- Email: [email protected]
Proteinase, phosphatase and glucuronidase activities in the growing mandible and temporomandibular joint of the guinea pig
= = = = = = = = = = ANNALS Of ANATOMY = = = = = = = = = =
Proteinase, phosphatase and glucuronidase activities in the growing mandible and temporomandibular joint of the guinea pig Kauko P. Isotupa*, Kauko K. Miikinen* and David S. Carlson**
* Department of Biologic and Materials Sciences, School of Dentistry, and ** Center for Human Growth and Development, The University of Michigan, Ann Arbor, Michigan 48109-1078, USA
Summary. This study investigated the effect of muscle function and occlusal form on the activity levels of several enzymes present in the mandible or temporomandibular joint of the guinea pig. Restriction of maxillary width and asymmetric function of the mandible was caused in 12 animals at the age of 10 days, as described in the accompanying paper (Isotupa et al. 1992). Tissue samples from six anatomical sites were obtained from the animals sacrificed 4, 8, or 12 weeks after manipulation (4 animals in each group). Six agematched animals acted as controls. Five samples were from the following sites of the mandible: the condylar cartilage, the lower and upper parts of processus angularis, and the anterior and posterior halves of the condyle neck. A sample was also obtained from fossa mandibularis of the temporal bone. Buffer extracts of powderized samples were studied for acid and alkaline phosphatase, glucuronidase and two types of proteolytic activity. Although the asymmetric manipulation of occlusion caused observable, localized asymmetric growth, the enzyme findings were not consistently asymmetric. Manipulation generally increased all enzyme activities regardless of whether apposition or resorption was involved. The activities of alkaline. phosphatase more consistently showed this pattern, and changes in enzyme activities seemed to be sensitive, reflecting cellular or molecular level of growth. The enzyme activities may also reflect a certain type of after-effect of irritation, or a healing period. The inclusion of several types of enzyme determinations is recommended to complete macroscopic measurements. Subsequent chromatographic and electrophoretic studies are also useful. Key words: Enzymes - Guinea pig - Bone growth Facial skeleton
~I
Ann. Anat. (1992) 174: 441-446 Gustav Fischer Verlag Jena
Introduction In bone growth studies, the amount of growth during a certain time period has generally been determined by linear measurements from the bone itself or from the radiographic picture of the bone. More accurate results have been obtained by using certain markers placed in the bone, such as amalgam shots or pins (Bjork 1955). By using vital staining methods, it is possible to mark the entire calcified bone surface; the new bone growth that happened after staining is discernible and measurable (MassIer and Schour 1951; Hoyte 1956; Baer and Ackerman 1966). These methods are usable when studying appositional bone growth mostly on the macroscopic level. However, the resorption of the bone - a very important part of bone growth studies - is difficult to estimate accurately with these methods. Using histological procedures, it is possible to observe the resorption areas in the bone and on the bone surface, but the estimation of the extent and rate of the process - and the changes in the extent and rate, cannot still be performed with sufficient accuracy. Recently, there have been attempts to utilize histochemical (Nielsen and Magnusson 1981) and biochemical techniques (Copray et al. 1985) in bone growth studies. By measuring activities of certain enzymes related to the growth of bone and cartilage, studies of bone remodeling have been conducted (Reddi and Huggins 1972; Stutzman and Petrovic 1984; Bollen et al. 1989). Measurement of the activities of acid phosphatase and glucuronidase has been assumed to elucidate the bone resorption areas themselves and even the rate of bone resorption (Stutzman and Petrovic 1984). In our previous paper (Isotupa et al. 1992) we elucidated the effect of muscle activity on the growing facial skeleton in the guinea pig. In that study, tissue samples from six
different bone areas were obtained - five from the mandible and one from the skull - after 4, 8 and 12 weeks following the manipulation made when the animals were ten days old. The samples were studied for hydrolytic enzyme activities that broadly represented the proteolytic, phosphorolytic and glucuronolytic capability of the tissues. The rationale behind these enzyme determinations was the need to study whether any of the macroscopic findings described in the previous paper were reflected in the biochemical enzyme measurements. The primary purpose of this study was thus to investigate whether certain enzyme determinations could be utilized to complete other procedures in order to obtain a better idea of growth in the guinea pig facial skeleton.
Materials and methods General study design The animals and the experimental arrangements were explained in detail in our previous paper (Isotupa et al. 1992). Eighteen male guinea pigs were divided into experimental (n = 12) and control groups (n = 6). At the age of ten days the guinea pigs were anaesthetized and the radiocephalograms were obtained as described in the above paper. In the experimental animals, the mouth was propped open and a l-cm incision was made longitudinally in the midline of the palate. The midpalatal suture was cleared of periosteum, and cyanoacrylate glue was injected in the soft sutural tissue and onto the bone surface across the midpalatal suture. The suture was then closed by pressing the flaps against the palatal bones. With the mouth remaining open, a dental handpiece and a carborundum grinding wheel were used to grind away the buccal aspect of the lower molars and the palatal aspect of the upper molars on the right side. The incisors were ground at an angle slanting up toward the left (Isotupa et al. 1992). The dental grinding was repeated under anaesthesia every two weeks. Four experimental and two control animals were killed at 4, 8 and 12 weeks postoperatively. The mandible was detached and bisected. The remaining tissues were carefully removed, after which samples of bone and cartilage were taken for the studies described below.
Fig. 1. Designation of the anatomical sites of the mandible (1 - 3, 5, 6) and the fossa mandibularis of the temporal bone (4), used in enzyme studies.
