The Role of Albumin in Developing Rodent Dental Enamel: A Possible Explanation for White Spot Hypoplasia C. ROBINSON, J. KIRKHAM, S.J. BROOKES, and R.C. SHORE Department of Oral Biology, University of Leeds, Clarendon Way, Leeds, LS2 9LU, United Kingdom
The uptake of serum albumin by maturation-stage rodent enamel and the resulting effects on the growth ofenamel crystallites were investigated in vitro. Albumin uptake was demonstrated by means of gel electrophoresis and confirmed by Western blotting with use of monoclonal antibodies. Measurement of crystal size was carried out by direct TEM measurement ofenamel crystallite outlines after incubations in metastable solutions of calcium phosphate. The ability of endogenous enamel enzymes to degrade albumin was investigated by substrate-specific zymography. The results showed that albumin could be taken up by maturationstage enamel and produce inhibition of crystallite growth. There was no detectable proteolytic activity in the enamel against albumin substrate, which suggests that albumin entering enamel by extravasation in vivo may produce incomplete tissue maturation, resulting in a white, opaque appearance on eruption. J Dent Res 71(6):1270-1274, June, 1992
Introduction.
Kirkham, 1984; Kirkham et al., 1988). Other serum proteins have since been reported in developing bovine tissue (Strawich and Glimcher, 1990). While serum proteins may not be produced by the ameloblasts themselves (Couwenhoven et al., 1989), their reported presence in developing enamel raises the possibility that albumin or other serum proteins might adventitiously leak into the enamel, inhibit crystal growth, and impair the maturation process. Albumin is known to bind to hydroxyapatite (Hlady and Furedi-Milhofer, 1979; Menanteau et al., 1987). Warshawsky et al. (1984) have also implied that enamel apatite in particular will bind serum albumin. Albumin has also been identified in other mineralized tissues (Owen et al., 1973; Thomas and Leaver, 1975). The aim of the present study was to investigate albumin uptake by developing rat incisor enamel in vitro and to determine the effect of such uptake on crystal growth in solutions of metastable calcium phosphate. Several workers have demonstrated the presence of serine proteases in enamel at the maturation stage (den Besten and Heffernan, 1989; Overall and Limeback, 1988; Robinson et al., 1990). Robinson et al. (1990) have also shown the effectiveness of serine proteases in promoting crystal growth in vitro, presumably by removing inhibitory enamel peptides. Efforts have therefore also been made to assess the effects of endogenous enamel proteases on albumin in maturation-stage enamel, in order to determine whether albumin entering the enamel during maturation would be degraded by endogenous tissue proteases.
During its development, dental enamel passes through a series of biochemically and visibly definable stages (Robinson et al., 1981; Robinson and Kirkham, 1985): (1) Secretion: Partially mineralized matrix is secreted, and the enamel has a translucent appearance. (2) Transition: Replacement of degraded matrix by tissue fluid occurs, and the enamel has a translucent/corrugated appearance. (3) Maturation: Residual matrix is replaced by tissue fluid which is in turn replaced by massive mineral uptake associated Materials and methods. with crystal growth in width and thickness. The enamel has an Preparation of material.