Advances in Biotechnology for Tissue Engineering of Bone

U. Ripamonti*, and J. R Tasker

Bone Research Unit, MRC/University of the Witwatersrand, Johannesburg, South Africa

 

 

*Address correspondence to this author at the Bone Research Unit, MRC/University of the Witwatersrand, 7 York Road, Parktown 2193, Johannesburg, South Africa; Tel:/Fax: + 27 11 647-2300; E-mail: 177RIPA@chiron.wits.ac.za

 

Abstract:  Tissue engineering is a  rapidly developing field applying the disciplines of cell biology, developmental biology, molecular biology and biomimetic engineering to regenerate new tissues for replacement therapies in clinical contexts. To aid in the elicitation and reiteration of the processes of morphogenesis of tissues, the cascade of chemotaxis of progenitor cells, their differentiation and pattern formation is redeployed in postnatal tissues, effected by a variety of ever-increasing morphogens and biomaterials. The extensive recent progress in elucidating  the molecular biology of BMPs and their receptors shall aid in promoting and extending the great operational future of this field. Although the BMP family of proteins and osteogenesis have been the subject of several recent reviews, we focus here on their activity in primates and on the novel localization of BMPs in the cerebellum and other areas of the nervous system, and the “mosaicism” of their localisation in the periodontal tissues followed by a discussion on the use of BMPs in periodontal regeneration. Lastly, we report on the unique osteoinductive activity of TGF-b proteins in heterotopic sites of primates and their synergistic interaction with a  recombinant human BMP, and finally we present unique data on novel biomaterials endowed with intrinsic osteoinductive activity, capable of initiating de novo bone formation in heterotopic sites even in the absence of exogenously applied BMPs, and the results of a clinical trial in humans using naturally-derived BMPs.

Introduction

     Recent advances in the mechanisms of bone morphogenesis have now laid the foundation for novel approaches in clinical tissue engineering of bone as a by product of biotechnological advances in the field. Perhaps one of the most important discoveries in the field of osteogenesis has been the discovery that the induction of bone formation in post-natal life is not a prerogative that belongs only to the bone morphogenetic proteins (BMPs), but has also been extended to the other BMP-related members, the transforming growth factors-b1 and b2,  specifically only in primates, Fig. (1A) [1,2,3]. These were previously unknown to induce de novo bone formation in extraskeletal heterotopic sites of rodents [4,5].

     Together with the observation that the decapentaplegic (Dpp) and 60A gene products of Drosophila melanogaster [6] and growth and differentiation  factor 5 (GDF-5) [7] are osteoinductive, the fact that the TGF-b proteins induce de novo bone formation in heterotopic sites of the primate indicates multiple interactions during endochondral bone formation and morphogenesis, Fig. (1A). This raises critical questions about the biological relevance of this apparent redundancy, moreover suggesting a site, tissue and organ specific activity of BMP family members, that will only be unraveled by in vivo studies on their structure/activity relationships.

Fig. (1A).  Endochondral bone formation induced upon heterotopic implantation of 25 mg hTGF-b2 in conjunction with insoluble collagenous bone matrix. Arrows point to foci of cartilage formation.

     We now understand that the molecular mechanisms deployed during endochondral bone formation are very similar to the very mechanisms deployed for morphogenesis of other tissues and organs, again requiring a cascade of molecular events pre-dating the resulting morphogenetic events and controlled by a specific family of proteins or morphogens, capable of engineering morphogenetic responses which have a critical clinical application in morphogenesis and regeneration.

     Although the BMP family of proteins and osteogenesis have been the subject of several recent reviews, we shall focus here on their activity in primates and on the novel localization of BMPs in the cerebellum and other areas of the nervous system, and the “mosaicism” of their localisation in the periodontal tissues followed by a discussion on the use of BMPs in periodontal regeneration, Fig. (2). Lastly, we will report on the unique osteoinductive activity of TGF-b proteins in heterotopic sites of primates and their synergistic interaction with a  recombinant human BMP, Fig. (1B), and finally we will present unique data on novel biomaterials endowed with intrinsic osteoinductive activity, capable of initiating de novo bone formation in heterotopic sites even in the absence of exogenously-applied BMPs, Fig. (3), and the results of a clinical trial in humans using naturally-derived BMPs.