Preparation of tissue samples for biochemical studies The tissue samples for biochemical studies were obtained from six anatomical sites (five from the mandible and one from the cranium) of all animals. These sites are shown in Fig. 1, and in
Chemical methods The activities of acid and alkaline phosphatase were determined in 0.05 M ~,~-dimethylglutarate buffer (pH 5.0) and in 0.05 M carbonate-bicarbonate buffer (pH 9.9), respectively. The reaction mixtures consisted of 0.3 ml of buffer, 0.1 rn1 of 6.0 mM pnitrophenyl phosphate, 0.15 rn1 of water and of 50!t1 of enzyme. The reaction time was 60 min at 37°C. The activity of ~ glucuronidase was determined in 0.05 M acetate buffer, pH 4.5. The reaction mixtures consisted of 0.3 ml of buffer, of 0.1 rn1 of 6 mM p-nitrophenyl-~-D-glucuronide, of 0.15 rn1 of water, and of 50 !tl of enzyme. The reaction time was 180 min at 37°C. In the above cases the reactions were arrested either with 0.4 ml of 0.1 M NaOH (acid phosphatase and glucuronidase) or with 0.4 ml of 25 mM EDTA (alkaline phosphatase) and the readings were made at 405 nm. The activity of proteolytic enzymes was studied using azocasein and glycyl-L-prolyl-4-methoxy-2-naphthylamide (GPm2NA) as substrates. In the former case, the method was used as presented by Seppa and Jarvinen (1978),· and in the latter procedure the enzyme activities were determined as described previously (Makinen and Makinen 1978). The protein was measured according to the Pierce (Rockford, Illinois, USA) Micro BCA Assay Reagent. If not otherwise mentioned, all chemicals were obtaind from Sigma. The water used in the studies was prepared with a Millipore-Milli-Q system and had a resistivity of 18 MQm- 1 •
442
exhibited about two times higher acid phosphatase activity than the controls. The glucuronidase activities followed the All extracts were subjected to SDS-PAGE with PhastGel gradient same pattern, although the 12-week groups did not differ as media (8 - 25 % ), using Coomassie Blue in staining. much as in the case of acid phosphatase (the increase was about 35 % in favor of the treated animals). The activities of alkaline phosphatase also followed this pattern, but the 12Results week old treated animals exhibited about 6-fold higher activities than the controls. The enzymes hydrolyzing azocaAs expected, the specific enzyme activities measured from sein showed up to 3-fold difference in favor of the treated the left and right sides of the control animals (two in each animals at the 4-week stage, but this difference leveled off age group, providing a total offour samples from each ofthe six anatomical sites in each group) were very similar (Fig. later. Enzymes hydrolyzing GPm-2NA showed no signifi2). Therefore, the means of these values were used below to cant differences between groups at any time point. It was difficult to observe any correlation between represent the enzyme activities of untreated controls. In the enzyme activities and macroscopic measurements. Howtreated animals (n = 2 in the two-week group; n = 4 in the ever, the right molars were initially ground leaving only a eight- and 12-week groups), the activities of the left and light contact between mandibular and maxillary teeth. When right sides were first compared in order to elucidate the the treated animals wore out all interferences of occlusion correlation between asymmetric growth and enzyme brought about by the grinding procedure, the right molars activities. When the control enzyme activities in each age obviously grew faster with a concomitant, increased pressure group were subsequently compared with those of the treated in the right TMJ compared with the left one. Macroscopic animals, the means ofthe specific activities from the left and measurements showed that the vertical growth in the right right sides of the latter animals were used. Condylar cartilage. Both control and experimental half of the mandible was indeed slightly lower than in the left groups displayed similar acid phosphatase activity after 4 half, although the enzyme activities did not differ between and 8 weeks. In the 12-week group, the treated animals right and left. Changes in all enzyme activities were in general in the same direction: increases in the activities of alkaline phosphatase, which is normally associated with 3,1 appositional bone growth, were always accompanied by ACID PHOSPHATASE 2.4 5.0 4. increases in the activities of acid phosphatase and glucuronidase, in spite of the fact that the two latter enzymes 2.0 4.0 are frequently associated with osteoclastic functions. 6. Lower part of processus angularis. Acid phosphatase 1.6 2. and glucuronidase activities did not differ appreciably as a 3.0 5. function of time and treatment. Alkaline phosphatase 1.2 1. activities tended to be higher in the treated animals after 2.0 0.8 eight and 12 weeks. Of the proteolytic activities, those determined with azocasein were about 2.5 times higher in 1.0 2. (/) 0.4 the treated animals compared with controls, in all age Q) 5. :;:::; 6. groups. All activities determined with GPm-2NA were os: 0 0 on similar. 12 4 8 4 8 12 « The lower part of the processus angularis represents the 0 ;;::: insertion of the m. pterygoideus med., which appeared to be AZOCASEIN ALKAUNE ° 3. 0Q) 2.5 0.2 PROTEASE PHOSPHATASE strongly affected by the occlusal deviation brought about by a. the dental grinding. In extreme cases, hyperactivity led to C/J 6. 2.0 .16 such a narrowing of the mandible that the entire processus 1. angularis was resorbed. The enzyme studies did not provide 5. 1.5 a logical, consistent result with regard to asymmetry, despite the retarded vertical growth of the ramus observed macro2. scopically in the mandible. However, the manipulation of 1.0 .08 occlusion did exert an exceptional effect on enzyme activities. While manipulation generally induced elevated 0.5 enzyme activities in the tissues studied, the lower part of processus angularis displayed reduced activities of acid o L..-.l...-_ _L - _ - - - l phosphatase and glucuronidase. These observations should 12 4 12 4 8 8 indicate increased de novo formation of bone. This may TIME (weeks) quite well have been the case at the time of tissue sampling, when the appositional growth of the condylar area had Fig. 2. Specific enzyme activities of six anatomical sites of control animals given as a function of animal age. Each experimental point slowed down because of grinding and the consequentional represents 12 half-jaws obtained from six animals. pressure of the condyle against the fossa. Therefore, sub-
Electrophoresis
443
