| Sign In to gain access to subscriptions and/or personal tools. |
Biomineralization: On the Trail of the Phosphate. Part II: Phosphophoryn, the DMPs, and MoreDepartment of Cell and Molecular Biology, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611; aveis{at}northwestern.edu
Key Words: dentin • matrix proteins • amelogenesis • phosphoproteins Part I of this essay described my rather unplanned and devious path into the arenas of dental science and mineralized tissue biology. Looking back, I can hardly believe I had the temerity to join Northwestern in Biochemistry. As a physical chemistry PhD, and chemical engineering undergraduate, I had only one course in "biochemistry", Irv Klotz’ graduate Physical Biochemistry, which was really just an "applied" physical chemistry course. I had not studied biochemistry or biology of any sort; I had to learn "on the job", so to speak, as I taught those subjects. It was hard work but always exciting, fresh, and wonderfully educational for me in the truest sense. I am sure that this different background also had an impact on my perspective and approach to the work on dentin.
Highlights of the early work introduced in Part I were the observations that part of the phosphate in the dentin matrix was covalently bound (Veis and Perry, 1967; Carmichael et al., 1971, 1977; Volpin and Veis, 1971, 1973), and was related to the major soluble phosphorylated protein of the ECM, which I named "phosphophoryn" (PP) (Veis et al., 1972; DiMuzio and Veis, 1978a,b). The PP was shown to be associated with mineralized dentin (Carmichael et al., 1977) and absent from the predentin. The soluble PP from rat incisors was not homogeneous, shown both on gels and by biosynthetic labeling studies (DiMuzio and Veis, 1978a,b; Maier et al., 1981, 1983; Veis et al., 1982). 31P-NMR studies verified that the phosphates were phosphate esters, with pK While the phosphoprotein work progressed, I continued work on collagen with equal intensity. My book The Macromolecular Chemistry of Gelatin was published in 1964. A third project returned to the thermodynamics of polyion interactions and consequent phase separation (complex coacervation). Not many people in the collagen or biomineralization fields know of that work. It went well, and in 1967 I received a John Simon Guggenheim Fellowship to support a sabbatical year with polymer chemist Paul J. Flory at Stanford Nobel Prize, 1968). That was a remarkable year with one of the greatest intellects I have encountered. At Stanford, I finished editing Biological Polyelectrolytes, which included a chapter of my own on "Phase Equilibria in Systems of Interacting Polyelectrolytes" (Veis, 1970). Unfortunately, in the fall of 1968, I learned that my grant to continue work on the coacervation problem was not funded. The Study Section critique suggested that I did not have a sufficiently strong mathematical background to carry out the work proposed. I reluctantly decided not to pursue that work for two reasons: PhD students interested in physical chemistry were not easy to find at the Medical School; and I was asked to join the Medical School Administration to manage both the Honors Program in Medical Education and the Graduate School operations on the Chicago Campus. I served as Associate Dean in the Medical and Graduate Schools until 1977, when I took a second sabbatical, as a Fogarty Senior International Fellow: six months in the European Molecular Biology Laboratory, in Grenoble, France, to work with Andrew Miller on collagen structure using small-angle x-ray diffraction and neutron-scattering techniques; and six months with Wolfie Traub at the Weizmann Institute of Science in the Protein Structure Laboratory (Veis et al., 1979). At the Weizmann, I encountered Steve Weiner, whom I had met earlier. Steve was in the Isotope Department working on invertebrate mineralization, continuing along the line of his PhD studies at Cal Tech. With Traub, I was trying to crystallize PP and get some wide-angle x-ray data. That did not succeed, but I did bring Steve and Wolfie together, and they have collaborated with spectacular success. I consider that as much a contribution to science as I have achieved with my own work. Before I left for my sabbatical, Dean Olsen of the NU Dental School had invited me to organize an Oral Biology Department in the Dental School. After some deliberation, I accepted and joined NUDS on my return at the end of 1977. I stayed with the Dental school, through thick and thin, until I retired to become Emeritus Professor in 1997, at about the same time that the Dental School was closed. Happily, I had maintained my joint appointment in the Medical School, and the Medical School Deans have been kind enough to allow me to retain much of my laboratory space so that my work continues. As I considered the data on the PP noted above, I had begun to think about biomineralization in broader terms than just PP and collagen. At the Weizmann, Steve Weiner and I took coffee together most mornings, and our conversations frequently turned to our differences in approach to the phosphate and carbonate mineralization systems. After returning home, I decided that most of the people in the field were indeed compartmentalized, and that we needed to have more "idea and concept" discussions. At breakfast while in New York at an AAAS meeting, I broached that subject with William Butler and sought his help in organizing an international meeting that would consider all aspects of biomineralization. Bill could not have been more supportive or helpful, and together we brought into being the First International Conference on the Chemistry and Biology of Mineralized Tissues, held in 1981 in Chicago at the Dental School, coinciding with the opening of the new laboratories of the Oral Biology