1.                   Introduction 

The major component of mineralized dentin tissue was found to be collagen type I. Besides the collagenous proteins, acidic non collagenous proteins were identified and Dentin Sialoprotein (DSP) and phosphophoryn (PP) were found to be the two most abundant acidic non collagenous proteins in dentin [1,2], which were postulated to play significant roles during tissue mineralization. PP was identified in 1967 by Veirs and Perry [3]. PP is an extremely acidic protein and well established as a mineral nucleator for dentin mineralization [4-6]. DSP was identified in 1981 [7]. 

Since the reports of DSP-PP cDNAs from rat, mouse, human and porcine, and that DSP and PP are derived from a DSP-PP gene [8-14], numerous efforts have focused on the DSP-PP cleavage site and on the protease responsible for DSP-PP cleavage. Despite intensive investigations since 2001, there was no direct evidence proving that G447|D448 is the DSP-PP cleavage site and only Western blot analysis to substantiate the claim that BMP1 can correctly cleave DSP-PP. The understanding of this crucial process was rather limited due to a lack of large quantities of DSP-PP precursor protein readily available from tissues or cultured cells [15]. To overcome this limitation, we recently described a method using the Sf9 baculovirus expression system, in which we infected Sf9 cells with baculovirus, a family of viruses known to infect insects and represents a robust method for producing recombinant proteins, containing recombinant DSP-PP240 to yield large amounts of DSP-PP240 precursor protein. 

We reported that DSP-PP240 produced by baculovirus-infected Sf9 cells can be cleaved accurately into DSP and PP by the endogenous processing enzyme Sf9 Tolloid-Related 1 (TLR1), a homolog for human Bone Morphogenic Protein 1 (BMP1), the proposed protease to cleave DSP-PP in human. It was also discovered via mass spectrometric analysis that the specific cleavage occurred at the site: SMQG447|D448DPN [15,16]. In addition, we reported that any mutations within the DSP-PP P4 to P4’cleavage site can block, impair or accelerate DSP-PP cleavage [17,18], which suggest that its BMP1 cleavage site is highly conserved to regulate its cleavage efficiency. Furthermore, mutations outside of the DSP-PP P4 to P4' cleavage site can impair or accelerate DSP-PP cleavage [17], which suggest that these distal conserved residues may participate in exosite interactions or affect conformation at the cleavage site that is important for DSP-PP catalysis. 

The C-terminal 18 residues of DSP-PP240 are most highly conserved among the sequences of all six-mammalian species (Figure 1). Strong conservation of this region suggests the possible presence of a functional domain. We hypothesized that the C-terminal peptide D686D687 in DSP-PP actively participates in controlling DSP-PP cleavage. Here, we investigate the role of the highly conserved DSPP C-terminal region in DSP-PP cleavage. We generated a DSP-PP C-terminal mutation by substituting the terminal’s two aspartate residues for two histamine residues (DD/HH-DSP-PP). To test the impact of the DD/HH mutant on DSP-PP cleavage, we used the Sf9 expression system’s endogenous TLR1 and recombinant BMP1.

2.                   Materials and Methods

2.1.              Site-Directed Mutagenesis

Mutations were created in the DSP-PP240 cDNA in the vector pGEM7Z (+) using 35-nucleotide sense and antisense mutagenic primers following the Strata gene quick change site-directed mutagenesis protocol (Stratagene, La Jolla, CA). The 35-nucleotide sense (5’AACCACTCAACCAGTCATCATTAGAGCAGAGAACC3’) and antisense (5’GGTTCTCTGCTCTAATGATGACTGGTTGAGTGGTT3’) mutagenic primers were used to generate the D686D687/H686H687 mutant from DSP-PP240 cDNA (Figure 1). These two mutant sites are located at the C-terminal conserved region of DSP-PP precursor protein (Figure 1). Mutations were verified by DNA sequence analysis, and mutated cDNAs were sub cloned into the baculovirus recombination vector pVL1392 to generate infectious baculovirus containing the mutated cDNA.

2.2.              Baculovirus Expression System

The mutant cDNA from site-directed mutagenesis was verified by DNA sequence analysis and sub cloned into the baculovirus recombination vector pVL1392 to generate infectious baculovirus containing the mutated cDNA. The baculovirus was then used to infect Sf9 cells to allow the expression of the mutated recombinant protein.

