Spatial survey of non-collagenous proteins in mineralizing and non-mineralizing vertebrate tissues ex vivo

Abstract

Bone biomineralization is a complex process in which type I collagen and associated non-collagenous proteins (NCPs), including glycoproteins and proteoglycans, interact closely with inorganic calcium and phosphate ions to control the precipitation of nanosized, non-stoichiometric hydroxyapatite (HAP, idealized stoichiometry Ca₁₀(PO₄)₆(OH)₂) within the organic matrix of a tissue. The ability of certain vertebrate tissues to mineralize is critically related to several aspects of their function. The goal of this study was to identify specific NCPs in mineralizing and non-mineralizing tissues of two animal models, rat and turkey, and to determine whether some NCPs are unique to each type of tissue. The tissues investigated were rat femur (mineralizing) and tail tendon (non-mineralizing) and turkey leg tendon (having both mineralizing and non-mineralizing regions in the same individual specimen). An experimental approach ex vivo was designed for this investigation by combining sequential protein extraction with comprehensive protein mapping using proteomics and Western blotting. The extraction method enabled separation of various NCPs based on their association with either the extracellular organic collagenous matrix phases or the inorganic mineral phases of the tissues. The proteomics work generated a complete picture of NCPs in different tissues and animal species. Subsequently, Western blotting provided validation for some of the proteomics findings. The survey then yielded generalized results relevant to various protein families, rather than only individual NCPs. This study focused primarily on the NCPs belonging to the small leucine-rich proteoglycan (SLRP) family and the small integrin-binding ligand N-linked glycoproteins (SIBLINGs). SLRPs were found to be associated only with the collagenous matrix, a result suggesting that they are mainly involved in structural matrix organization and not in mineralization. SIBLINGs as well as matrix Gla (γ-carboxyglutamate) protein were strictly localized within the inorganic mineral phase of mineralizing tissues, a finding suggesting that their roles are limited to mineralization. The results from this study indicated that osteocalcin was closely involved in mineralization but did not preclude possible additional roles as a hormone. This report provides for the first time a spatial survey and comparison of NCPs from mineralizing and non-mineralizing tissues ex vivo and defines the proteome of turkey leg tendons as a model for vertebrate mineralization.

Keywords: B, rat bone; BSP, bone sialoprotein; DCN, decorin; E, EDTA extract; ECM, extracellular matrix; G, guanidine-HCl-only extract (for non-mineralizing tissues); G1, first guanidine-HCl extract; G2, second guanidine-HCl extract; Gla, gamma-carboxylated glutamic acid; MGP, matrix Gla protein; MT, turkey mineralizing tendon; Mineralization; NCP, non-collagenous protein; NMT, turkey never-mineralizing tendon; NT, turkey not-yet-mineralized tendon; Non-collagenous protein; OCN, osteocalcin; OPN, osteopontin; Proteomics; SIBLING, small integrin-binding ligand N-linked glycoprotein; SLRP, small leucine-rich proteoglycan; T, rat tail tendon; TLT, turkey leg tendon (gastrocnemius); TNAP, tissue-nonspecific alkaline phosphatase; Type I collagen; Vertebrate.

Fig. 1

Schematic diagram of protein extraction and proteome analysis procedures for rat and turkey mineralizing and non-mineralizing tissues. To study the protein inventory, tissues were isolated from ten-week-old normal male Sprague-Dawley rats and sixteen-week-old male domestic turkeys (a). The femoral bone (B) and tail tendon (T) were dissected from rats. The dissected leg tendons of turkey were divided into three regions: a mineralizing (MT) region, a region that has not-yet-mineralized (NT) at the time the turkey was sacrificed, and a never-mineralizing region (NMT) (b). These tissue sections were then pulverized prior to the protein extraction procedure. Sequential extraction steps for mineralizing tissues consisted initially of guanidine-HCl, followed by EDTA, and subsequently a second guanidine-HCl treatment yielding the protein extracts G1, E, and G2, respectively. The non-mineralizing tissues were extracted using only guanidine-HCl (G) (c). Proteins were then identified by proteomics and Western blotting (d). After each treatment, the protein extracts were tryptically digested (e) and the resulting peptides separated by nanoLC hyphenated to an Orbitrap Fusion mass spectrometer and analyzed by tandem mass spectrometry (MS/MS) (f, g). Proteins were identified and quantified through a label-free approach using the PEAKS Studio 8.5 software platform (h). P: proximal; D: distal; ACN: acetonitrile; DDA: data-dependent acquisition.

Fig. 2

Venn diagrams of the differential distribution of proteins identified in rat (a) and turkey (b) mineralizing and non-mineralizing tissues. Numbers in parentheses indicate the total proteins that were identified in each extract. Numbers in overlapping regions of the Venn diagrams show the similar proteins that were identified in more than one extract. B: rat bone; T: rat tail tendon; MT: mineralizing TLT; NT: not-yet-mineralized TLT; NMT: never-mineralizing TLT; G, G1, G2, and E: extraction steps as defined in the text and in Fig. 1.

Fig. 3

The relative distribution of SLRP proteins in extracts derived from rat (a) and turkey (b) mineralizing and non-mineralizing tissues. Validation of decorin (DCN) localization by Western blotting (c). B: rat bone; T: rat tail tendon; MT: mineralizing TLT; NT: not-yet-mineralized TLT; NMT: never-mineralizing TLT. G, G1, G2, and E: extraction steps as defined in the text and in Fig. 1. 1, 2: Specimens 1 and 2.

Fig. 4

The relative distribution of SIBLING proteins in extracts derived from rat (a) and turkey (b) mineralizing and non-mineralizing tissues and Western blot-based validation of the localization of OPN in rat (c) and turkey (d) and BSP in rat (e) and turkey (f) tissues. B: rat bone; T: rat tail tendon; MT: mineralizing TLT; NT: not-yet-mineralized TLT; NMT: never-mineralizing TLT. G, G1, G2, and E: extraction steps as defined in the text and in Fig. 1. 1, 2: Specimens 1 and 2.

Fig. 5

The identification of vitamin K-dependent NCPs in rat and turkey and a comparison of MGP abundance in rat (a) and turkey (b). The identification of OCN in rat and turkey is validated by Western blotting (c). Western blot results of tissue-nonspecific alkaline phosphatase (TNAP) in rat bone and tail tendon (d). B: rat bone; T: rat tail tendon; MT: mineralizing TLT; NT: not-yet-mineralized TLT; NMT: never-mineralizing TLT. G, G1, G2, and E: extraction steps as defined in the text and in Fig. 1. 1, 2: Specimens 1 and 2.