|Home | About | Journals | Submit | Contact Us | Français|
It is known that dentin sialophosphoprotein (DSPP) is processed into NH2- and COOH-terminal fragments, but its key cleavage site has not been identified, nor has its full-length form been discovered. The objectives of this study were to identify the key cleavage site during DSPP processing and to search for full-length DSPP in vivo. We generated a construct encoding DSPP, in which Asp452, a cleavage site residue, was replaced by Ala452. The pulp-odontoblast complex and dentin were extracted, chromatographically separated, and assessed by Stains-All staining, Western immunoblotting, and mass spectrometry. These studies showed that the substitution of Asp452 by Ala452 completely blocks the cleavage of mouse DSPP in the transfected cells, indicating that the NH2-terminal peptide bond of Asp452 is essential for the initiation of DSPP proteolytic processing. The results of this study revealed the presence of full-length DSPP and its processed fragments in extracts from the pulp/odontoblast and dentin.
The importance of dentin sialophosphoprotein (DSPP) in the formation of dentin has been illustrated by human and mouse genetic studies, which showed the association of the DSPP gene mutations or ablations with dentin defects (Xiao et al., 2001; Zhang et al., 2001; Sreenath et al., 2003). The expression of DSPP was originally thought to be tooth-specific. Recent studies demonstrated the presence of this gene in several other tissues (Qin et al., 2002; Alvares et al., 2006).
DSPP is a large protein that is processed into dentin sialoprotein (DSP) and dentin phosphoprotein (DPP), with the DSP sequence at the 5′ end and the DPP at the 3′ end of the DSPP transcript (MacDougall et al., 1997). Although derived from the same mRNA, the chemical structures of DSP and DPP are very different (Qin et al., 2004). On 7.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), the migration rate of rat DSP is ~95 kDa, while that of DPP is ~90 kDa. Thus, the full-length form of DSPP is expected to be around 185 kDa on SDS-PAGE. Analysis of data obtained through in vitro mineralization studies indicated that DPP is an important initiator and modulator for the formation and growth of hydroxyapatite crystals (Boskey et al., 1990; Saito et al., 1997; He et al., 2005). A recent in vivo study indicated that DSP may be involved in the initiation of dentin mineralization, but not in the maturation of this tissue (Suzuki et al., 2009). While DSP and DPP are found abundantly in the extracellular matrix (ECM) of dentin, the full-length form of DSPP has not been identified.
Protein chemistry work (Butler et al., 1983; Qin et al., 2001a), in combination with data from the cDNA-deduced amino acid sequence (Ritchie et al., 2001), indicates that proteolytic processing of rat DSPP to DSP and DPP involves the cleavage of the peptide bonds Gly447-Asp448, Tyr438-Asp439, and His423-Ser424. Among these identified cleavage sites, the amino acid sequences surrounding Gly447-Asp448 are highly conserved. The first 4 residues of rat DPP are Asp448-Asp-Pro-Asn451 (Butler et al., 1983), which is conserved in mouse, human, and porcine DSPP (MacDougall et al., 1997; Gu et al., 2000; Yamakoshi et al., 2003). Based on these data, we speculate that the peptide bond at the NH2-terminus of Asp452 in mouse DSPP (corresponding to Asp448 in rat DSPP) is likely the key cleavage site that is critical for the initiation of DSPP processing. This investigation tested the above hypothesis, and searched for the full-length form of DSPP in the extracts from the rodent dental tissues.
Two constructs were generated: one expressing the normal full-length DSPP and the other expressing DSPP, in which Asp452 was replaced by Ala452 (D452A substitution). To generate the normal DSPP construct, the Dspp cDNA released from the pBluescriptII SK-DSPP construct (MacDougall et al., 1997) was subcloned into the mammalian expression vector pcDNA3 (Invitrogen, Carlsbad, CA, USA), producing the pcDNA3-DSPP construct. A second construct encoding DSPP with the D452A substitution (referred to as “D452A-DSPP”) was generated with a site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA), according to the manufacturer’s instructions.
The human embryonic kidney cells HEK293 that constitutively express the Epstein-Barr Virus Nuclear Antigen (HEK293-EBNA, American Type Culture Collection, Manassas, VA, USA) and that are widely used for the production of recombinant proteins (Wright et al., 2003) were cultured in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal bovine serum (Invitrogen). The pcDNA3-DSPP, pcDNA3-D452A-DSPP, or pcDNA3 (control) plasmids were transiently transfected into the cells with lipofectamine 2000 (Invitrogen). The medium was replaced with serum-free medium 16-18 hrs after transfection, and then the transfected cells were cultured for an additional 48 hrs. The medium was collected and analyzed for DSPP and/or its processed fragments by Stains-All staining and Western immunoblotting. To test if DSPP was cleaved by bone morphogenetic protein 1 (BMP-1), we treated the medium containing the mutant or normal DSPP with recombinant human BMP-1 (R&D Systems, Minneapolis, MN, USA), as reported previously (Steiglitz et al., 2004).