     The members of the bone morphogenetic protein (BMP) family belong to the larger TGF-b superfamily of soluble secreted proteins, which act as powerful regulators of development and  tissue repair in both vertebrates and invertebrates [8,9,10, 11,12,13,14,15]. The BMPs share sequence homologies in their carboxy-terminal domains with TGF-bs themselves. They are synthesised as large precursors and the mature protein, derived from the carboxy-terminal region, is released after proteolytic cleavage. These 30-38 kDa dimeric proteins are divided into subfamilies according to their amino acid sequence similarities [9,10,14,15]. The BMP family members are multifunctional morphogens that control the development and apoptosis of a variety of cell types including osteoblasts, chondroblasts, neural cells, and epithelial cells.

Tissue Engineering

     Tissue engineering is defined as the science of fabrication of new tissues for replacement and regeneration, [16] and is based on principles of developmental and molecular biology, signal transduction and cell biology, including the supramolecular assembly of the extracellular matrix, or ECM. Morphogenesis, a complex cascade of determination of cell lineages by induction, pattern formation and progressive cell differentiation, is initiated during the process of embryogenesis and culminates in the achievement of the adult form [17]. The three most important  ingredients, that come into play in tissue engineering, are the regulatory morphogenetic signals, the responding cells and the ECM [10,13,18].  In the formation of endochondral bone, cartilage is first laid down and forms the patterning anlage for the subsequent differentiation and deposition of bone. The ECM of cartilage consists of a constellation of macromolecules such as collagens, proteoglycans, and glycoproteins and  is to these ECM components that morphogens bind to assemble a morphogenetic scaffold [8,9,10,13,15].

     In vertebrate development, the formation of bone is a continual process, initiated during fetal development and continued in postnatal tissues in the processes of remodelling and repair [9,10,13,19]. The remarkable capacity of skeletal tissues to regenerate has led to the theory that the molecular signalling pathways regulating skeletogenesis are reiterated after fetal development in the process of adult wound healing. Individual BMPs are markedly expressed at many sites in the developing embryo and are likely to be key regulators of early development and organogenesis  [9,10,15].  Their role, in inducing the formation of bone, cartilage and connective tissues associated with the skeleton, is well established. The study of fetal skeletogenesis is important in order to gain insights into fracture repair processes and advances in this field will aid in the development of therapeutic agents for the creation of new bone useful in the treatment of skeletal injuries and metabolic bone diseases and in oral and maxillofacial applications.

     The analysis of embryonic induction has already pointed to the importance of the antagonistic roles played by secreted inducing morphogenetic factors and their soluble inhibitory binding proteins.  These have been particularly well characterised in patterning the primary axes of insects and vertebrates,  similar antagonistic relationships also playing a role in a number of  later events of embryogenesis. Important progress has been made from recent studies on the patterning of embryonic ectoderm, most of which have been done in Xenopus. Gradual progress is being made in unravelling the molecular basis of the cell fate decisions occuring during vertebrate gastrulation, cells of the embryonic ectoderm giving rise to epidermal progenitors ventrally and neural progenitors dorsally. Epidermal  induction  appears to require instructive positive signals while the establishment of neural fate has been found to occur by default by inhibition of epidermal inducers, [22] challenging previous thinking in this field. Substantial progress has also been made by the finding in amphibians that BMP-4 acts as a neural inhibitor and epidermal inducer and that endogenous antagonists of BMPs are secreted by the organiser, earlier upstream events also being required before BMP inhibition stabilises neural fates [23].

     The secreted  inductive proteins sonic and Indian hedgehog are expressed in early embryogenesis and exert their effects on pattern formation and chondrogenesis in the appendicular skeleton partly through their regulation of BMPs and parathyroid hormone-related peptide (PTHrP). The transcription factor Cbfa1 has also been found to play a critical role in the process of chondrocyte differentiation and ossification i.e., in their elimination by apoptosis and replacement by bone after chondrocyte  maturation and hypertrophy. From studies on the expression patterns of these genes during fracture healing they are proposed to play an important and equivalent role in adult wound repair [24].