stitutional vertical growth most likely took place in the lower edge of the ramus and on its lateral surface. Upper part of processus angularis. The activities of acid phosphatase were higher in the treated animals, especially in the 8-week group where a 3-fold increase compared with controls was observed. In the glucuronidase determinations, the 8-week group showed a large 5-fold difference while the 4- and 12~week groups and controls were essentially the same. In alkaline phosphatase determinations, the 4-week groups showed no difference, whereas after 8 and 12 weeks there was a clear 4- and 3-fold difference, respectively, in favor of the treated animals. The treatment increased the activities with both proteolytic enzyme substrates, even more so when GPm-2NA was used as substrate. The upper part of processus angularis, which represents an area of appositional growth, exhibited increased activity of enzymes which are normally involved in the growth and degradation of bone. The macroscopic measurements showed that the activation of m. pterygoideus was followed by medial bending of the angular process. In extreme situations this led to the resorption of the entire process. Another important observation was the slight retardation of the vertical growth of the process in the treated animals compared with controls. If acid phosphatase and glucuronidase can be assumed to be associated with bone resorption, the increase in the activities of these enzymes in the treated animals was thus expected and logical. The increased activity of alkaline phosphatase could be explained in terms of the resorption of the medial side of the processus angularis being followed by concomitant growth of bone on the lateral side. Although the final result was an angular process that was lower than normal, it is possible that its maintenance required accelerated appositional growth. Fossa mandibularis. The treated animals showed after 4 and 8 weeks a clear 2.5- and 2-fold increase, respectively, in the acid phosphatase activities compared with controls. In the 12-week old animals this trend had leveled off, reflecting most likely the fact that in younger animals the growth processes were more active than in the older ones, especially in the temporal portion of the temporomandibular joint. The glucuronidase pattern followed that of acid phosphatase. Activities of alkaline phosphatase showed a 2-fold increase in favor of the treated animals after 8-week-treatment, but the groups did not differ at 4- and 12-weeks. The activities tested with azocasein did not differ at all, while those tested with GPm-2NA showed, after 4 weeks'-treatment, an about 2-fold increase compared with controls. These tissue samples were the largest in size studied in this analysis, and contained both cartilage and bone. The measured enzyme activities were generally very high, a finding also observed in the condylar cartilage. The differences between the enzyme activities of the latter and fossa mandibularis may have resulted from the presence of bone in the fossa. The asymmetric manipulation of occlusion was not reflected in enzyme activities. However, manipUlation caused a general increase in the activities, the degree of which varied between the right and left sides. Manipulation also increased the enzyme activities more clearly in younger
animals, whereas the corresponding activity increase in the condylar cartilage was observable first in the older age groups. The reason for this finding remains to be elucidated. Anterior half of the condyle neck. Acid phosphatase activities were low in the anterior region of the condyle neck, but the treated animals always displayed higher activities than the control animals. In the glucuronidase determinations, there was a strong, 5-to 6-fold increase after 8 weeks in the treated animals, but the 4-week and the 12week groups did not differ. Alkaline phosphatase exhibited a clear 6-fold increase after 8 weeks'-treatment, and an about 2-fold increase after 12 weeks' treatment. Azocasein hydrolysis was about 50 % higher in all treated animals. The enzymes hydrolyzing GPm-2NA showed the highest activities - about 3.5-fold compared to controls - after 12 weeks of treatment. ' In normal growth, the anterior half of the condyle neck represents an area of resorption. In the enzyme activities this was seen with increasing age of the animals in the form of decreased activities of alkaline phosphatase and elevated activities of glucuronidase. The specific enzyme activities were generally low, indicating the involvement of smaller growth changes than were expected in the growth area. It should indeed be noted that, in the treated group, the activities of both enzymes increased, but the balance was still in favor of resorption. Posterior half of the condyle neck. The posterior region of the condyle neck also showed low acid phosphatase activity, but the treated animals exhibited 3- to 4-times higher activity than the controls at all ages. The activities of glucuronidase increased as a function of age, but they did not differ significantly between the treatment and the control groups, except in one aspect: the 12-week-old treated animals showed a clearly lower activity than the controls. This marked one of the very few exceptions where the control tissues showed an elevated enzyme activity in this study. The activity of alkaline phosphatase followed the same pattern: the controls exerted, as an exception from the general rule, about 3-fold higher activity than the treatment group after 4 weeks (after 8 and 12 weeks the treatment had caused an increase in enzyme activities, in line with most of the cases mentioned above). Azocasein hydrolysis showed an increase in treated animals at all ages, but the differences from controls were not significant. All activities tested with azocasein increased as a function of animal age. The activities tested with GPm-2NA also increased as a function of age, and were about 3 times higher in animals treated for 12 weeks than in the controls; after 4 and 8 weeks the activities were relatively similar. It is known that when the mandible grows in length, the posterior half of the condyle neck represents an area of appositional growth, although the longitudinal growth of the mandible and the skull is relatively minor in the guinea pig compared with the rat. The low enzyme activities were expected as a result of the dominant proportion of mature, mineralized bone in the samples studied. Therefore, the decrease in the activities of alkaline phosphatase with increasing age may be understandable. The experimental
444
5.0
3
2 • Control
4.0
o Experimental
3.0
(/)
2.0
Q)
+:: .>
1.0
0
0
+::
4
0
;;::
Fig. 3. Specific activities of alkaline phosphatase measured in six anatomical sites of control and treated animals. In the experimental group, each point represents four half-jaws at four weeks, and eight half-jaws at eight and 12 weeks. The number of samples studied in the control group was four at all time points.