Department. Heinz Lowenstam presented the Plenary lecture and discussed the wide range of mineralized tissues and wide variety of minerals created in different organisms. That lecture, combined with the ideas fermenting in the background from my discussions with Steve Weiner, brought me to consider the broader question as to whether there was a common theme or strategy for biomineralization in general. Those ideas were presented in "Bone and Tooth Formation: Insights into Mineralization Strategies" at a meeting in The Netherlands in 1982 (Veis and Sabsay, 1982). That paper set forth our matrix-structure-mediated, NCP-mediated mechanism for vertebrate tissue mineralization processes, with the implication of generalization to all mineralization processes. At the same meeting, Lowenstam and Weiner presented "Mineralization by Organisms and the Evolution of Biomineralization", elaborating on Lowenstam’s proposal of the organic matrix-mediated mineralization in the invertebrate systems. All of those ideas have guided my present work, especially the branching out into consideration of mineralization in sea urchin teeth (Veis et al., 1986, 2002; Stock et al., 2002a,b, 2003a, Stock et al., b), work presently very active in the laboratory. The 1978 move to the Dental School and Oral Biology had other effects. The necessity to organize the courses and teach in Oral Biology at the dental student and post-graduate levels required me to educate myself in the subject, and the related cell biology, molecular biology, and developmental biology. This affected my outlook and perspective in many ways and led to changes in my approach to phosphophoryn studies. It was clear that if we stuck to the biochemistry approach alone, the sequencing of the PP would have required the efforts of everyone in the lab, so we finally turned to molecular biology and cloning. John S. Evans (DDS) came to the lab to work on collagen biosynthesis, but he was also eager to work on the cloning of the PP. Although I was not knowledgeable in that arena, our cloning effort began. John isolated the mRNA from rat incisors and prepared a cDNA library for screening, but with important gaps in knowing which probes to use, and how to do this most effectively. John moved to Harold Slavkin’s laboratory, but Philip Simonian joined the group for a one-year break in his dental curriculum. One of the best moves I made was to send Phil to Marian Young’s laboratory at NIDR, where he spent the first two months of his year being educated in the details of library screening. Marian’s contribution was enormous, and she was most generous with her time and effort. At the same time I sent Przemko Tylzanowski, a PhD student working on collagen mRNA translation, to Gideon Rodan’s laboratory at Merck to learn the latest techniques of full-length mRNA isolation. When both returned, we combined forces, isolated the mRNA from rat incisors, prepared the corresponding cDNA library, and screened it with antibodies to riPP and with nucleotide probes representing potential sequences as suggested in our biochemical work (Sabsay et al., 1991). The screening provided 7 unique clones. At that point, Philip returned to the regular program and Przemko to his collagen work to complete his dissertation. There has always been regular, formal communication among the people in my lab via weekly meetings, so that everyone is aware of what is going on. Anne George, then a physical chemistry post-doc, had completed a two-year study of collagen fibrillogenesis by FTIR. Anne was attracted by the ongoing cloning of the PP and asked if she could do a second post-doc and complete the PP cloning. I was about to hire a molecular biologist, but, impressed with Anne’s ability, diligence, and intelligence, I agreed, even though it again meant "square one" training. I have not regretted that decision. It may have taken us longer, but the result was the first of the dentin matrix proteins, Dentin Matrix Protein 1 (DMP 1), on human chromosome 4q21 (George et al., 1992, 1993, 1994, 1995), followed by the phosphophoryn part of what is now called DSPP, DMP 2 (George et al., 1996; Veis et al., 1998), also on 4q21, and the full rat DSPP, DMP 3 (George et al., 1999). Helena Ritchie (1994) in Bill Butler’s lab used our rat incisor cDNA library and sequenced DSP. By the time we had accomplished these studies, others had entered into similar efforts. Mary MacDougall (MacDougall et al., 1997) was the first to report that DSP and DPP were localized as a single gene product. We had evidence that this was the case, but for once I was too cautious to move quickly to tie our work to that of Helena Ritchie on DSP. I was rather severely chided for being too stubborn to adopt the DSPP concept (Simmer, 2000) as the sole gene product, but I had, and still have, some reservations based on the large difference in contents of DSP and DPP in dentin shown by the biochemical studies, and the fact that some in situ hybridization studies indicate that probes to the supposed DSP and PP message sequences do not detect these message sequences in parallel in identical locations in the developing tooth (Kulkarni et al., 2000). Now that, as Professor Emeritus, I have a smaller laboratory, I have had to refocus my work on the PP. Since so many good molecular biologists are working on the cloning, I decided to put my efforts back into the mineralization mechanisms, and returned to the interaction of the phosphophoryn with collagen to initiate mineralization (Traub et al., 1992; Veis, 1997; Dahl et al., 1998; Weiner et al., 1999; Veis et al., 2000b; Beniash et al., 2000; Dahl and Veis, 2003). Currently, recombinant DMP2 is being used to represent the PP. We are attempting to phosphorylate it in vitro to the same extent as in vivo. This is proving to be a