BMP Signalling-Molecular Studies

     Understanding of  the signalling of BMPs as ligands has been gained from the molecular cloning and characterisation of many TGF-b superfamily member receptors, including those of the BMPs (reviewed by ten Dijke et al., 1996 [24]). BMPs, and most of the members of the TGF-b superfamily act as ligands via the heteromultimerisation of two kinds of serine/threonine kinase receptors: type 1, of  50-55 kDa, and type II, of more than 75 kDa [26]. Advances have been made in identifying the intracellular signalling molecules of the TGF-b signal transduction pathway and their regulators. Molecular studies into the signalling of the decapentaplegic (Dpp) morphogen, a BMP homologue in Drosophila, have led to the discovery of Smad proteins as central mediators of signal transduction by TGF-b family members. These consist of the so-called pathway-restricted, common-mediator and inhibitory Smads [26, 27,28]. Work in mammalian cell culture has identified a biochemical model of the signal transduction mechanism in which the activation of receptor serine-threonine kinase activity leads to phosphorylation of specific Smad proteins and translocation of heteromeric Smad protein complexes to the nucleus. Once in the nucleus Smad proteins interact with other DNA-binding proteins to regulate transcription of specific target genes. The recent findings in Drosophila and vertebrates suggest that the BMP signalling can be modulated extracellularly and intracellularly by the availability of BMP inhibitors and Smads respectively [29]. Two BMP type I receptors and a BMP type II receptor have been identified in mammals. The BMP receptors types I and II bind BMPs and act in concert to transduce the phosphorylation of Smad 1 and 5. These then translocate to the nucleus complexed with Smad 4, where they activate transcription of various genes to initiate BMP responses including fracture healing [30]. More specifically, in osteogenesis, it has been found that Smads 1 and 5 mediate the intracellular signalling of BMP, whereas Smads 2 and 3 transduce TGF-b signalling, Smad 4 being the common mediator required for both pathways. Both Smads 6 and 7 inhibit signalling by members of the TGF-b superfamily, acting in autoregulatory feedback loops to reduce the duration and/or intensity of the signal. They have been found to be strongly expressed in cartilage containing mature chondrocytes, pointing along with other evidence, to the morphogenic role played by Smads during endochondral bone formation [31].  Activities of BMPs are extracellularly regulated by BMP-binding proteins such as noggin and chordin, and these, and  additional regulators of BMPs identified thus far, including cerberus, dan and gremlin may be harnessed as therapies to offset calcification, such as that encountered after total hip arthroplasties [32].

Periodontal Regeneration

     The periodontium is comprised of the gingiva, cementum, alveolar bone and periodontal ligament. The initiating event in periodontal regeneration  is transitory and leads to a number of cellular events which in turn stimulate a number of subsequent events such as chemotaxis, proliferation, differentiation and/or angiogenesis, leading to the further progression of  tissue formation. Periodontal regeneration is defined as rapid colonisation and ECM synthesis by cementoblasts and their precursors along the exposed root surface, followed by the development and insertion of Sharpey’s fibres into the newly-forming cementum, Fig. (2). BMPs initiate both cementum and periodontal ligament  regeneration [33]. Naturally-sourced BMPs and particularly recombinant human osteogenic protein-1 (hOP-1 or BMP-7) appear particularly effective in initiating cementogenesis.

     By preparing semi-thin undecalcified sections of the periodontal, dental and bone tissues, foci of nascent mineralisation could be seen within newly-formed cementoid, the as yet to be mineralised collagenic material, Fig. (2) [33,34]. It has been shown in surgically created furcation defects that OP-1 (BMP-7) in association with a collagenous matrix as carrier induces morphogenesis of periodontal ligament, cementogenesis and insertion of Sharpey’s fibres into cementum [8,33,34,35].

     Classic receptor-mediated peptides or ECM molecules for hard and soft tissues appear  permissive to trigger tissue formation cascades. rhOP-1 (BMP-7), rhBMP-2 and rhBMP-4 singly applied are equally capable of reparative dentinogenesis as well as bone induction [35], raising the question of possible reasons for this apparent redundancy [34,35,36]. Furthermore,  limited promiscuity among receptors in binding BMP-4 and OP-1 (BMP-7) also reflects BMPs’ different in vivo functions [24].