-For all of the investigations in this opaque/white porous appearance. study, mandibular incisors from 150-g male Wistar rats were During maturation (stage 3), the thin, plate-like crystals, used. The animals were killed by anesthetic overdose, and the deposited during the secretion phase, increase greatly in thick- incisors were dissected out of the alveolar bone. Adhering enamel ness and width, displace fluid, and attain the final dimensions organ and soft tissue were removed with a moist paper towel. characteristic of mature tissue. Recent information has shown Drying the enamel in air for one min revealed the location of the that the residual matrix present at this stage is capable of maturation stage (Hiller et al., 1975; Robinson and Kirkham, exerting an inhibitory effect on the growth of apatite crystals 1984) as a white, opaque area approximately 1/3 of the way along (Robinson et al., 1989a). the tooth from the root apex. For crystal-growth studies, a strip The porosity of this developmental stage resulting from matrix of enamel, 0.5 mm in width, was removed at a distance of 0.5 mm loss produces a white, opaque appearance when the enamel is from the transition/maturation boundary in the direction of the dried (Robinson et al. , 1978; Robinson and Kirkham, 1985). When maturing tissue. this appearance occurs in erupted teeth, it is indicative of incomIn vitro incubation ofenamel in metastable calcium phosphate pletely mature tissue. This can be seen in fluorosis (Fejerskov et solutions.-The protocol used was based upon that described by al., 1977) and various kinds of amelogenesis imperfecta (Witkop, Robinson et al. (1989a). 1989). However, it is also associated with a wide variety of other Controls.-(1) For measurement of normal crystallite sizes in conditions (Small and Murray, 1978), all apparently associated the maturation stage, enamel specimens prepared as described with incomplete crystal growth during the maturation stage. above were used and measured as described below. (2) AlternaThe reasons for this incomplete tissue maturation are unclear, tively, the enamel specimens were incubated in metastable calbut recent studies have identified serum albumin in porcine cium phosphate solutions at pH 7.2 (Bachra and Fischer, 1968) enamel (Limeback et al., 1989; Strawich and Glimcher, 1990) ([Cal = 3.75 mmol/L, [P1 = 1.67 mmol/L) for seven days at 370C which, even in normal circumstances, does not appear to mature prior to measurement of crystallite size. (3) Further specimens completely compared with enamel in other species (Robinson and were incubated for 30 min in metastable calcium phosphate solutions made 8 mol/L with respect to urea, in an attempt to remove residual matrix protein while preventing any loss of Received for publication September 9, 1991 mineral. These samples were then either prepared directly for Accepted for publication December 24, 1991 This investigation was supported by the Medical Research Council of measurement of crystallite size, or incubated in metastable calGreat Britain (Grant No. G90136725D). cium phosphate solution as described above.
1270
Downloaded from jdr.sagepub.com at UCSF LIBRARY & CKM on December 15, 2014 For personal use only. No other uses without permission.
Vol. 71 N. 6
a
ALBUMIN IN DEVELOPIPG RODENT ENAMEL b
C
d
A
Mr
FIG I Fig; 1SDS electrophoretogram of maturation stag enamel particles following incubation with 4% albumin (a) Enamel extract after incubation oftissue with 4% albumin solution Material at Mr 65 K can be seen, alone ; with residual matrix proteins at Mr < 30 (K) Enamel extract after treatment of tissue with 8 mol/ urea lowed by incubation with 4% albumin solution Material at Mr 65 Kean be seen, but the residual matrix proteins were presumably removed by the urea. (c) Enamel extract after treatment tissue with 8 moL urea (control. No gained material can be seen. (d)Enamelextractwith nopriortreatmentoftissue (control). Residual matix proteins at Mr < 30 K can be seen, but no band at 65 K can be een. A: Serum albumin. Mr: Molecular-weight markers.