·0 Q)
0-
W
4.0
8
12
8
4
12
8
12
6
5
4
4
3.0 2.0 1.0
o '---'-__--'-__........_ 4
12
8
1.--1-_ _-1-._ _- - 1 . _ L--....L.._ _-I-._ _- - I . _
4
8
12
4
8
12
TIME (weeks)
arrangements rendered somewhat elevated enzyme activities possible in this area of the condyle neck, although this elevation did concern more significantly the hydrolysis of GPm-2NA in the 12-week-treated group. A graphical representation of the effect of the treatment on the activities of alkaline phosphatase is shown in Fig. 3. Electrophoresis. The aqueous extracts representing all six anatomical sites were subjected to SDS-PAGE followed by Coomassie Blue staining. The purpose of these experiments was to study whether any of the macroscopic and enzymatic findings were reflected in the electrophoretograms of the soluble, stainable proteins present in the extracts. These experiments were performed considering the fact that most proteins derived from the above tissue samples, and which were soluble and stainable under the present conditions, should be similar and represent various dominant intracellular and extracellular proteins and peptides whose relative levels in the extracts studied could be assumed to remain virtually unchanged regardless of the treatment carried out. Likewise, it was assumed that the concentrations of the enzymes hydrolyzing the above mentioned substrates were too low in the tissue extracts to be demonstrated as stainable bands in the present electrophoretograms. In spite of these reservations, it was possible to make a number of consistent observations which are summarized below. Condylar cartilage. The control animals showed no significant differences as a function of age; the 4-, 8- and 12week-old animals gave a similar electrophoretic pattern. Also, the same protein pattern was present in the samples obtained from the left and the right halves of the skull in the treated animals. Comparison between treated and control animals showed the following general trend. There were differences in the protein pattern after four weeks. After eight weeks those differences had partly disappeared and after 12 weeks there were no observable electrophoretic differences between controls and treated animals. The above
differences concerning the 4-week-old animals were related to the presence of bands in the two extreme ends of the gels, indicating that soluble proteins with relatively low and high molecular weights had undergone concentration changes because of the treatment. In the 4-week-old treated animals, one band, representing a small-molecular protein, was consistently absent. Also, one high-molecular weight band was missing. Both bands were clearly present in the electrophoretograms of control samples. Lower part of processus angularis. The treated animals showed a difference between the left and the right sides: there was a high-molecular weight band present in the 4week-old animals after treatment. After eight weeks the intensity of this band was clearly decreased, and after 12 weeks there was no longer any difference between the left and right sides. Fossa mandibularis. These samples showed an extra band in the small-molecular weight region of the gel, after four weeks of treatment in the right side of the mandible; the intensity of this band was clearly higher than in the controls. There were no differences between the 8-week-old animals. After 12 weeks, the left side samples showed an extra band in the low-molecular weight region of the gels. It is obvious that more detailed chromatographic studies using sensitive techniques will further elucidate metabolic and molecular aspects of bone growth in guinea pigs.
Discussion The guinea pig was chosen as the experimental animal because of its size; when the right and left sides were to be studied separately, the tissue sample had to be large enough for biochemical studies and the preceding preparations. On the other hand, the large number of anatomical sites involved, and the resulting numerous enzyme determinations in'volving laborious pulverization and extraction procedures,
445
stipulated the number of animals that could be investigated. The tissue samples were obtained from areas where the putative alterations of muscle activity, caused by manipulation of occlusions , could induce growth changes. The purpose of the enzyme assays was thus to study whether the activity levels of acid and alkaline phosphatase, glucuronidase and certain proteinases could be used as additional tools in bone growth research. The enzyme activities that were associated with normal growth (determined with control animals), showed a relatively good "symmetric" result, i. e., both the right and left sides expectedly yielded similar specific activities in the same animal. There were differences between individual animals even in the control group, which also was expected. However, because the deviations between the size of individual bones were rather ,small at any given age of the animals, one could assume the changes in enzyme activities to be subtle and perhaps short-term in duration. A change in specific activity may not~ therefore, reflect a measurable change in bone growth. , Excluding a few exceptions, the enzyme activities of the treated animals were higher than those of control animals, indicating a clearly accelerated turnover of enzymes that are normally associated with the growth and degradation of bone. Because of the obviously quite diversified metabolic roles of the enzymes studied, it is not justifiable to draw conclusions about bone growth basing on one enzyme alone. Focusing on several enzymes with different metabolic functions, a somewhat better picture of bone growth could be obtained. At best, however, even these methods cannot offer more than a glimpse into the prevailing, and not so much into the past, situation. The asymmetric manipulation of occlusion elicited observable, although small, localized asymmetric growth. The enzyme findings from the corresponding areas were not consistently asymmetric. One has to observe, however, that the macroscopic measurements indicated growth that had already taken place, while the enzyme activities reflected the nature of ongoing growth. The asymmetric manipUlation was expected to exert a more long-term, even a continuous effect, forcing the mandible to grow asymmetrically to the left. The guinea pigs were able, however, to grind the interference away in about two days, and the effects remained obviously too short-term considering the present enzyme study. It is possible, therefore, that the determination of enzyme activites - two weeks following the last grinding - took place too late to reflect the asymmetric growth described in the accompanying paper. The present enzyme activities may rather reflect a certain type of after-effect of irritation, or a healing period, when asymmetric observations may be possible. On the other hand, if the enzyme activities had been performed soon after treatment, they would perhaps have reflected the direction of ongoing growth which we would have been unable to observe, however, by measuring appositional growth. Based on the present study, it appears to be possible to elucidate bone growth by means of enzyme activity determinations, but with certain reservations . A complicating factor concerns the relatively high individual differences. Furthermore , a relatively mild irritation could cause strong changes
in enzyme activities. All enzyme activities were increased regardless of whether apposition or resorption was to be the dominant event in the area. However, changes in enzyme activities seemed to be sensitive and reflected the microscopic or molecular level of growth (connoting that regular histological procedures would represent the macroscopic level where apposition and resorption had been observed, as reported in the accompanying paper). This indicates that only a slight and transient irritation may induce changes in enzyme activities, and that a real observable growth change presumed a long-term force and irritation. If only a minor irritation induced alterations in enzyme activities, it is possible that the biologic changes of bone, involved in orthodontic practice, could be brought about by really weak forces , provided that such forces are sufficiently long-term.