     An important question is whether the presence of multiple forms of BMPs has a therapeutic significance. Our experiments indicate that hOP-1 (BMP-7), at the dose of 100 and 500 mg per gram of matrix as carrier preferentially induced cementogenesis [34]. Future research should focus on molecular combinations, developing a structure- activity profile amongst the BMP family members [35].

BMP Proteins and their Receptors: Potential Functions in the Brain

     The superfamily of TGF-b-related genes include over 25 members in mammals, several of which are expressed in the growing nervous system and serve important functions in regionalising the early CNS. Recent results have revealed that TGF-bs, activins, and BMPs selectively signal to the responding cells via different hetero-oligomeric complexes of type I and type II serine/threonine kinase receptors. From immunolocalisation studies in the mouse [37], BMP-2 and OP-1 have been found in spiral ligament and interdentate cells of the cochlea, while BMP-3 was restricted to the spiral ganglion. BMP-3 was also found  in the Purkinje cells of the cerebellum, nerve fibres of the cerebellum and hemispheres, the afferent cells of the dorsal root ganglia, inferior alveolar nerve and peripheral processes of the vestibulocochlear nerve amongst many other cell types of non-neuronal origin [37].

     It  is known that the mature brain expresses high levels of the BMP receptor, BMPR-II, and two activin receptors, ActR-1 and ActR-II. This  indicates that OP-1/BMP-7, BMP-2 and BMP-4 as well as activins may serve important functions for brain neurons, which are proposed to use the receptors BMPR-II and ActR-I to sense the presence of BMPs [38]. Preliminary in situ data furthermore reveal the BMP relatives GDF-1 and GDF–10 are differentially expressed at high levels in neurons expressing these receptors. From these studies, BMPs and their signalling systems are proposed to constitute a  novel pathway for control of neural activity and may offer means for pharmacological interventions rescuing brain neurons [38].

     BMP ligands and receptor subunits are present within discrete regions of the embryonic brain throughout neural development and within neural crest-derived migratory zones. During gastrulation, BMPs initially inhibit the formation of neuroectoderm while within the neural tube, they act in gradients in promoting the differentiation of dorsal and intermediate cell types through interactive signalling. In the peripheral nervous system, BMPs act as instructive signals for the commitment of neuronal lineages and the promotion of graded stages of neuronal differentiation. In addition, BMPs act on more lineage-restricted embryonic central nervous system progenitor cells to promote regional neuronal survival and cellular differentiation. The pluripotency of these cytokines is further shown by their ability to selectively induce apoptosis of certain neural crest-associated cell populations. These observations reflect the ability of BMPs to promote a broad range of context-specific effects during multiple stages of neural development [39].

Bone Inductive Activity of TGF-b Proteins

     The de novo induction of bone formation, originally and solely ascribed to the BMP family members, has now been extended to other TGF-b, family members, the decapentaplegic (Dpp) and 60A gene products expressed early in Drosophila development [6] and  to the growth and differentiation factor-5 (GDF-5) [7]. More recent experiments have additionally indicated a direct induction of osteogenic differentiation by Hedgehog proteins, interestingly without affecting the expression level of BMPs [42,43].

     The presence of multiple molecular forms of morphogens with bone inductive activity, besides pointing to a possible redundancy in osteogenic differentiation, indicates on the other hand the evolutionary conservation of related proteins from phylogenetically distinct species, and moreover points to synergistic interactions during endochondral bone formation [1,2,13]. Indeed in recent experiments in the primate, we have reported the de novo induction of endochondral bone formation by recombinant human (h) and native platelet-derived TGF-b1 in intramuscular sites of the baboon (Papio ursinus) [1,2] and also the heterotopic osteoinductivity of hTGF-b2, Fig. (1A) [44]. Moreover, we have shown a potent synergistic induction of endochondral bone formation with the binary application of TGF-b1 and a recombinant hBMP (osteogenic protein-1, OP-1, also known as BMP-7) both in heterotopic,  Fig. (1B) and orthotopic sites of the baboon [1,2]. Significantly, however, both hTGF-b1 and -b2 had rather limited bone inductive activity in orthotopic sites, [34,44] raising the concept of site and tissue specificity in bone induction by TGF-b proteins, Fig. (1C) [1,2,44].