1271
h-
IRG 2 Fir 2"Western blot of extracts of enamel nartiles against monoclonal antibodies to smmr albuir fo lowing SDS gel electrophoresis as shown in Fig. 1. (a) Enamel extract after incbation of tissue with 4% albumir solution. Cross reactive at Mr 65 K cn h sen. (e ) Enamel extract after treatment of issue with 8 mol/L urea followed by incubation with 4% albumin solution Cross-reactiity at Mr 65 K can be see (c) Enamel extract after treatment of issue with 8 molL urea (control No cross reactiity can be sen. d Enamel extract wth no prior tissue treatment (control No cross reactivity can be een A. Serum albumin
grinding the enamel with 50 vtL SDS Treatment buffer (0.5 moI Ths-HC, 1% SDS, pH 6.8 on a watch glass. Duplicate gels (10% Experimental.--Enamel specimens prepared and deproteinated acrylaiide) were run with 15 AL of sample extract loaded per lane. with urea, as described above, were then placed fo 30 min in similar The 4% albumin Solution described above was also included as a solutions but with the addition of 4% (w) serum albumin (.e., positive control. Thegelswererunat75Vandat4 Cforabout l1US h. One of the duplicates was stained with Coomassie Brilliant serum concentrations) at room temperature prior to a further incubation in metastable calcium phosphate alone, for seven days at Blue R250 in 12.5% trichloroacetic acid for 48 h and de-stained 37C. Crystallite dimensions were the measured as described in 7 acetic acid. The remaining gel was electroblotted at 120 mA by the method of Towbin et al. (1979). Protein transfer below. Determination of crystallite size, i.e., crystal growth.Enamnel to nitrocellulose was checked by Ponceau Red staining. The specimens, treated as described above, were removed from treat nitrocellulose sheets were then incubated in 10 mmol/L Trisment solutions and washed with water (pH 7) three times for HC1, 0.9% NaCI, pH 7.4, including 4% soybean protein for removal of excess calcium phosphate. They were subsequently prevention of non-specific binding. After the nitrocellulose was blocked, it was probed with a placed in 200 pL of 0.25 moI glutaraldehyde in 0.1 mol/L phosphate buffer, pH 7.2, for 60 min. Dehydration through graded mouse monoclonal antibody specific for human albumin (Bioalcohols was followed by immersion in propylene oxde and hen genesis, Bournemouth, UK). Areas of cross-reactivity were 50% propylene oxide/epon resin After 18 h, the specimens were visualized with a second antibody conjugated to horseradish infiltrated in fresh resin fbr six h, then finally transferred to fresh peroxidase Sigma, Poole, UK) and the chromogenic substrate resin in polypropylene embedding capsules and accurately oriented. amino-ethyl carbazole. Effect of endogenois enamel enzymes on albumin. A Polymerization was carried out at 60 C overnight. Sections (80-90 nm) were cut with a diamond knife and mounted zymographic technique was used fbr assessment of the efects on Forimvar/crbon grids. Crysli cross-sections were photo- on albumin of endogenous enamel enzymes at each stage of graphed and printed at a constant magnification, and the cross tissue development (Banda et r1., 1987). A 12% acrylamide sectional dimensions of their hexagonal outlines were measured gel was prepared and 1 mg/mL rat serum albumin was included with a Magiscan (Joyce Loebel, UK) image analysis system. Consis- as substrate. Contiguous particles of developing fresh incisor tent magnification was ensured by use of a cross-grating replica enamel from youngest to oldest tissue were prepared as de(2160 lines/mm) for calibration. Results were analyzed by one-way scribed by Hiller et acr (1975). After electrophoresis at 4 C, the gels were washed in 2.5% Triton for 2 x 15 min and then analysis of variance. Determination of albumin uptake into &veloping enamel in incubated in 50 mmol/L Tris buffer, pH 8.0, including 5 mmol/ vitro. Treatment ofspecimens.Enarmel samples from the matura L CaCl 2 at 37 C for up to three days. The gel was then stained tion stage were prepared as described above. Eac sample was then with Coomassie Brilliant Blue R250, and areas of proteolytic immersed directly in 4% serum albumin (Sigma, human fraction V) activity were identified at specific molecular-size locations as for 30 min or was pre-treated with 8 moL urea for 30 mi prior to clear regions on the gel against the stained background. being washed three times with water (pH 7), flowed by immersion in 4% albumin fbr 30 mn. Specimens were washed for a further Results. three times after treatment with albumin. Idntifcationand determination olbumin.-Specimeinstreated Uptake ofalbumin b develping enamel. Fig. 1 shows gel electroas described above were subjected to SDS g1 electrophoresis phoretograms of maturation-stage enamel pre-treated with albu(Laemmh, 1970) on mini gels. The samples were extracted by mm both with and without prior deproteinization with urea and Downloaded from jdr.sagepub.com at UCSF LIBRARY & CKM on December 15, 2014 For personal use only. No other uses without permission.