References Baer MJ, Ackerman LL (1966) A longitudinal vital staining method for the study of apposition in bone,. In : Studies on the anatomy and function of bone and joints. Evans FG (ed) Springer, New York, pp 81-92 Bjork A (1955) Facial growth in man, studied with the aid of metallic implants . Acta Odont Scand 13: 9-34 Bollen AM, Makinen KK, Makinen PL, Carlson DS (1989) Collagenolytic and phosphatase activity in the rat mandible after functional protrusion. Archs Oral BioI 34: 267-273 Copray JCVM, Janssen HWB, Duterloo HS (1985) Effect of compressive forces on phosphatase activity in mandibular condylar cartilage of the rat in vitro . J Anat 140: 479-489 Hoyte DAN (1956) A histological study of bone growth using Alizarin red S. J Anat 90: 585 Isotupa K, Carlson DS , Makinen KK (1992) Influence of asymmetric occlusal relationships and decreased maxillary width on the growth of the facial skeleton in the guinea pig . Anat Anz 174: 447-451 Makinen KK, Makinen PL (1978) Purification and characterization of two human erythrocyte arylamidases preferentially hydrolysing N-terminal arginine or lysine residues. Biochem J 175: 1051-1067 MassIer M, Schour I (1951) The growth pattern of the cranial vault in the albino rat as measured by vital staining with Alizarin red S. Anat Rec 110: 83-101 Nielsen R, Magnusson BC (1981) Enzyme histochemical studies of induced heterotopic cartilage and bone formation in guinea pigs with special reference to acid phosphatase. Scand J Dent Res 89 : 491-498 Reddi AH, Huggins C (1972) Biochemical sequences in the transformation of normal fibroblasts in adolescent rats. Proc Nat! Acad Sci 69: 1601-1605 Seppa HE, Jarvinen M (1978) Rat skin main neutral protease: purification and properties. J Invest Dermatol 70: 84-89 Stutzman J, Petrovic A (1984) Human alveolar bone tum-over rate. A quantitative study of spontaneous and therapeutically-induced variations. In : McNamara, Jr. JA , Ribbons KA (eds) Malocclusion and Periodontium. Craniofacial Growth Series, Center of Human Growth and Development, Ann Arbor, Michigan , pp 185-212 Accepted October 30, 1991
446
Proteinase, phosphatase and glucuronidase activities in the growing mandible and temporomandibular joint of the guinea pig Kauko P. Isotupa*, Kauko K. Miikinen* and David S. Carlson**
* Department of Biologic and Materials Sciences, School of Dentistry, and ** Center for Human Growth and Development, The University of Michigan, Ann Arbor, Michigan 48109-1078, USA
Summary. This study investigated the effect of muscle function and occlusal form on the activity levels of several enzymes present in the mandible or temporomandibular joint of the guinea pig. Restriction of maxillary width and asymmetric function of the mandible was caused in 12 animals at the age of 10 days, as described in the accompanying paper (Isotupa et al. 1992). Tissue samples from six anatomical sites were obtained from the animals sacrificed 4, 8, or 12 weeks after manipulation (4 animals in each group). Six agematched animals acted as controls. Five samples were from the following sites of the mandible: the condylar cartilage, the lower and upper parts of processus angularis, and the anterior and posterior halves of the condyle neck. A sample was also obtained from fossa mandibularis of the temporal bone. Buffer extracts of powderized samples were studied for acid and alkaline phosphatase, glucuronidase and two types of proteolytic activity. Although the asymmetric manipulation of occlusion caused observable, localized asymmetric growth, the enzyme findings were not consistently asymmetric. Manipulation generally increased all enzyme activities regardless of whether apposition or resorption was involved. The activities of alkaline. phosphatase more consistently showed this pattern, and changes in enzyme activities seemed to be sensitive, reflecting cellular or molecular level of growth. The enzyme activities may also reflect a certain type of after-effect of irritation, or a healing period. The inclusion of several types of enzyme determinations is recommended to complete macroscopic measurements. Subsequent chromatographic and electrophoretic studies are also useful. Key words: Enzymes - Guinea pig - Bone growth Facial skeleton
~I
Ann. Anat. (1992) 174: 441-446 Gustav Fischer Verlag Jena
Introduction In bone growth studies, the amount of growth during a certain time period has generally been determined by linear measurements from the bone itself or from the radiographic picture of the bone. More accurate results have been obtained by using certain markers placed in the bone, such as amalgam shots or pins (Bjork 1955). By using vital staining methods, it is possible to mark the entire calcified bone surface; the new bone growth that happened after staining is discernible and measurable (MassIer and Schour 1951; Hoyte 1956; Baer and Ackerman 1966). These methods are usable when studying appositional bone growth mostly on the macroscopic level. However, the resorption of the bone - a very important part of bone growth studies - is difficult to estimate accurately with these methods. Using histological procedures, it is possible to observe the resorption areas in the bone and on the bone surface, but the estimation of the extent and rate of the process - and the changes in the extent and rate, cannot still be performed with sufficient accuracy. Recently, there have been attempts to utilize histochemical (Nielsen and Magnusson 1981) and biochemical techniques (Copray et al. 1985) in bone growth studies. By measuring activities of certain enzymes related to the growth of bone and cartilage, studies of bone remodeling have been conducted (Reddi and Huggins 1972; Stutzman and Petrovic 1984; Bollen et al. 1989). Measurement of the activities of acid phosphatase and glucuronidase has been assumed to elucidate the bone resorption areas themselves and even the rate of bone resorption (Stutzman and Petrovic 1984). In our previous paper (Isotupa et al. 1992) we elucidated the effect of muscle activity on the growing facial skeleton in the guinea pig. In that study, tissue samples from six
different bone areas were obtained - five from the mandible and one from the skull - after 4, 8 and 12 weeks following the manipulation made when the animals were ten days old. The samples were studied for hydrolytic enzyme activities that broadly represented the proteolytic, phosphorolytic and glucuronolytic capability of the tissues. The rationale behind these enzyme determinations was the need to study whether any of the macroscopic findings described in the previous paper were reflected in the biochemical enzyme measurements. The primary purpose of this study was thus to investigate whether certain enzyme determinations could be utilized to complete other procedures in order to obtain a better idea of growth in the guinea pig facial skeleton.
Materials and methods General study design The animals and the experimental arrangements were explained in detail in our previous paper (Isotupa et al. 1992). Eighteen male guinea pigs were divided into experimental (n = 12) and control groups (n = 6). At the age of ten days the guinea pigs were anaesthetized and the radiocephalograms were obtained as described in the above paper. In the experimental animals, the mouth was propped open and a l-cm incision was made longitudinally in the midline of the palate. The midpalatal suture was cleared of periosteum, and cyanoacrylate glue was injected in the soft sutural tissue and onto the bone surface across the midpalatal suture. The suture was then closed by pressing the flaps against the palatal bones. With the mouth remaining open, a dental handpiece and a carborundum grinding wheel were used to grind away the buccal aspect of the lower molars and the palatal aspect of the upper molars on the right side. The incisors were ground at an angle slanting up toward the left (Isotupa et al. 1992). The dental grinding was repeated under anaesthesia every two weeks. Four experimental and two control animals were killed at 4, 8 and 12 weeks postoperatively. The mandible was detached and bisected. The remaining tissues were carefully removed, after which samples of bone and cartilage were taken for the studies described below.
