HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (11) Citing Articles via Google Scholar Google Scholar Articles by Chen, H. Articles by Clarkson, B.H. Search for Related Content PubMed PubMed Citation Articles by Chen, H. Articles by Clarkson, B.H.

Interaction of Dendrimers (Artificial Proteins) with Biological Hydroxyapatite Crystals

¹ University of Michigan School of Dentistry, 1011 N. University, Ann Arbor, MI 48109-1078;
² University of Michigan Department of Chemistry;
³ University of Michigan Department of Physics;
⁴ University of Michigan, Internal Medicine; and
⁵ Center for Biologic Nanotechnology;

*corresponding author, bricla{at}umich.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
This investigation sets out to mimic protein-crystal interaction during biomineralization with the use of artificial proteins (dendrimers). It is hypothesized that these interactions depend on the surface charge of hydroxyapatite crystals. This was investigated with the use of dendrimers with capped surfaces of different charges to probe the surface. We used AFM images of crystal-bound dendrimers to determine the distribution of the surface charge, and its magnitude was correlated to the binding capacity of the dendrimers to the surface. The binding capacity of the dendrimers in ascending order at pH 7.4 was: acetamide-capped, -NHC(O)CH3, neutral charge; carboxylic-acid-capped, -COOH, negative charge; and amine-capped, -NH2, positive charge. AFM images of the crystals showed dendrimers spaced equally along the crystal. The results suggest that the crystal surface has alternating bands of positive and negative charge or a differential charge array, i.e., alternating bands of either more or less positive or negative charge.

KEY WORDS: dendrimers • interaction • hydroxyapatite • crystals

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Protein-mineral binding is likely to play a central role in the biomineralization process by modulating both mineral nucleation and crystal growth. Wierzbicki et al. (1994), using atomic force microscopy (AFM), described oyster shell protein as having a globular appearance on calcite crystal surfaces, whereas poly (Asp) was extended on the surface. More recently, the conformation of the non-collagenous proteins of both enamel and dentin when adsorbed to hydroxyapatite crystals has been investigated. These include observations in the AFM (Wallwork et al., 2002), which would suggest that when the dentin matrix protein phosphophoryn (PP) interacts with hydroxyapatite crystals, it has a globular form, and that the nanospheres are similar in size to those reported in previous AFM studies of bovine serum albumin (BSA) monomers (Mori and Imae, 1997;Wallwork et al., 2001) and the amelogenin aggregates (Kirkham et al., 2000). The Wallwork et al. (2002) study also showed that the distribution of dentin sialoprotein (DSP) and especially PP on the crystal surface gave it a distinct "banded" appearance, and was similar to that described for amelogenin (Kirkham et al., 2000) and for BSA (Wallwork et al., 2001). These bands seemed to be of a periodicity similar to that of the charge arrays found by chemical force microscopy.

Binding of enamel proteins and the non-collagenous proteins of dentin to hydroxyapatite crystal surfaces may depend on their structural conformation. This complicates the interpretation of experiments, because the size and geometry of the active functional groups are not yet fully determined. The use of a macromolecule with a well-defined and controllable structure would allow the functional activity to be measured for a variety of known configurations. It is this potential we sought by synthesizing dendrimers to mimic those enamel and non-collagenous dentin proteins that may modulate crystal nucleation and growth.

One of the best candidates for such a nanomaterial is the general class of dendritic polymers or dendrimers. These polymers are synthesized as defined spherical structures ranging from 1 to 20 nanometers in diameter and have been called "artificial proteins". Several generations of polyamidoamine (ß-alanine subunit) dendrimers are depicted in Fig. 1a. Molecular weight and the number of terminal groups increase exponentially as a function of generation (the number of layers) of the polymer. Different types of dendrimers can be synthesized based on the core structure that initiates the polymerization process. These core structures dictate several characteristics of the molecule, such as the overall shape, density, and surface functionality (Tomalia et al., 1990).

View larger version (74K): [in this window] [in a new window]   Figure 1. Structure and size comparison of polyamidoamine dendrimer. (a) Polyamidoamine (PAMAM) dendrimers, also known as "artificial proteins", with a well-defined structure, can act as a nanoprobe to probe specific surface domains on the surfaces of enamel crystals. Three types of PAMAM dendrimers, with amine-, carboxylic-acid-, and acetamide-capped surfaces, respectively, were used in this study. Tapping-mode AFM images of enamel crystals (control). (b) Enamel crystals on a mica surface, imaged in air prior to the introduction of dendrimers. (c) Enamel crystals on a mica surface, imaged in air after the sample (b) was rinsed with distilled water (pH 7.4) and dried in a desiccator. AFM images show the topographical image (1 µm x 1 µm) on the left, and phase image (1 µm x 1 µm) on the right.