t;:xpC\
: 'D;
The uptake of serum albumin by maturation-stage rodent enamel and the resulting effects on the growth ofenamel crystallites were investigated in vitro. Albumin uptake was demonstrated by means of gel electrophoresis and confirmed by Western blotting with use of monoclonal antibodies. Measurement of crystal size was carried out by direct TEM measurement ofenamel crystallite outlines after incubations in metastable solutions of calcium phosphate. The ability of endogenous enamel enzymes to degrade albumin was investigated by substrate-specific zymography. The results showed that albumin could be taken up by maturationstage enamel and produce inhibition of crystallite growth. There was no detectable proteolytic activity in the enamel against albumin substrate, which suggests that albumin entering enamel by extravasation in vivo may produce incomplete tissue maturation, resulting in a white, opaque appearance on eruption. J Dent Res 71(6):1270-1274, June, 1992
Introduction.
Kirkham, 1984; Kirkham et al., 1988). Other serum proteins have since been reported in developing bovine tissue (Strawich and Glimcher, 1990). While serum proteins may not be produced by the ameloblasts themselves (Couwenhoven et al., 1989), their reported presence in developing enamel raises the possibility that albumin or other serum proteins might adventitiously leak into the enamel, inhibit crystal growth, and impair the maturation process. Albumin is known to bind to hydroxyapatite (Hlady and Furedi-Milhofer, 1979; Menanteau et al., 1987). Warshawsky et al. (1984) have also implied that enamel apatite in particular will bind serum albumin. Albumin has also been identified in other mineralized tissues (Owen et al., 1973; Thomas and Leaver, 1975). The aim of the present study was to investigate albumin uptake by developing rat incisor enamel in vitro and to determine the effect of such uptake on crystal growth in solutions of metastable calcium phosphate. Several workers have demonstrated the presence of serine proteases in enamel at the maturation stage (den Besten and Heffernan, 1989; Overall and Limeback, 1988; Robinson et al., 1990). Robinson et al. (1990) have also shown the effectiveness of serine proteases in promoting crystal growth in vitro, presumably by removing inhibitory enamel peptides. Efforts have therefore also been made to assess the effects of endogenous enamel proteases on albumin in maturation-stage enamel, in order to determine whether albumin entering the enamel during maturation would be degraded by endogenous tissue proteases.
During its development, dental enamel passes through a series of biochemically and visibly definable stages (Robinson et al., 1981; Robinson and Kirkham, 1985): (1) Secretion: Partially mineralized matrix is secreted, and the enamel has a translucent appearance. (2) Transition: Replacement of degraded matrix by tissue fluid occurs, and the enamel has a translucent/corrugated appearance. (3) Maturation: Residual matrix is replaced by tissue fluid which is in turn replaced by massive mineral uptake associated Materials and methods. with crystal growth in width and thickness. The enamel has an Preparation of material.