     The bone inductive activity of TGF-b isoforms is noteworthy and importantly and significantly the induction of endochondral bone results only upon implantation in heterotopic sites of primates [1,2, 44], since the subcutaneous implantation of TGF-b proteins in rodents induces only granulation tissue with potentially marked fibrosis [4,5]. The site and tissue specific inductivity of TGF-b proteins in the primate requires a full understanding of the mechanistic insights regulating such specificity, since eventually the TGF-b proteins could routinely be used in orthotopic sites of human patients. Possibilities that deserve investigation are the identification of possibly disparate responding cells in different anatomical locations, the presence of inhibitory binding proteins and the regulation mediated by antagonists of TGF-b signalling, the down-stream effectors, Smads 6 and 7 [27,45,46].

Biomaterials with Intrinsic Osteoinductive Activity

     A very important issue not only for the tissue engineering of bone but also for morphogenesis and tissue regeneration in general is the development and use of biomimetic biomaterials that will optimally deliver the biological activity of BMPs and related members. This is a very important concept since we have now learned that the optimal induction of bone formation is dependent on the combined action of morphogens such as BMPs and TGF-bs, the soluble signals, and a complementary substratum, the insoluble signal [10,12,13,18,47]. From a thera-peutic perspective, a carrier substratum is required for local delivery of recombinant hBMPs to evoke a desired osteogenic response, since it is the composite of a biomaterial carrier together with hBMPs that triggers the bone induction cascade [12,10]. In previous studies, we have reported that porous biomaterials of coral-derived hydroxyapatite acted as ideal delivery systems for BMPs, providing evidence that the osteogenic activity of BMPs could be restored and delivered by an  insoluble substratum other than the commonly used insoluble collagenous bone matrix [48,49,50,51,52,53]. Further experiments were conducted  in our laboratories to create and study biomaterials that in their own right, when implanted heterotopically  in  recipient animals, can induce specific morphogenetic responses from the host tissues without the addition of exogenously applied BMPs or TGF-bs, i.e. biomaterials with intrinsic osteoinductive activity [18,54,55,56,57]. We have reported that the surface geometry of our biomimetic carriers is of critical importance for the expression of the osteogenic phenotype and the morphogenesis of bone, Fig. (3) [55,57]. The incorporation of specific biological activities into biomaterials, achieved by manipulating the geometry and the surface characteristics of the substratum, defined as the geometric induction of bone formation [55,57] will result in predictable bone morphogenesis and growth for the treatment of human bone defects, Fig. (3). An effective carrier/delivery system is still a very important challenge to be met. Essential requirements for an osteogenic delivery system is the initiation of optimal osteoinductivity with relatively low doses of recombinant BMPs along with favourable  surface characteristics and the geometry of the substratum, the insoluble signal [57].

     The experimental studies performed in primates and the ongoing clinical trials of recombinant hBMPs are assisting dramatically our understanding of the biological and therapeutic significance of hBMPs in clinical contexts, since ultimately these morphogens will be an essential armamentarium for multiple tissue and organ engineering and regeneration. So much so that after several years of experimentation in primates, under the aegis of the University of the Witwatersrand, Johannesburg, our Bone Research Unit initiated and implemented in 1996-97 and 1998 a clinical trial in humans using naturally-sourced BMPs delivered by human demineralized bone matrix in craniofacial defects in man [58,59].  The accrued information will help to design novel therapeutic strategies based on information on gene regulation by soluble signals and particularly the substratum surface characteristics and geometry.  Breakthroughs in tissue engineering of bone and related tissues may be expected, if the concepts outlined in this review, are developed both biologically and clinically.

Acknowledgments

     Our work is supported by the South African Medical Research Council, the University of the Witwatersrand, Johannesburg and the Foundation for Research and Development and also by an ad hoc grant from the Dentistry Development of South Africa to several post-graduate students. For their invaluable help in the past years we wish to thank Barbara van den Heever, Jean Crooks and Drs. Ilario dal Mas and Carlo Ferretti.