Fig. 1. Designation of the anatomical sites of the mandible (1 - 3, 5, 6) and the fossa mandibularis of the temporal bone (4), used in enzyme studies.
Preparation of tissue samples for biochemical studies The tissue samples for biochemical studies were obtained from six anatomical sites (five from the mandible and one from the cranium) of all animals. These sites are shown in Fig. 1, and in
Chemical methods The activities of acid and alkaline phosphatase were determined in 0.05 M ~,~-dimethylglutarate buffer (pH 5.0) and in 0.05 M carbonate-bicarbonate buffer (pH 9.9), respectively. The reaction mixtures consisted of 0.3 ml of buffer, 0.1 rn1 of 6.0 mM pnitrophenyl phosphate, 0.15 rn1 of water and of 50!t1 of enzyme. The reaction time was 60 min at 37°C. The activity of ~ glucuronidase was determined in 0.05 M acetate buffer, pH 4.5. The reaction mixtures consisted of 0.3 ml of buffer, of 0.1 rn1 of 6 mM p-nitrophenyl-~-D-glucuronide, of 0.15 rn1 of water, and of 50 !tl of enzyme. The reaction time was 180 min at 37°C. In the above cases the reactions were arrested either with 0.4 ml of 0.1 M NaOH (acid phosphatase and glucuronidase) or with 0.4 ml of 25 mM EDTA (alkaline phosphatase) and the readings were made at 405 nm. The activity of proteolytic enzymes was studied using azocasein and glycyl-L-prolyl-4-methoxy-2-naphthylamide (GPm2NA) as substrates. In the former case, the method was used as presented by Seppa and Jarvinen (1978),· and in the latter procedure the enzyme activities were determined as described previously (Makinen and Makinen 1978). The protein was measured according to the Pierce (Rockford, Illinois, USA) Micro BCA Assay Reagent. If not otherwise mentioned, all chemicals were obtaind from Sigma. The water used in the studies was prepared with a Millipore-Milli-Q system and had a resistivity of 18 MQm- 1 •
442
exhibited about two times higher acid phosphatase activity than the controls. The glucuronidase activities followed the All extracts were subjected to SDS-PAGE with PhastGel gradient same pattern, although the 12-week groups did not differ as media (8 - 25 % ), using Coomassie Blue in staining. much as in the case of acid phosphatase (the increase was about 35 % in favor of the treated animals). The activities of alkaline phosphatase also followed this pattern, but the 12Results week old treated animals exhibited about 6-fold higher activities than the controls. The enzymes hydrolyzing azocaAs expected, the specific enzyme activities measured from sein showed up to 3-fold difference in favor of the treated the left and right sides of the control animals (two in each animals at the 4-week stage, but this difference leveled off age group, providing a total offour samples from each ofthe six anatomical sites in each group) were very similar (Fig. later. Enzymes hydrolyzing GPm-2NA showed no signifi2). Therefore, the means of these values were used below to cant differences between groups at any time point. It was difficult to observe any correlation between represent the enzyme activities of untreated controls. In the enzyme activities and macroscopic measurements. Howtreated animals (n = 2 in the two-week group; n = 4 in the ever, the right molars were initially ground leaving only a eight- and 12-week groups), the activities of the left and light contact between mandibular and maxillary teeth. When right sides were first compared in order to elucidate the the treated animals wore out all interferences of occlusion correlation between asymmetric growth and enzyme brought about by the grinding procedure, the right molars activities. When the control enzyme activities in each age obviously grew faster with a concomitant, increased pressure group were subsequently compared with those of the treated in the right TMJ compared with the left one. Macroscopic animals, the means ofthe specific activities from the left and measurements showed that the vertical growth in the right right sides of the latter animals were used. Condylar cartilage. Both control and experimental half of the mandible was indeed slightly lower than in the left groups displayed similar acid phosphatase activity after 4 half, although the enzyme activities did not differ between and 8 weeks. In the 12-week group, the treated animals right and left. Changes in all enzyme activities were in general in the same direction: increases in the activities of alkaline phosphatase, which is normally associated with 3,1 appositional bone growth, were always accompanied by ACID PHOSPHATASE 2.4 5.0 4. increases in the activities of acid phosphatase and glucuronidase, in spite of the fact that the two latter enzymes 2.0 4.0 are frequently associated with osteoclastic functions. 6. Lower part of processus angularis. Acid phosphatase 1.6 2. and glucuronidase activities did not differ appreciably as a 3.0 5. function of time and treatment. Alkaline phosphatase 1.2 1. activities tended to be higher in the treated animals after 2.0 0.8 eight and 12 weeks. Of the proteolytic activities, those determined with azocasein were about 2.5 times higher in 1.0 2. (/) 0.4 the treated animals compared with controls, in all age Q) 5. :;:::; 6. groups. All activities determined with GPm-2NA were os: 0 0 on similar. 12 4 8 4 8 12 « The lower part of the processus angularis represents the 0 ;;::: insertion of the m. pterygoideus med., which appeared to be AZOCASEIN ALKAUNE ° 3. 0Q) 2.5 0.2 PROTEASE PHOSPHATASE strongly affected by the occlusal deviation brought about by a. the dental grinding. In extreme cases, hyperactivity led to C/J 6. 2.0 .16 such a narrowing of the mandible that the entire processus 1. angularis was resorbed. The enzyme studies did not provide 5. 1.5 a logical, consistent result with regard to asymmetry, despite the retarded vertical growth of the ramus observed macro2. scopically in the mandible. However, the manipulation of 1.0 .08 occlusion did exert an exceptional effect on enzyme activities. While manipulation generally induced elevated 0.5 enzyme activities in the tissues studied, the lower part of processus angularis displayed reduced activities of acid o L..-.l...-_ _L - _ - - - l phosphatase and glucuronidase. These observations should 12 4 12 4 8 8 indicate increased de novo formation of bone. This may TIME (weeks) quite well have been the case at the time of tissue sampling, when the appositional growth of the condylar area had Fig. 2. Specific enzyme activities of six anatomical sites of control animals given as a function of animal age. Each experimental point slowed down because of grinding and the consequentional represents 12 half-jaws obtained from six animals. pressure of the condyle against the fossa. Therefore, sub-
Electrophoresis
443