-For all of the investigations in this opaque/white porous appearance. study, mandibular incisors from 150-g male Wistar rats were During maturation (stage 3), the thin, plate-like crystals, used. The animals were killed by anesthetic overdose, and the deposited during the secretion phase, increase greatly in thick- incisors were dissected out of the alveolar bone. Adhering enamel ness and width, displace fluid, and attain the final dimensions organ and soft tissue were removed with a moist paper towel. characteristic of mature tissue. Recent information has shown Drying the enamel in air for one min revealed the location of the that the residual matrix present at this stage is capable of maturation stage (Hiller et al., 1975; Robinson and Kirkham, exerting an inhibitory effect on the growth of apatite crystals 1984) as a white, opaque area approximately 1/3 of the way along (Robinson et al., 1989a). the tooth from the root apex. For crystal-growth studies, a strip The porosity of this developmental stage resulting from matrix of enamel, 0.5 mm in width, was removed at a distance of 0.5 mm loss produces a white, opaque appearance when the enamel is from the transition/maturation boundary in the direction of the dried (Robinson et al. , 1978; Robinson and Kirkham, 1985). When maturing tissue. this appearance occurs in erupted teeth, it is indicative of incomIn vitro incubation ofenamel in metastable calcium phosphate pletely mature tissue. This can be seen in fluorosis (Fejerskov et solutions.-The protocol used was based upon that described by al., 1977) and various kinds of amelogenesis imperfecta (Witkop, Robinson et al. (1989a). 1989). However, it is also associated with a wide variety of other Controls.-(1) For measurement of normal crystallite sizes in conditions (Small and Murray, 1978), all apparently associated the maturation stage, enamel specimens prepared as described with incomplete crystal growth during the maturation stage. above were used and measured as described below. (2) AlternaThe reasons for this incomplete tissue maturation are unclear, tively, the enamel specimens were incubated in metastable calbut recent studies have identified serum albumin in porcine cium phosphate solutions at pH 7.2 (Bachra and Fischer, 1968) enamel (Limeback et al., 1989; Strawich and Glimcher, 1990) ([Cal = 3.75 mmol/L, [P1 = 1.67 mmol/L) for seven days at 370C which, even in normal circumstances, does not appear to mature prior to measurement of crystallite size. (3) Further specimens completely compared with enamel in other species (Robinson and were incubated for 30 min in metastable calcium phosphate solutions made 8 mol/L with respect to urea, in an attempt to remove residual matrix protein while preventing any loss of Received for publication September 9, 1991 mineral. These samples were then either prepared directly for Accepted for publication December 24, 1991 This investigation was supported by the Medical Research Council of measurement of crystallite size, or incubated in metastable calGreat Britain (Grant No. G90136725D). cium phosphate solution as described above.
1270
Downloaded from jdr.sagepub.com at UCSF LIBRARY & CKM on December 15, 2014 For personal use only. No other uses without permission.
Vol. 71 N. 6
a
ALBUMIN IN DEVELOPIPG RODENT ENAMEL b
C
d
A
Mr
FIG I Fig; 1SDS electrophoretogram of maturation stag enamel particles following incubation with 4% albumin (a) Enamel extract after incubation oftissue with 4% albumin solution Material at Mr 65 K can be seen, alone ; with residual matrix proteins at Mr < 30 (K) Enamel extract after treatment of tissue with 8 mol/ urea lowed by incubation with 4% albumin solution Material at Mr 65 Kean be seen, but the residual matrix proteins were presumably removed by the urea. (c) Enamel extract after treatment tissue with 8 moL urea (control. No gained material can be seen. (d)Enamelextractwith nopriortreatmentoftissue (control). Residual matix proteins at Mr < 30 K can be seen, but no band at 65 K can be een. A: Serum albumin. Mr: Molecular-weight markers.