challenging problem and may ultimately shed more light on the mechanisms of phosphorylation as well as on the mineralization mechanisms. While not dealing specifically with the phosphoproteins, another branch of our work on dentin has opened up an entirely new area, and it deserves mention. Early studies by Marshall Urist showed that dentin matrix had a strong BMP-like activity relative to bone induction at ectopic sites. Kari Koskinen (PhD, DDS), an endodontist at the University of Helsinki, came for a year. When we discussed his project, Kari posed the problem of the repair of dentin defects using BMP. This was in the early 1980s, before BMP was cloned and made available as a recombinant protein, so we had the not-too-brilliant idea of using the dentin matrix as our source of BMP. I preferred that we use in vitro assays rather than implants, so we developed in vitro assays and showed that the dentin matrix proteins could induce cultured fibroblasts to express a heightened production of sulfated glycosaminoglycans and type II collagen, indicative of a switch to the chondrogenic phenotype (Koskinen et al., 1985). The relevant BMP activity was in a fraction that we had been discarding during PP extraction. Bryan Sires (MD-PhD student) and John Clohisy (MD student) isolated the BMP-active fraction (Veis et al., 1989, 1990). Salomon Amar (PhD, DDS), from Strasbourg, France, showed definitively that the active rat incisor dentin component was not a BMP (Amar et al., 1991). It contained no cysteine. Unfortunately, we were unable to determine the sequence. For her PhD work, Denise Nebgen (DDS, MD) shifted to bovine dentin to obtain larger yields and finally isolated the active protein. To our surprise, the active fraction was identified as one of the small splice products of the amelogenin gene (Nebgen et al., 1999). Had we spent all that effort isolating an amelogenin impurity in our dentin preparation? Many in the amelogenin community seemed to think that was the case. However, our work on the DSPP and DMP1 showed that during tooth development these "dentin-specific" proteins and their messages were transiently expressed in ameloblasts. It seemed to me that we should check to see if amelogenins were similarly expressed in odontoblasts. I was confident that any ameloblasts had been removed before RNA extraction in the preparation of our rat incisor odontoblast cDNA library. Thus, we used PCR to probe that library for amelogenin messages, and found that the small amelogenin splice product mRNAs, corresponding to the rat 59- (LRAP, [A-4]) and 73-amino-acid ([A+4]) peptides, were indeed present (Veis et al., 2000a). Recombinant peptides were produced and shown to have the predicted activities in both in vitro chondrogenesis experiments and ectopic muscle pouch implants. We also showed that the two peptides acted differently but that both were transcription factors, directly or indirectly inducing the expression of Cbfa1 and Sox 9, required for bone and cartilage production. In situ hybridization showed the transient appearance of the messages for the amelogenin peptides in the odontoblasts of mouse molars and incisors. My ideas regarding the significance of the transient presence of the amelogenin peptides in the odontoblasts have been summarized recently (Veis, 2003). Currently, with PhD student Kevin Tompkins (DMD), we are examining the effects of the peptides on tooth bud development and determining the nature of the amelogenin peptide receptors (Tompkins and Veis, 2002). It is most interesting that [A-4] has an inhibitory effect on the polarization of pre-ameloblasts. My latest hypothesis (Veis, 2003) is that this may be related to the relative timing of odontoblast and ameloblast maturation to the matrix-producing secretory stage. The [A+4] peptide, on the other hand, stimulates collagen and PP production, that is, dentin matrix production. Kari Koskinen’s quest for a method for dentin bridge repair that had started us down this path may finally be realized. A great friend and collaborator, Michel Goldberg, in Paris, has been doing these experiments in his laboratory, and the amelogenin peptide implants in mice seem to do the job (Veis et al., 2001; Goldberg et al., 2002, 2003; Six et al., 2002). It has been good to feel that I have been a part of the start of new fields and new ideas, and quite rewarding to have made so many friends in all parts of the world and to have had so many individuals provide help at crucial times. When I started out on a career in science, it seemed that it would be all about facts and experiments and papers. But that is really not the case. The real rewards come from the interactions with others who share the same intellectual passions. One may not see a colleague for months at a time, but pick up on the conversation without pause at the next meeting—with more intensity than with most neighbors one sees every day. The other amazing thing is that although I have been essentially working on the same problems for the past 50 years, it has never been boring or tedious. The ground keeps shifting, and new directions keep opening up, so that the future looks just as fresh, interesting, and challenging as it did at the start. My next task is to learn how to stop. Received for publication July 26, 2003. Accepted for publication September 3, 2003. REFERENCES
Journal of Dental Research, Vol. 83, No. 1,
6-10 (2004)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
= 6.8 (
1.5 x 105, nearly 90% of the Ser residues were phosphorylated, and the combined phosphate and carboxylate residues yielded a net negative charge (at neutral pH) of ~ –1.3/amino acid residue. The high P-Ser and Asp content made conventional sequencing difficult, but we showed that there were specific domains of different character (