stitutional vertical growth most likely took place in the lower edge of the ramus and on its lateral surface. Upper part of processus angularis. The activities of acid phosphatase were higher in the treated animals, especially in the 8-week group where a 3-fold increase compared with controls was observed. In the glucuronidase determinations, the 8-week group showed a large 5-fold difference while the 4- and 12~week groups and controls were essentially the same. In alkaline phosphatase determinations, the 4-week groups showed no difference, whereas after 8 and 12 weeks there was a clear 4- and 3-fold difference, respectively, in favor of the treated animals. The treatment increased the activities with both proteolytic enzyme substrates, even more so when GPm-2NA was used as substrate. The upper part of processus angularis, which represents an area of appositional growth, exhibited increased activity of enzymes which are normally involved in the growth and degradation of bone. The macroscopic measurements showed that the activation of m. pterygoideus was followed by medial bending of the angular process. In extreme situations this led to the resorption of the entire process. Another important observation was the slight retardation of the vertical growth of the process in the treated animals compared with controls. If acid phosphatase and glucuronidase can be assumed to be associated with bone resorption, the increase in the activities of these enzymes in the treated animals was thus expected and logical. The increased activity of alkaline phosphatase could be explained in terms of the resorption of the medial side of the processus angularis being followed by concomitant growth of bone on the lateral side. Although the final result was an angular process that was lower than normal, it is possible that its maintenance required accelerated appositional growth. Fossa mandibularis. The treated animals showed after 4 and 8 weeks a clear 2.5- and 2-fold increase, respectively, in the acid phosphatase activities compared with controls. In the 12-week old animals this trend had leveled off, reflecting most likely the fact that in younger animals the growth processes were more active than in the older ones, especially in the temporal portion of the temporomandibular joint. The glucuronidase pattern followed that of acid phosphatase. Activities of alkaline phosphatase showed a 2-fold increase in favor of the treated animals after 8-week-treatment, but the groups did not differ at 4- and 12-weeks. The activities tested with azocasein did not differ at all, while those tested with GPm-2NA showed, after 4 weeks'-treatment, an about 2-fold increase compared with controls. These tissue samples were the largest in size studied in this analysis, and contained both cartilage and bone. The measured enzyme activities were generally very high, a finding also observed in the condylar cartilage. The differences between the enzyme activities of the latter and fossa mandibularis may have resulted from the presence of bone in the fossa. The asymmetric manipulation of occlusion was not reflected in enzyme activities. However, manipUlation caused a general increase in the activities, the degree of which varied between the right and left sides. Manipulation also increased the enzyme activities more clearly in younger
animals, whereas the corresponding activity increase in the condylar cartilage was observable first in the older age groups. The reason for this finding remains to be elucidated. Anterior half of the condyle neck. Acid phosphatase activities were low in the anterior region of the condyle neck, but the treated animals always displayed higher activities than the control animals. In the glucuronidase determinations, there was a strong, 5-to 6-fold increase after 8 weeks in the treated animals, but the 4-week and the 12week groups did not differ. Alkaline phosphatase exhibited a clear 6-fold increase after 8 weeks'-treatment, and an about 2-fold increase after 12 weeks' treatment. Azocasein hydrolysis was about 50 % higher in all treated animals. The enzymes hydrolyzing GPm-2NA showed the highest activities - about 3.5-fold compared to controls - after 12 weeks of treatment. ' In normal growth, the anterior half of the condyle neck represents an area of resorption. In the enzyme activities this was seen with increasing age of the animals in the form of decreased activities of alkaline phosphatase and elevated activities of glucuronidase. The specific enzyme activities were generally low, indicating the involvement of smaller growth changes than were expected in the growth area. It should indeed be noted that, in the treated group, the activities of both enzymes increased, but the balance was still in favor of resorption. Posterior half of the condyle neck. The posterior region of the condyle neck also showed low acid phosphatase activity, but the treated animals exhibited 3- to 4-times higher activity than the controls at all ages. The activities of glucuronidase increased as a function of age, but they did not differ significantly between the treatment and the control groups, except in one aspect: the 12-week-old treated animals showed a clearly lower activity than the controls. This marked one of the very few exceptions where the control tissues showed an elevated enzyme activity in this study. The activity of alkaline phosphatase followed the same pattern: the controls exerted, as an exception from the general rule, about 3-fold higher activity than the treatment group after 4 weeks (after 8 and 12 weeks the treatment had caused an increase in enzyme activities, in line with most of the cases mentioned above). Azocasein hydrolysis showed an increase in treated animals at all ages, but the differences from controls were not significant. All activities tested with azocasein increased as a function of animal age. The activities tested with GPm-2NA also increased as a function of age, and were about 3 times higher in animals treated for 12 weeks than in the controls; after 4 and 8 weeks the activities were relatively similar. It is known that when the mandible grows in length, the posterior half of the condyle neck represents an area of appositional growth, although the longitudinal growth of the mandible and the skull is relatively minor in the guinea pig compared with the rat. The low enzyme activities were expected as a result of the dominant proportion of mature, mineralized bone in the samples studied. Therefore, the decrease in the activities of alkaline phosphatase with increasing age may be understandable. The experimental
444
5.0
3
2 • Control
4.0
o Experimental
3.0
(/)
2.0
Q)
+:: .>
1.0
0
0
+::
4
0
;;::
Fig. 3. Specific activities of alkaline phosphatase measured in six anatomical sites of control and treated animals. In the experimental group, each point represents four half-jaws at four weeks, and eight half-jaws at eight and 12 weeks. The number of samples studied in the control group was four at all time points.