2.3.              Partial Purification of Mutated Recombinant Proteins using Polyanion Extraction

Recombinant acidic proteins DSP-PP precursor, DSP, and PP are soluble in 5% TCA, which was used to isolate the recombinant proteins. DSP-PP related a recombinant protein in 5% TCA was then neutralized with 3M Tris. HCl pH 8.8 and precipitated with 1M CaCl2. The pellet was re-suspended in 0.1 M EDTA.

2.4.              Recombinant DSP-PP240 Protein Preparation for Human BMP1 Cleavage Studies

Secretion of TLR1 protease from Sf9 cells diminished 3 days after infection. After 3 days, fresh Grace Medium was added to the culture of baculovirus-infected Sf9 cells and incubated at 28°C for an additional 4 days. This new 4-day Condition Medium (CM), called CM3-7d, was mostly unprocessed due to the lack of TLR1 and was used as a substrate for cleavage reactions by BMP1.

3.                   Results

3.1.              Wild-Type (wt) Recombinant DSP-PP240 Precursor Protein was Cleaved into DSP430 and PP240 in 4 day Sf9 Culture Medium.

wt DSP-PP240 cDNA in baculovirus vector expresses a recombinant DSP-PP240 precursor protein, which was cleaved into DSP430 and PP240 by TLR1 in 4 day sf9 culture medium (Figure 2).

Mutant DSP-PP240 cDNA (see Figure 1) in baculovirus vector expresses only a recombinant DSP-PP240 precursor protein. No DSP430 and PP240 cleavage products were detected. Thus mutant DSP-PP240 was not cleaved at all (Figure 2) by TLR1 in the culture medium.

3.2.              Can longer incubation of mutant DSP-PP240 from 4d Sf9 culture medium enhance cleavage by TLR1 in the medium?

Mutant DSP-PP240 in 4d Sf9 culture medium showed only DSP-PP240 precursor protein. No DSP430 or PP240 cleaved products were detected (Figure 3, Lane 2). Previously we reported that TLR1 is the protease responsible for wt DSP-PP240 cleavage and the longer incubation time with conditional medium enhanced wt DSP-PP240 precursor protein cleavage into DSP430 and PP240. Here we tested whether longer incubation of mutant DSP-PP240 with TLR1 in the Sf9 culture medium would affect the cleavage. As shown in (Figure 3 Lane 3), only mutant DSP-PP240 precursor protein was present. Thus longer incubation did not enhance mutant DSP-PP240 cleavage. Thus TLR1 in the culture medium could not cleave mutant DSP-PP240 precursor protein. 

3.3.              Can BMP1 Cleave Mutant DSP-PP240 from 4d Sf9 Culture Medium?

The TRL1 in the culture medium did not enhance mutant DSP-PP240 cleavage. An analysis of a partial TLR1 cDNA from Sf9 cells indicates that residues that line the substrate-binding cleft of Sf9 TLR1 and human BMP1 are nearly perfectly conserved, offering an explanation of why Sf9 cells so accurately process mammalian DSP-PP [16]. We further tested whether the addition of human recombiant BMP1 protein could affect mutant DSP-PP240 precursor protein cleavage. As shown in (Figure 3), Lane 4, the addition of BMP1 into culture medium did not cleave the mutant DSP-PP240 precursor protein. In other words, the presence of both BMP1 and TLR1 did not lead to DSP-PP240 cleavage.

3.4.              Can human recombinant BMP1 cleave wild type and mutant DSP-PP240 in CM3-7d which contains no or low amount of TLR1 protease?

Recombinant wt and mutant DSP-PP240 proteins (see Figure 1) were from 3-7d condition medium (CM3-7d). contains no or low endogenous Sf9 TLR1[16]. From Figure 4 wt lane, we see no cleavage of recombinant wt DSP-PP240 derived from CM3-7d due to this lack of protease. When the wt CM3-7d substrate was incubated with BMP1 (170 g/ml) at 37oC for 24 hours, DSP-PP240 precursor protein was greatly reduced and the cleaved product PP240 appeared (Figure 4, wt+BMP1 lane). Recombinant mutant DSP-PP240 from CM3-7d culture medium only showed uncleaved DSP-PP240 precursor protein (Figure 4 Lane mutant). Recombinant mutant DSP-PP240 CM3-7d was incubated with BMP1 (170 g/ml), no DSP-PP240 cleavage occurred and no cleaved product PP240 appeared (Figure 4, Lane mutant+BMP1). Thus BMP1 could not cleave recombinant mutant protein in CM3-7d medium.

