previous studies [21, 23, 24]. Although the values of WT and Het were essentially identical to each other at all analyzed sites, Lys modifications were affected in a site-specific manner in the absence of CypB (Table 1).…”

“…At α2 Lys-933, it was 30.4% in KO and 100% in WT/Het. For α1 Lys-930, using collagenase/pepsin digest as previously reported [23], we analyzed Lys modifications in the peptide containing α1 Lys-918/930 (GD K GETGEQGDRGI K GHR). In WT/Het, these sites were almost fully hydroxylated (Lys + Lys = 0%, Lys + Hyl = 4.1%, and Hyl + Hyl = 95.9%), however, at least 66.4% of those in KO was nonhydroxylated (Lys + Lys), indicating that α1 Lys-930 in KO was significantly underhydroxylated compared to WT/Het (Table 1).…”

“…In tissues of CypB-null (hereafter referred to as CypB KO) mice, a model of recessive type IX OI [21], prolyl 3-hydroxylation at α1(I)-986, the major site for this modification, is severely suppressed [21–24]. Furthermore, a series of recent studies on the CypB KO mouse tissues have provided evidence that CypB also modulates Lys hydroxylation in a tissue- and molecular site-specific manner affecting cross-linking chemistry, fibrillogenesis and tissue formation [21, 23, 24]. However, the extent of alterations in Lys modifications, the molecular mechanism of CypB-controlled LH functions and the functional outcome in different tissues are still not well understood.…”