	MATERIALS & METHODS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Enamel Crystals
Individual crystals from maturation-stage enamel were obtained from the mandibular incisors of four-week-old male Sprague Dawley rats (Robinson et al., 1974;Hiller et al., 1975). (The animal use protocol was reviewed and approved by the University’s committee on Use and Care of Animals).

All detectable traces of matrix protein were removed from the enamel samples by a sequential extraction procedure described by Robinson et al. (1995). Briefly, enamel particles were first extracted with 0.1 M phosphate buffer, pH 7.4, to desorb mineral-bound proteins and components dissolved in enamel fluid. After centrifugation, ww re-extracted the insoluble pellet by re-suspending it in fresh phosphate buffer. This was repeated a total of 6 times. The pelleted material was then further extracted with the use of 50 mM Tris containing 4 M urea at pH 7.4 to dissolve aggregated precipitated protein. The insoluble residue was then re-extracted for a further 6 times with 0.1 M phosphate buffer, pH 7.4, to ensure final desorption and dissolution of any mineral-bound components. The final residue was washed with distilled water with the pH adjusted to 7.4 so that all traces of buffer and urea would be removed. The crystals were then treated with 3% hypochlorite for oxidization of traces of the organic materials and washed with distilled water. The crystals were finally dispersed in HPLC-grade methanol by sonication.

Dendrimer-Crystal Binding
The enamel crystals were sonicated for 2 min in HPLC-grade methanol to reduce aggregation. Two 5-µL quantities of this suspension were then pipetted onto freshly cleaved mica. The methanol was evaporated rapidly, leaving a coating of dispersed hydroxyapatite crystals. The dendrimers were suspended in water, at pH 7.4, and then pipetted onto the crystals. A 120-second exposure period was chosen because binding of the nascent dentin proteins to the crystal surfaces appeared to be almost instantaneous (Wallwork et al., 2002). The fluid was wicked off, and the specimen was placed in a desiccator for 12 hrs and subsequently imaged by AFM.

Force Microscopy
All samples were imaged in tapping mode in air, by means of a Nanoscope IIIa Multimode AFM and controller (Digital Instruments, Santa Barbara, CA, USA) equipped with a 120 µm x 120 µm J-type scanner. Commercially available tapping mode cantilevers, TESP, were used (Digital Instruments).

Desorption of Dendrimers from Crystal Surfaces
Desorption of dendrimers from the crystal surfaces was achieved with the use of 100 and 200 mM phosphate buffer at pH 7.4. The binding capacity of the protein was directly correlated with the molarity of the phosphate solution.

Polyamidoamine (PAMAM) Dendrimers
Generation 7-PAMAM dendrimers capped with amine (-NH₂), carboxylic acid (-COOH), and acetamide (-NHC(O)CH₃) groups were synthesized in the laboratory of the Center of Biologic Nanotechnology at the University of Michigan.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Figs. 1b and 1c show a tapping-mode AFM image of the enamel crystals prior to (Fig. 1b) and after being rinsed with distilled water pH 7.4 (control) (Fig. 1c). The crystal surfaces are clean and free of protein deposits. The surface morphology seen in Fig. 1b is the innate surface roughness of the crystal. The acetamide (-NHC(O)CH₃)-capped G7-PAMAM dendrimer solution was allowed to flow over the crystals, allowed to remain for 2 min, and wicked off. Fig. 2a shows the crystals and mica surfaces very sparsely covered with dendrimers deposited after the addition of a 1 µL drop of 7-nM dendrimer solution to the crystals. A much greater coverage of dendrimers on all surfaces was obtained by the addition of 5 µL of the same solution (Fig. 2b). After the crystals were washed with distilled water, all the dendrimers were removed (Fig. 2c). In Figs. 3a, 3b, and 3c, hydroxyapatite crystals are shown coated with carboxylic acid (-COOH)-capped dendrimers deposited after the addition of 1 µL, 1.5 µL, and 2 µL of dendrimer-containing solutions of 570 nM, 285 nM, and 570 nM concentrations, respectively. Their appearance is similar to that seen for the acetamide-capped dendrimers, although, in Fig. 3a, the dendrimers are spaced at regular intervals at the crystal interfaces, and in Fig. 3b they are spaced regularly along the surface of a crystal. Washing the sample imaged in Fig. 4c with distilled water (pH 7.4) did not remove the dendrimers (Fig. 3d). However, rinsing with 100 mM phosphate buffer (pH 7.4) did remove the dendrimers from the crystal surfaces (Fig. 3e). Treatment of the crystals with 1-µL (Fig. 4a) and 2-µL (Fig. 4b) quantities of a 78-nM solution of amine-capped dendrimers resulted in the crystal surfaces being coated with nanospheres. This appearance was similar to that seen for the acetamide- and carboxylic-acid-capped dendrimers. Rinsing with distilled water (image not shown) and then phosphate solutions, at either 100-mM or 200-mM concentration, did not remove the dendrimers (Figs. 4c and 4d, respectively). Rinsing the crystals after the 100-mM phosphate treatment with 4 M urea, the same concentration of urea used to strip proteins from the enamel crystals when they were prepared for experimentation, did remove the dendrimers. The results are the same as that shown for clean crystals in Figs. 1b and 1c.