4.                   Discussion

Previously we reported that wt DSP-PP240 cDNA in baculovirus expression system expresses DSP-PP240 precursor protein which undergoes cleavage to generate cleavage products DSP430 and PP240 via the TLR1 protease present in Sf9 culture medium. Incubation of the recombinant DSP-PP240 precursor protein in the presence of TLR1 results in further DSP-PP240 cleavage. The exact cleavage site was established as SMQG|DDPN [16].

The C-terminal 18 residues of DSP-PP240 are highly conserved among the sequences of all seven mammalian species (Figure 1). Strong conservation of this region suggests the possible presence of a functional domain. Indeed, C-terminal mutation (i.e., D687 to M) generated a new, more slowly migrating minor band just above PP240, suggesting the possibility that a novel upstream cleavage site was activated [17]. Since single C-terminus mutation D687M affected DSP-PP240 precursor protein cleavage, we further examined the conservation of D686D687 on DSP-PP240 precursor protein cleavage. We mutated D686D687 to H686H687 and found these double mutations totally blocked DSP-PP240 cleavage in Sf9 baculovirus expression system. This double mutant DSP-PP240 precursor protein was not able to cleave by endogenous TRL1 secreted by Sf9 cells into the culture medium. Mutant DSP-PP240 recombinant protein substrate from CM3-7d contains low level of TRL1 or no TRL1 was added with human recombinant BMP1 to test whether BMP1 is able to cleave the mutant DSP-PP240 substrate. We found that wt DSP-PP240 can be cleaved by BMP1, in contrast mutant DSP-PP240 was not. Both single C-terminal D687M and double C-terminal D686D687/H686H687 mutants affect DSP-PP240 precursor protein cleavage. We speculate that these C-terminal D686 and D687 residues may participate in exosite interactions or affect conformation at the cleavage site that is important for DSP-PP catalysis. The conservation of C-terminal region likely play a crucial role in controlling proper DSP-PP precursor protein cleavage. Future work needs to further dissect the mechanism how C-terminal conservation in substrate conformation and possible excite interactions with BMP1.

In summary, wt DSP-PP240 can be cleaved by both TLR1 and BMP1 to generate DSP430 and PP240. The C-terminal double mutation of D686D687/H686H687 completely shut down the cleavage process. Thereby this finding affirms our hypothesis that the C-terminal peptides D686D687 actively participates with BMP1 in controlling DSP-PP cleavage and that they must be conserved for the cleavage process to take place normally.

5.                   Acknowledgement

We want to thank Dr. David Ritchie for helpful discussions in experimental design and help with editing the manuscript. This work was supported by NIH RO1 DE18901. The authors deny any conflicts of interest.

Figure 1: Alignment of DSP-PP₂₄₀ C-terminal sequences from marsupial and placental mammals. The heavy bar indicates the residues flanking the cleavage site between DSP and PP from P₄ to P₄'. Mutations (i.e., D⁶⁸⁶D⁶⁸⁷/H⁶⁸⁶H⁶⁸⁷) were generated in the C-terminal (see arrows). Purple highlights amino acid residues perfectly conserved in all seven species. Yellow highlights residues identical in six or seven species. Blue highlights residues identical in four species. The alignment for the C-terminal 18 amino acids is shown for the six placental mammalian species.

Figure 2: Recombinant wild type (wt) DSP-PP₂₄₀ and D⁶⁸⁶D⁶⁸⁷/H⁶⁸⁶H⁶⁸⁷mutant DSP-PP₂₄₀ protein expression. M: size marker. Wild type (wt) DSP-PP₂₄₀ cDNA in baculovirus expression vector: DSP-PP₂₄₀, precursor band, cleaved products DSP₄₃₀ and PP₂₄₀ bands were detected in 4d sf9 culture medium. D⁶⁸⁶D⁶⁸⁷/H⁶⁸⁶H⁶⁸⁷mutant DSP-PP₂₄₀ cDNA: Only DSP-PP₂₄₀ precursor band was detected in mutant lane. Neither cleaved product DSP₄₃₀ nor PP₂₄₀was found in the mutant lane.