References

[1]     Ripamonti, U.; Duneas, N.; van den Heever, B.; Bosch, C. and Crooks, J. (1997) J. Bone Miner. Res., 12, 1584-1595.

[2]     Duneas, D.; Crooks, J. and Ripamonti, U. (1998) Growth Factors, 15, 259-277.

[3]     Ripamonti, U.; Crooks, J.; Matsaba, T., and Tasker, J.  (1999) Growth Factors, (in press).

[4]     Roberts, A.B.; Sporn, M.B.; Assoian, R.K.; Smith, J.M.; Roche, N.S.; Wakefield, L.M.; Heine, U.I.; Liotta, L.A.;  Falanga, V.; Kehrl J.H. and Fauci, A.S. (1986) Proc. Natl. Acad. Sci. USA. 83, 4167-4171.

[5]     Shinozaki, M.; Kawara, S.; Hayashi, N.; Kakinuma, T.; Igarashi, A. and Takehara, K. (1997) Biochem. Biophys. Res. Commun., 237, 292-296.

[6]     Sampath, T.K.; Rashka, K.E.; Doctor, J.S.; Tucker, R.F. and Hoffmann, F.M. (1993)  Proc. Natl. Acad. Sci., USA. 90, 6004‑6008.

[7]     Hotten, G.C.; Matsumoto, T.; Kimura, M.; Bechtold, R.F.; Kron, R.; Ohara, T.; Tanaka, H.;  Satoh, Y.; Okazaki, M.; Shirai, T.;  Pan, H.; Kawai, S.; Pohl, J.S. and Kudo, Growth Factors, 13, 65-74.

[8]     Reddi, A.H. (1994) Curr. Opin. Genet. Dev. 4, 737-744.

[9]     Reddi, A.H. (1997) Cytokine Growth Factor Rev. 8, 11-20.

[10] Reddi, A.H. (1998) Nature Biotechnology 16, 247-252.

[11] Ripamonti, U. and Vukicevic, S. (1995) S. Afr. J. Sci., 91, 277-279.

[12] Ripamonti, U. and Reddi, A.H. (1995) Adv. Plast. Reconstr. Surg., 11,47-73.

[13] Ripamonti, U. and Duneas, N. (1998) Plast. Reconstr. Surg., 101, 227-239.

[14] Wozney, J.M.; Rosen, V.; Celeste, A.J.; Mitsock, L.M.; Whitters, M.J.; Kriz, R.W.; Hewick, R.M. and Wang, E.A. (1988) Science (Wash. DC). 242, 1528‑1534.

[15] Wozney, J.M. (1992) Molec. Reprod. Dev. 32, 160-167.

[16] Reddi, A. H. (1994) J. Cell Biochem., 56, 192-195.

[17] Reddi, A.H.(1984) Extracellular matrix and development in Extracellular matrix biochemistry, (Piez K.A. and Reddi, A. H., ed.), Elsevier, New York, pp. 375-412.

[18] Ripamonti, U. and Duneas, N. (1996) Mat. Res. Soc. Bull., 21, 36-39.

[19] Ferguson, C.M.; Miclau, T.; Hu, D.; Alpern, E., and Helms, J.A. (1998) Ann. N.Y.Acad. Sci. Oct. 23, 857, 33-42.

[20] Wozney, J.M.; Rosen, V.; Celeste, A.J.; Mitsock, L.M.; Whitters, M.J.; Kriz, R.W.; Hewick, R.M. and Wang, E.A. (1988) Science (Wash. DC). 242, 1528‑1534.

[21] Wozney, J.M. (1992) Molec. Reprod. Dev. 32, 160-167.

[22] Chang, C.and Hemmati-Brivanlou, A.(1998) J. Neurobiol. 36(2), 128-151.

[23] Streit, A. and Stern, C. D. (1999) Trends Genet. 15(1), 20-24.

[24] ten Dijke, P.; Miyazono, K. and Heldin, C.-H. (1996) Curr. Opin. Cell Biol. 8, 139-145.