View larger version (68K): [in this window] [in a new window]   Figure 2. Tapping-mode AFM images of enamel crystals after interaction with acetamide-capped PAMAM dendrimers. A solution containing G7 PAMAM dendrimers with an acetamide (-NHC(O)CH3)-capped surface was allowed to flow over the crystals and allowed to remain for 2 min before being wicked off. The sample was dried in a desiccator before being imaged in air. (a) After exposure to 1 µL 7-nM dendrimer solution. (b) After exposure to 5 µL 7-nM dendrimer solution. (c) After sample (b) was rinsed with distilled water (pH 7.4). AFM images show the topographical image (1 µm x 1 µm) on the left, and phase image (1 µm x 1 µm) on the right.

View larger version (109K): [in this window] [in a new window]   Figure 3. Tapping-mode AFM images of enamel crystals after interaction with carboxylic-acid-capped PAMAM dendrimers. A solution containing G7 PAMAM dendrimers with carboxylic-acid (-COOH)-capped surface was allowed to flow over the crystals and allowed to remain for 2 min before being wicked off. The sample was dried in a desiccator before being imaged in air. (a) Crystal surfaces after exposure to 1 µL 570-nM dendrimer solution (arrows denote dendrimers at crystal interface). (b) Crystal surfaces after exposure to 1.5 µL 285-nM dendrimer solution (arrows denote dendrimers on crystal surface). (c) Crystal surfaces after exposure to 2 µL 570-nM dendrimer solution. (d) After sample (c) was rinsed with distilled water (pH 7.4). (e) After sample (d) was rinsed with 100 mM phosphate buffer (pH 7.4). AFM images show the topographical image (1 µm x 1 µm) on the left, and phase image (1 µm x 1 µm) on the right.

View larger version (92K): [in this window] [in a new window]   Figure 4. Tapping-mode AFM images of enamel crystals after interaction with amine-capped PAMAM dendrimers. A solution of G7 PAMAM dendrimers with an amine (-NH2)-capped surface was allowed to flow over the crystals and allowed to remain for 2 min before being wicked off. The sample was dried in a desiccator before being imaged in air. (a) Crystal surfaces after exposure to 1 µL 78-nM dendrimer solution. (b) Crystal surfaces after exposure to 2 µL 78-nM dendrimer solution. (c) After the distilled-water-washed sample was rinsed with 100 mM phosphate buffer (pH 7.4). (d) After the sample (c) was rinsed with 200 mM phosphate buffer (pH 7.4). AFM images show the topographical image (1 µm x 1 µm) on the left, and phase image (1 µm x 1 µm) on the right.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