Twenty-five 12-week-old C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME, USA) were used for the extraction and separation of NCPs. The animal protocol was approved by the Animal Welfare Committee of Baylor College of Dentistry, Texas A&M Health Science Center. The dental pulp/odontoblast complex was separated from the dentin of the mouse incisors according to the following protocol. Immediately after being immersed in 4 M guanidine-hydrochloride (Gdm-HCl) solution, the incisors were ground into pieces for approximately 2 min. After the Gdm-HCl solution was stirred with a magnetic bar for 3 min, the dentin pieces precipitated, while the dental pulp/odontoblast complex portion floated on top of the solution. The floating portion was quickly removed from the dentin pieces and reserved for the extraction of the NCPs. We repeated this procedure 3 times to ensure that the dental pulp cells and odontoblasts were separated from the dentin ECM. The NCPs in the dental pulp/odontoblast complex were extracted by 4 M Gdm-HCL without EDTA for 24 hrs. The precipitated dentin pieces were extracted by 4 M Gdm-HCl/0.5 M EDTA solution for 48 hrs, as previously described (Qin et al., 2001b). These 2 extracts were separated by Q-Sepharose (Amersham Biosciences, Uppsala, Sweden) chromatography with a gradient ranging from 0.1–0.8 M NaCl in 6 M urea. During this chromatography, DSPP and its processed fragments (eluted in earlier fractions) were separated from the proteoglycan form of DSP (in later fractions).
We performed Stains-All staining and Western immunoblotting to visualize DSPP and/or its processed fragments. Western immunoblotting was carried out with an anti-DSP polyclonal antibody (Butler et al., 1992) at a dilution of 1:3000, according to the protocols as previously described (Qin et al., 2001b).
To obtain a sufficient amount of DSPP for mass spectrometry (MS) analyses, we used incisors from 40 rats as the starting materials for NCP extraction. Following Q-Sepharose chromatography, the corresponding fractions containing DSPP were loaded onto 7.5% SDS-PAGE gel. The protein band with a migration rate of 185 kDa was cut off from the gel and analyzed by MS. For MS analyses, the gel band was first cut into small pieces and then digested by trypsin overnight. The resulting tryptic peptides were extracted, de-salted with C18 ZipTips, and analyzed by high-performance liquid chromatography tandem MS.Mascot 2.1 (Matrix Science, London, UK). The peptides identified for the proteins of interest were manually verified by a routine method (Chen et al., 2005). The results were compared with the amino acid sequence deduced from rat DSPP cDNA.
The Stains-All staining and Western immunoblotting results showed that the culture media of the cells transfected with the pcDNA3-DSPP or pcDNA3-D452A-DSPP constructs contained significant amounts of DSPP and/or its fragments, whereas the media from the cells transfected with empty vector did not contain any DSPP or its fragments. Stains-All staining (Fig. 1A) detected a single blue-stained protein band in the culture medium from the cells transfected with the pcDNA-D452A-DSPP construct; the migration rate of this blue-stained protein band (185 kDa) matched that of full-length DSPP (DSP + DPP = 185 kDa). In the normal DSPP sample, full-length DSPP as well as DPP were clearly visible. Since DSP is less acidic and less sensitive to Stains-All staining, this fragment in the sample (Fig. 1A) was not visualized. The results from Western immunoblotting (Fig. 1B) with the anti-DSP antibody confirmed the presence of DSP in the normal DSPP sample and its absence in the media of the cells transfected with the pcDNA-D452A-DSPP construct. These findings indicated that the D452A substitution blocked the proteolytic processing of DSPP in these cells.
Q-Sepharose chromatography separated the pulp-odontoblast or dentin extracts into 120 fractions, and the chromatographic profiles for the two types of extracts showed apparent differences (Fig. 2). In the chromatogram of the pulp-odontoblast extract, the most prominent UV absorbance peak was found in the earlier fractions (around fraction 25); these fractions contained large amounts of less acidic proteins that could be visualized by Coomassie brilliant blue. In the dentin ECM extract, the highest peak emerged in fractions in which DPP eluted (around fraction 55).
The recombinant full-length D452A-DSPP was used as a positive control in our search for the full-length form of DSPP in the extracts from the dental pulp/odontoblast complex and dentin. In the extract from the dental pulp-odontoblast complex, a blue-stained protein band with a migration rate (185 kDa) identical to that of D452A-DSPP was clearly detected by Stains-All staining (Fig. 3A) in fractions 42 to 52. This 185-kDa protein band from the pulp-odontoblast was recognized by the anti-DSP antibody (Fig. 3C), and by two other types of anti-DSP monoclonal antibodies (data not shown). We believe that this 185-kDa protein was the full-length form of DSPP, which was further confirmed by MS analysis (see below). In the dentin extract, DSPP was also detected by Stains-All staining (Fig. 3B) and Western immunoblotting (Fig. 3D). When mouse DSP was used as an internal control for normalization, the signal for the full-length form of DSPP appeared much weaker in the dentin than in the pulp-odontoblast extract. In addition, when neither the samples in the dental pulp-odontoblast nor those in the dentin extracts were treated with β-mercaptoethanol (β-ME), significant amounts of DSP in the pulp-odontoblast or dentin extract occurred in the dimeric form (Figs. 3E, ,3F),3F), and the migration rate of the DSP dimer (close to 200 kDa) was slower than that of the full-length form of DSPP on SDS-PAGE. Furthermore, we also dissected the pulp from the rat incisor using surgical scalpels. This alternative dissection protocol also showed the presence of DSPP fragments as well as its full-length form in the extract from the dental pulp-odontoblast complex.