[25] Yamashita, H.; ten Dijke P.; Heldin, C.-H. and Miyazono, K (1996) Bone 19(6), 569-574.

[26] Heldin, C. H.; Miyazono, K. and ten Dijke, P. (1997) Nature 390, 465-471.

[27] Whitman, M. (1997) Nature 389, 549-551.

[28] Massague, J. (1998) Annu. Rev. Cell Biol. 6, 597-641.

[29] Cho, K.W. and Blitz I.L. (1998) Curr. Opin. Genet. Dev. 8(4), 443-449.

[30] Kawabata, M.; Imamura, T. and Miyazono, K. (1998) Cytokine Growth Factor Rev.  9(1), 49-61.

[31] Sakou, T; Onishi, T.; Yamamoto, T.; Nagamine T.; Sampath, T.K. and ten Dijke, P. (1999) J. Bone Min. Res. 14(7), 1145- 1152.

[32] Schmitt, J. M.; Hwang, K.; Winn, S. R. and Hollinger, J. O. (1999) J. Orthop. Res. 17(2), 269-278.

[33] Ripamonti, U.; Heliotis, M.; van den Heever, B. and Reddi, A. H. (1994) J. Period. Res. 29, 439- 445.

[34] Ripamonti, U. (1996) in Bone Morphogenetic Proteins: Biology, Biochemistry and Reconstructive Surgery, (Lindholm, T.S., Ed.), Academic Press, Austin, Texas.

[35] Ripamonti, U. and Reddi, A. H. (1997) Crit. Rev. Oral Biol. Med. 8(2), 154-163

[36] Ripamonti, U. and Reddi, A. H. (1994) J. Period. Res. 29, 225- 235.

[37] Thomadakis, G.; Ramoshebi, L.N.; Crooks, J.; Rueger, D. C. and Ripamonti, U. (1999) Eur. J. Oral Sciences, 107, 368-377.

[38] Ebendal, T.; Bengtsson, H. and Soderstrom, S. (1998) J. Neurosci. Res. Jan 15, 51(2), 139-146.

[39] Mehler, M.F.; Mabie, P.C.; Zhang, D. and Kessler, J.A. (1997) Trends Neurosci. 20(7), 309-317.

[40] Sampath, T.K. and Reddi, A.H. (1984) J. Cell Biol., 98, 2192-2197.

[41] Reddi, A.H. (1974). Adv. Biol. Med. Phys. 15, 1-18.

[42] Kinto, N.; Iwamoto, M.; Enomoto-Iwamoto, E.; Noji, S.; Ohuchi, H.; Yoshioka, H.; Kataoka, H.; Wada, Y.; Yuhao, G.; Takahashi, H.E.; Yoshiki, S. and Yamaguchi, A. (1997) Febs. Letters, 404, 319-323.

[43] Takashi, N.; Aikawa, T.; Iwamoto-Enomoto, M.; Iwamoto, M.; Higuchi, Y.; Pacifici, M.; Kinto, N.; Yamaguchi, A.; Noji, S.; Kurisu, K. and Matsuya, T. (1997) Biochem. Biophys. Res. Commun., 237, 465- 469.

[44] Ripamonti, U.; Crooks, J.; Matsaba, T., and Tasker, J.  (1999) Growth Factors, (in press).

[45] Imamura, T.; Takase, M.; Nishihara, A.; Oeda, E.; Hanai, J.; Kawabata, M. and Miyazono, K. (1997) Nature, 389, 622-626.

[46] Nakao, A..; Afrakhte, M.; Moren, A.; Nakayama, T.; Christian, J. L.; Heuchel, R.; Itoh, S.; Kawabata, M.; Heldin, N.; Heldin, C. and ten Dijke, P. (1997) Nature, 389, 631-635.

[47] Ripamonti, U. (1999) South African Dental Journal 54, 257-262.

[48] Ripamonti, U. (1991) J. Bone Joint Surg. 73-A, 692 -703.

[49] Ripamonti, U.; Ma, S.; van den Heever, B.; and Reddi, A. H. (1992) Plast. Reconstr. Surg., 90, 382-393.

[50] Ripamonti, U.; Ma, S. and Reddi, A. H. (1992)