The apparent stronger binding capacity of the positively charged dendrimer (-NH₂-capped) compared with the negatively charged one (-COOH-capped) was unexpected. Previous results with the highly negatively charged human phosphophoryn molecule had shown that this protein bound at regular intervals on the crystals and with a binding capacity that resisted a 200-mM phosphate rinse (Wallwork et al., 2002). Analysis of these data suggests that the crystal surface had regularly spaced positively charged domains, perhaps calcium-rich. The binding of the (-COOH)-capped dendrimer would also suggest this, with the difference in the strength of the binding between the dendrimer and the nascent protein to the crystals being dependent on charge differential between the two molecules. The binding of the (-NH₂)-capped dendrimers indicated negatively charged regions (e.g., phosphate-rich) also being present on the crystal surface. This would explain why, during the CFM study reported by Kirkham et al. (2000), there was a preferred interaction for the amine-capped tip in the images. The overall results of this present study and those reported by Wallwork et al. (2002), using the highly electronegative protein PP, lend credence to the work of Chander and Fuerstenau (1984), who measured the zeta potentials of hydroxyapatite surfaces and claimed that they had both negative and positive charges. However, others have reported that the surfaces of biological hydroxyapatite crystals are positively charged, possibly reflecting a calcium-rich layer (Mafe, 1996;Zhang et al., 1999). The difference in binding capacity between the amine- and carboxylic-acid-capped dendrimers could also be explained by differing degrees of ionization. Another possible explanation involves the opposite charges present on the polymers. The negatively charged carboxylic-terminated polymers presumably bind to positively charged regions and thus would be most likely removed by phosphate buffer. The negatively charged phosphate would not be expected to compete effectively for the negatively charged surface sites to which the positively charged amine-terminated polymers would be expected to bind. The stronger binding of the amine-capped dendrimers may also indicate a different binding mechanism. Analysis of recent data has suggested that the adsorbed amino acid on the phosphoserine was located within the inner Helmholtz plane of the electrical double layer at the HAP/electrolyte interface, while the phosphate and carboxyl groups on the proteins were oriented away from the surface, due to electrical repulsion (Spanos et al., 2001).

According to the phase image, many of the crystal surfaces appeared to have an innate repeating nanoscale structure of approximately 20 nm (Fig. 1b). A further surface structure was suggested after the crystals were exposed to the low and medium concentrations of carboxylic acid dendrimers, which resulted in AFM images showing crystals with dendrimers spaced at approximately 48 nm, although there was a large distribution (from 20 to 80 nm) (Fig. 3b). These results are complementary to the results of Kirkham et al. (2000), who showed, using chemical force microscopy, alternating domains of surface charge comprised of broad bands, some 30 to 50 nm wide. Our results and those of earlier studies strongly suggest that the charge on both the crystal surface and of the protein play a significant role in the binding of the proteins to the crystal surface.

All the enamel proteins have post-translational modifications, and all are phosphorylated. It is also recognized that both the amelogenins and the enamelins are adsorbed onto hydroxyapatite surfaces (Doi et al., 1984;Bai and Warshawsky, 1985;Aoba et al., 1987;Aoba and Moreno, 1991;Fejerskov et al., 1994;Kirkham et al., 2000; Wallwork et al., 2001). Phosphorylation of these proteins and dentin matrix non-collagenous proteins may be an important factor in the binding of these proteins to enamel crystals. Wallwork et al. (2002) showed reduced binding capacity of PP after dephosphorylation. Dentin sialoprotein, which is not as highly phosphorylated as PP, also showed less binding affinity to the crystals than PP. However, relying solely on charge for the binding of nascent proteins to crystal surfaces may be too simplistic an explanation.

The results of this present work show the potential of these artificial proteins, dendrimers, to be used as nanoprobes to mimic the pattern and determine the strength of binding of proteins to crystal surfaces. As importantly, this study demonstrates that by changing the dendrimer’s functional groups, one alters their binding capacity to the crystal surface, thus perhaps influencing the dendrimers’ ability to initiate or control crystal growth. Synthetic scaffolds (self-assembled peptide-amphiphile molecules) have already been used to form a cross-linked mesh for the growth of oriented hydroxyapatite crystals (Hartgerink et al., 2001). The dendrimers can also be used to probe the surface to establish the distribution of charge domains, which, in the case of the naturally occurring enamel crystals used in this study, appears to be one alternating band of positive and negative charges.

	ACKNOWLEDGMENTS

 
This investigation was supported in part by USPHS Research Grant DE121899 from the National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD 20892, and the Department of Cariology, Restorative Sciences and Endodontics, the University of Michigan. Fig. 1a is reproduced with permission from James R. Baker, Jr., MD, Center for Biologic Nanotechnology, University of Michigan School of Medicine.

Received October 24, 2002; Last revision February 17, 2003; Accepted February 28, 2003

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Aoba T, Moreno EC (1991). Structural relationship of amelogenin proteins to their regulatory function of enamel mineralization. In: Surface reactive peptides and polymers. Sikes CS, Wheeler AP, editors. Washington, DC: American Chemical Society, pp. 85-106.