Figure 3: Effect of endogenous TLR-1 and-exogenous BMP1 on recombinant D⁴⁸⁶D⁴⁸⁷/H⁴⁸⁶H⁴⁸⁷ mutant DSP-PP₂₄₀ protein expression. Lane 1: size marker. Lane 2: Recombinant DSP-PP₂₄₀ Mutant protein from 4d Sf9 Culture Medium (CM). Lane 3: Mutant DSP-PP₂₄₀ in baculovirus expression vector from 4d Sf9 culture medium CM was incubated at 28^oC for an additional 24 hours. Lane 4: Mutant DSP-PP₂₄₀ in baculovirus expression vector from 4d Sf9 culture medium was incubated with BMP1 (170ng/ml) for 24 hours at 28^oC.

Figure 4: Effect of BMP1 on the cleavage of recombinant CM_3-7d substrates (containing no or few TLR-1) from wt and mutant. M: size marker. wt: wild type DSP-PP₂₄₀ precursor substrate(control) derived from CM_3-7d Sf9 condition medium. wt+Bmp1: wild type recombinant DSP-PP₂₄₀ proteins from CM_3-7d was incubated with BMP1 (170 ng/ml at 28⁰C) at 37^oC for 24 hours. mut: recombinant mutant D⁶⁸⁶D⁶⁸⁷/H⁶⁸⁶H⁶⁸⁷ DSP-PP₂₄₀ from CM_3-7d medium. mut+Bmp1: mutant DSP-PP₂₄₀ from CM_3-7d medium was incubated with BMP1 (170 ng/ml at 28⁰C) at 28^oC for 24 hours.

1.       Butler WT, Bhown M, Brunn JC, D'Souza RN, Farach-Carson MC, et al. (1992) Isolation, Characterization and Immunolocalization of a 53-kDal Dentin Sialoprotein (DSP). Matrix 12: 343-351.

2.       Linde A, Bhown M, Butler WT (1980) Noncollagenous proteins of dentin. J Biol Chem 255: 5931-5942.

3.       Veis A, Perry A (1967) The phosphoprotein of the dentin matrix. Biochemistry 6: 2409.

4.       Linde A, Lussi A (1989) Mineral induction by polyanionic dentin and bone proteins at physiological ionic conditions. Connective Tissue Research 21: 197-202.

5.       Veis A (1985) Phosphoproteins of Dentin and Bone Do they have a role in matrix mineralization?. In: Butler Wt(ed.). Chemistry and Biology of Mineralized Tissues. EBSCO Media, Birmingham, AL, USA. Pg No: 170-184.

6.       Linde A (1989) Dentin Matrix Proteins: Composition and Possible Functions in Calcification. Anatomical Record 224: 154-166.

7.       Butler WT, Bhown M, DiMuzio MT, Linde A (1981) Noncollagenous proteins of dentin. Isolation and partial characterization of rat dentin proteins and proteoglycans using a three-step preparative method. Collagen Re Res 1: 187-199.

8.       Gu K, Chang S, Ritchie HH, Clarkson BH. Rutherford RB (2000) Molecular cloning of a human dentin sialophosphoprotein gene. Eur J Oral Sci 108: 35-42.

9.       MacDougall M, Gu TT, Luan X, Simmons D, Chen J (1998) Identification of a novel isoform of mouse dentin matrix protein 1: spatial expression of in mineralized tissues. J Bone Miner Res 13: 422-431.

10.    Ritchie HH, Wang L-H (1996) Sequence Determination of an Extremely Acidic Rat Dentin Phosphoprotein. Journal of Biological Chemistry 271: 21695-21698.

11.    Ritchie HH, Wang LH (1997) A Mammalian bicistronic transcript encoding two dentin-specific proteins. Biochem Biophy Res Co 231: 425-428.

12.    Ritchie HH, Wang LH (2000) The presence of multiple rat DSP-PP transcripts. Biochim Biophys Acta 1493: 27-32.

13.    Ritchie HH, Wang LH, Knudtson K (2001) A Novel Rat 523 Amino Acid Phosphophoryn: Nucleotide Sequence and Genomic Organization. Biochim Biophys Acta 1520: 212-222.