1271
h-
IRG 2 Fir 2"Western blot of extracts of enamel nartiles against monoclonal antibodies to smmr albuir fo lowing SDS gel electrophoresis as shown in Fig. 1. (a) Enamel extract after incbation of tissue with 4% albumir solution. Cross reactive at Mr 65 K cn h sen. (e ) Enamel extract after treatment of issue with 8 mol/L urea followed by incubation with 4% albumin solution Cross-reactiity at Mr 65 K can be see (c) Enamel extract after treatment of issue with 8 molL urea (control No cross reactiity can be sen. d Enamel extract wth no prior tissue treatment (control No cross reactivity can be een A. Serum albumin
grinding the enamel with 50 vtL SDS Treatment buffer (0.5 moI Ths-HC, 1% SDS, pH 6.8 on a watch glass. Duplicate gels (10% Experimental.--Enamel specimens prepared and deproteinated acrylaiide) were run with 15 AL of sample extract loaded per lane. with urea, as described above, were then placed fo 30 min in similar The 4% albumin Solution described above was also included as a solutions but with the addition of 4% (w) serum albumin (.e., positive control. Thegelswererunat75Vandat4 Cforabout l1US h. One of the duplicates was stained with Coomassie Brilliant serum concentrations) at room temperature prior to a further incubation in metastable calcium phosphate alone, for seven days at Blue R250 in 12.5% trichloroacetic acid for 48 h and de-stained 37C. Crystallite dimensions were the measured as described in 7 acetic acid. The remaining gel was electroblotted at 120 mA by the method of Towbin et al. (1979). Protein transfer below. Determination of crystallite size, i.e., crystal growth.Enamnel to nitrocellulose was checked by Ponceau Red staining. The specimens, treated as described above, were removed from treat nitrocellulose sheets were then incubated in 10 mmol/L Trisment solutions and washed with water (pH 7) three times for HC1, 0.9% NaCI, pH 7.4, including 4% soybean protein for removal of excess calcium phosphate. They were subsequently prevention of non-specific binding. After the nitrocellulose was blocked, it was probed with a placed in 200 pL of 0.25 moI glutaraldehyde in 0.1 mol/L phosphate buffer, pH 7.2, for 60 min. Dehydration through graded mouse monoclonal antibody specific for human albumin (Bioalcohols was followed by immersion in propylene oxde and hen genesis, Bournemouth, UK). Areas of cross-reactivity were 50% propylene oxide/epon resin After 18 h, the specimens were visualized with a second antibody conjugated to horseradish infiltrated in fresh resin fbr six h, then finally transferred to fresh peroxidase Sigma, Poole, UK) and the chromogenic substrate resin in polypropylene embedding capsules and accurately oriented. amino-ethyl carbazole. Effect of endogenois enamel enzymes on albumin. A Polymerization was carried out at 60 C overnight. Sections (80-90 nm) were cut with a diamond knife and mounted zymographic technique was used fbr assessment of the efects on Forimvar/crbon grids. Crysli cross-sections were photo- on albumin of endogenous enamel enzymes at each stage of graphed and printed at a constant magnification, and the cross tissue development (Banda et r1., 1987). A 12% acrylamide sectional dimensions of their hexagonal outlines were measured gel was prepared and 1 mg/mL rat serum albumin was included with a Magiscan (Joyce Loebel, UK) image analysis system. Consis- as substrate. Contiguous particles of developing fresh incisor tent magnification was ensured by use of a cross-grating replica enamel from youngest to oldest tissue were prepared as de(2160 lines/mm) for calibration. Results were analyzed by one-way scribed by Hiller et acr (1975). After electrophoresis at 4 C, the gels were washed in 2.5% Triton for 2 x 15 min and then analysis of variance. Determination of albumin uptake into &veloping enamel in incubated in 50 mmol/L Tris buffer, pH 8.0, including 5 mmol/ vitro. Treatment ofspecimens.Enarmel samples from the matura L CaCl 2 at 37 C for up to three days. The gel was then stained tion stage were prepared as described above. Eac sample was then with Coomassie Brilliant Blue R250, and areas of proteolytic immersed directly in 4% serum albumin (Sigma, human fraction V) activity were identified at specific molecular-size locations as for 30 min or was pre-treated with 8 moL urea for 30 mi prior to clear regions on the gel against the stained background. being washed three times with water (pH 7), flowed by immersion in 4% albumin fbr 30 mn. Specimens were washed for a further Results. three times after treatment with albumin. Idntifcationand determination olbumin.-Specimeinstreated Uptake ofalbumin b develping enamel. Fig. 1 shows gel electroas described above were subjected to SDS g1 electrophoresis phoretograms of maturation-stage enamel pre-treated with albu(Laemmh, 1970) on mini gels. The samples were extracted by mm both with and without prior deproteinization with urea and Downloaded from jdr.sagepub.com at UCSF LIBRARY & CKM on December 15, 2014 For personal use only. No other uses without permission.
t;:xpC\
: 'D;