Aoba T, Fukae M, Tanabe T, Shimizu M, Moreno EC (1987). Selective adsorption of porcine-amelogenins onto hydroxyapatite and their inhibitory activity on hydroxyapatite growth in supersaturated solutions. Calcif Tissue Int 41:281–289.[ISI][Medline]

Bai P, Warshawsky H (1985). Morphological studies on the distribution of enamel matrix proteins using routine electron microscopy and freeze-fracture replicas in the rat incisor. Anat Rec 212:1–16.[Medline]

Betley TA, Babaszak Holl MM, Orr BG, Swanson DR, Tomalia DA, Baker JR Jr (2001). Tapping mode atomic force microscopy investigation of poly(amidoamine) dendrimers: effects of substrate and pH on dendrimer deformation. Langmuir 17:2768–2773.

Chander S, Fuerstenau DW (1984). Solubility and interfacial properties of hydroxyapatite: a review. In: Adsorption on and surface chemistry of hydroxyapatite. Misra DN, editor. New York: Plenum Press, p. 29-49.

Doi Y, Eanes ED, Shimokawa H, Termine JD (1984). Inhibition of seeded growth of enamel apatite crystals by amelogenin and enamelin proteins in vitro. J Dent Res63:98–105.[Abstract/Free Full Text]

Fejerskov O, Larsen MJ, Richards A, Baelum V (1994) Dental tissue effects of fluoride. Adv Dent Res 8:15–31.[Abstract]

Hartgerink JD, Beniash E, Stupp SI (2001). Self-assembly and mineralization of peptide-amphiphile nanofibers. Science 294:1684–1688.[Abstract/Free Full Text]

Hiller CR, Robinson C, Weatherell JA (1975). Variations in the composition of developing rat incisor enamel. Calcif Tissue Res 18:1–12.[ISI][Medline]

Kirkham J, Zhang J, Brookes SJ, Shore RC, Wood SR, Smith DA, et al. (2000). Evidence for charge domains on developing enamel crystal surfaces. J Dent Res 79:1943–1947.[Abstract/Free Full Text]

Mafe S, Manzanares JA, Reiss H, Thomann JM, Gramain P (1996). Model for the dissolution of calcium hydroxyapatite powder. J Phys Chem 96:861–866.

Mori O, Imae T (1997). AFM investigation of the adsorption process of bovine serum albumin on mica. Colloids and Surfaces B Biointerfaces 9:31–36.

Robinson C, Hiller CR, Weatherell JA (1974). Uptake of 32P-labelled phosphate into developing rat incisor enamel. Calcif Tissue Res 15:143–152.[ISI][Medline]

Robinson C, Kirkham J, Shore RC, Brookes SJ, Bonass WA (1995). Role of the extracellular matrix in enamel development. In: Dental enamel—formation to destruction. Robinson C, Kirkham J, Shore RC, editors. Boca Raton, FL: CRC Press, pp. 105-134.

Spanos N, Koutsoukos PG (2001). Model studies of the effect of ortho-phospho-l-serine on biological mineralization. Langmuir 17:866–872.

Tomalia DA, Naylor AM, Goddard WA III (1990). Starburst dendrimers: molecular-lever control of size, shape, surface chemistry, topology and flexibility from atoms to macroscopic matter. Angew Chem In Ed Engl 29:138–175.

Uppuluri S, Swanson DR, Piehler LT, Li J, Hagnauer GL, Tomalia DA (2000). Core-shell tecto(dendrimers): I. Synthesis and characterization of saturated shell models. Adv Mater 12:796–800.

Wallwork ML, Kirkham J, Zhang J, Smith DA, Brookes SJ, Shore RC, et al. (2001). Binding of matrix proteins to developing enamel crystals: an atomic force microscopy study. Langmuir 17:2508–2513.

Wallwork ML, Kirkham J, Chen H, Chang SR, Robinson C, Smith DA, et al. (2002). Binding of dentin noncollagenous matrix proteins to biological mineral crystals: an atomic force microscopy study. Calcif Tissue Int 71:249–255.[ISI][Medline]

Wierzbicki A, Sikes CS, Madura JD, Drake B (1994). Atomic force microscopy and molecular modeling of protein and peptide binding to calcite. Calcif Tissue Int 54:133–141.[ISI][Medline]

Zhang J, Kirkham J, Wallwork ML, Smith DA, Brookes SJ, Shore RC, et al. (1999). Use of self-assembled monolayers as substrates for atomic force imaging of hydroxyapatite crystals from mammalian skeletal tissues. Langmuir 15:8178–8183.

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (11) Citing Articles via Google Scholar Google Scholar Articles by Chen, H. Articles by Clarkson, B.H. Search for Related Content PubMed PubMed Citation Articles by Chen, H. Articles by Clarkson, B.H.