BMP-1 digestion experiments showed that normal DSPP was completely processed into fragments, while a portion of the mutant DSPP (D452A-DSPP) was cleaved by BMP-1 to form DSP (Fig. 3G).
The MS analyses identified 6 tryptic peptides with sequences originating from the NH2-terminal (DSP) region and 1 peptide from the COOH-terminal (DPP) portion of rat DSPP (Fig. 4). In fact, the identified DPP tryptic peptide was from the COOH-terminus of rat DSPP. Analysis of the MS data showed unequivocally that this 185-kDa protein band represents the native, full-length form of DSPP.
Previous studies by our group revealed that at least 3 peptide bonds are cleaved during the processing of rat DSPP (Butler et al., 1983; Qin et al., 2001a). A recent in vitro study showed that matrix metalloproteinases (MMP-2 and MMP-20) cleaved the NH2-terminal portion of porcine DSPP (namely, DSP and dentin glycoprotein) at multiple sites, yielding several low-molecular-weight fragments (Yamakoshi et al., 2006). Another in vitro study (Godovikova and Ritchie, 2007) showed that a recombinant rat DSP-PP (a shorter form of the rat DSPP variants) underwent self-processing; one of the self-cleavages appeared to be at Gly447-Asp448, which corresponds to Gly451-Asp452 in the mouse DSPP. The results from this study showed that the D452A substitution completely blocked the proteolytic processing of mouse DSPP in the eukaryotic cells, while normal DSPP was processed into DSP and DPP. These findings suggest that cleavage at the NH2-terminus of Asp452 of mouse DSPP is critical for the initiation of DSPP processing. We believe that this cleavage represents an initial, first-step scission in the whole cascade of DSPP processing, which is essential for the release of DSP and DPP.
BMP-1/Tolloid-like metalloproteinases have been shown to cleave mouse dentin matrix protein 1 (DMP1) at the Ser212-Asp213 bond (Steiglitz et al., 2004). Although DSPP and DMP1 belong to the same Small Integrin-Binding LIgand, N-linked Glycoprotein family, the sequence homology level between the 2 proteins is low. Nevertheless, the residues surrounding the 2 cleavage sites, Gly451-Asp452 in mouse DSPP and Ser212-Asp213 in mouse DMP1, are similar and highly conserved across species. In this investigation, we incubated the normal DSPP as well as the mutant D452A-DSPP (both made by HEK293-EBNA cells) with recombinant human BMP-1. While normal DSPP was completely cleaved by BMP-1, a portion of the mutant DSPP was also processed by this proteinase to form DSP. These findings suggest that in addition to Gly451-Asp452, there might be other peptide bond(s) close to Gly451-Asp452 that can be cleaved by BMP-1 in the in vitro environment. Clearly, further in vivo (transgenic mouse) studies are needed to test whether the D452A substitution blocks DSPP processing in the animals; such investigations may help explain the discrepancy that D452A-DSPP could not be processed by the eukaryotic cells, while a portion of this mutant protein was cleaved by BMP-1 in the test tube.
Additionally, we also incubated recombinant human BMP-1 with the chromatographic fractions containing DSPP and other NCPs extracted from the mouse and rat dental tissues; these experiments showed that DSPP isolated from these tissues was cleaved by the recombinant BMP-1 (data not shown).
In this investigation, we detected the full-length form of DSPP in the cellular (pulp-odontoblast) portion and the dentin ECM; the former had more full-length DSPP than the latter. It is obvious that the quantity of DSP and DPP in the dentin ECM is greater than the full-length form of DSPP. However, at this point, we are unable to calculate accurately the exact ratio of the full-length DSPP to its processed fragments in the dentin ECM or in the pulp-odontoblast complex.
The presence of DSP and DPP in the pulp-odontoblast portion suggests that the proteolyic processing of DSPP must begin in the odontoblasts that synthesize it. The fact that a small quantity of full-length DSPP is present in the dentin ECM indicates that a minor amount of the uncleaved, full-length DSPP is secreted into the dentin matrix. Recently, we observed that the majority of DSPP in the ECM of the rat condylar cartilage is not cleaved (Sun et al., unpublished observations). It is likely that the full-length form of DSPP found in the dental tissues and the condylar cartilage plays certain biological roles that are different from those of its cleaved products.
This work was supported by NIH Grant DE005092 (CQ).
A supplemental appendix to this article is published electronically only at http://jdr.sagepub.com/supplemental.