·0 Q)
0-
W
4.0
8
12
8
4
12
8
12
6
5
4
4
3.0 2.0 1.0
o '---'-__--'-__........_ 4
12
8
1.--1-_ _-1-._ _- - 1 . _ L--....L.._ _-I-._ _- - I . _
4
8
12
4
8
12
TIME (weeks)
arrangements rendered somewhat elevated enzyme activities possible in this area of the condyle neck, although this elevation did concern more significantly the hydrolysis of GPm-2NA in the 12-week-treated group. A graphical representation of the effect of the treatment on the activities of alkaline phosphatase is shown in Fig. 3. Electrophoresis. The aqueous extracts representing all six anatomical sites were subjected to SDS-PAGE followed by Coomassie Blue staining. The purpose of these experiments was to study whether any of the macroscopic and enzymatic findings were reflected in the electrophoretograms of the soluble, stainable proteins present in the extracts. These experiments were performed considering the fact that most proteins derived from the above tissue samples, and which were soluble and stainable under the present conditions, should be similar and represent various dominant intracellular and extracellular proteins and peptides whose relative levels in the extracts studied could be assumed to remain virtually unchanged regardless of the treatment carried out. Likewise, it was assumed that the concentrations of the enzymes hydrolyzing the above mentioned substrates were too low in the tissue extracts to be demonstrated as stainable bands in the present electrophoretograms. In spite of these reservations, it was possible to make a number of consistent observations which are summarized below. Condylar cartilage. The control animals showed no significant differences as a function of age; the 4-, 8- and 12week-old animals gave a similar electrophoretic pattern. Also, the same protein pattern was present in the samples obtained from the left and the right halves of the skull in the treated animals. Comparison between treated and control animals showed the following general trend. There were differences in the protein pattern after four weeks. After eight weeks those differences had partly disappeared and after 12 weeks there were no observable electrophoretic differences between controls and treated animals. The above
differences concerning the 4-week-old animals were related to the presence of bands in the two extreme ends of the gels, indicating that soluble proteins with relatively low and high molecular weights had undergone concentration changes because of the treatment. In the 4-week-old treated animals, one band, representing a small-molecular protein, was consistently absent. Also, one high-molecular weight band was missing. Both bands were clearly present in the electrophoretograms of control samples. Lower part of processus angularis. The treated animals showed a difference between the left and the right sides: there was a high-molecular weight band present in the 4week-old animals after treatment. After eight weeks the intensity of this band was clearly decreased, and after 12 weeks there was no longer any difference between the left and right sides. Fossa mandibularis. These samples showed an extra band in the small-molecular weight region of the gel, after four weeks of treatment in the right side of the mandible; the intensity of this band was clearly higher than in the controls. There were no differences between the 8-week-old animals. After 12 weeks, the left side samples showed an extra band in the low-molecular weight region of the gels. It is obvious that more detailed chromatographic studies using sensitive techniques will further elucidate metabolic and molecular aspects of bone growth in guinea pigs.
Discussion The guinea pig was chosen as the experimental animal because of its size; when the right and left sides were to be studied separately, the tissue sample had to be large enough for biochemical studies and the preceding preparations. On the other hand, the large number of anatomical sites involved, and the resulting numerous enzyme determinations in'volving laborious pulverization and extraction procedures,
445
stipulated the number of animals that could be investigated. The tissue samples were obtained from areas where the putative alterations of muscle activity, caused by manipulation of occlusions , could induce growth changes. The purpose of the enzyme assays was thus to study whether the activity levels of acid and alkaline phosphatase, glucuronidase and certain proteinases could be used as additional tools in bone growth research. The enzyme activities that were associated with normal growth (determined with control animals), showed a relatively good "symmetric" result, i. e., both the right and left sides expectedly yielded similar specific activities in the same animal. There were differences between individual animals even in the control group, which also was expected. However, because the deviations between the size of individual bones were rather ,small at any given age of the animals, one could assume the changes in enzyme activities to be subtle and perhaps short-term in duration. A change in specific activity may not~ therefore, reflect a measurable change in bone growth. , Excluding a few exceptions, the enzyme activities of the treated animals were higher than those of control animals, indicating a clearly accelerated turnover of enzymes that are normally associated with the growth and degradation of bone. Because of the obviously quite diversified metabolic roles of the enzymes studied, it is not justifiable to draw conclusions about bone growth basing on one enzyme alone. Focusing on several enzymes with different metabolic functions, a somewhat better picture of bone growth could be obtained. At best, however, even these methods cannot offer more than a glimpse into the prevailing, and not so much into the past, situation. The asymmetric manipulation of occlusion elicited observable, although small, localized asymmetric growth. The enzyme findings from the corresponding areas were not consistently asymmetric. One has to observe, however, that the macroscopic measurements indicated growth that had already taken place, while the enzyme activities reflected the nature of ongoing growth. The asymmetric manipUlation was expected to exert a more long-term, even a continuous effect, forcing the mandible to grow asymmetrically to the left. The guinea pigs were able, however, to grind the interference away in about two days, and the effects remained obviously too short-term considering the present enzyme study. It is possible, therefore, that the determination of enzyme activites - two weeks following the last grinding - took place too late to reflect the asymmetric growth described in the accompanying paper. The present enzyme activities may rather reflect a certain type of after-effect of irritation, or a healing period, when asymmetric observations may be possible. On the other hand, if the enzyme activities had been performed soon after treatment, they would perhaps have reflected the direction of ongoing growth which we would have been unable to observe, however, by measuring appositional growth. Based on the present study, it appears to be possible to elucidate bone growth by means of enzyme activity determinations, but with certain reservations . A complicating factor concerns the relatively high individual differences. Furthermore , a relatively mild irritation could cause strong changes
in enzyme activities. All enzyme activities were increased regardless of whether apposition or resorption was to be the dominant event in the area. However, changes in enzyme activities seemed to be sensitive and reflected the microscopic or molecular level of growth (connoting that regular histological procedures would represent the macroscopic level where apposition and resorption had been observed, as reported in the accompanying paper). This indicates that only a slight and transient irritation may induce changes in enzyme activities, and that a real observable growth change presumed a long-term force and irritation. If only a minor irritation induced alterations in enzyme activities, it is possible that the biologic changes of bone, involved in orthodontic practice, could be brought about by really weak forces , provided that such forces are sufficiently long-term.
References Baer MJ, Ackerman LL (1966) A longitudinal vital staining method for the study of apposition in bone,. In : Studies on the anatomy and function of bone and joints. Evans FG (ed) Springer, New York, pp 81-92 Bjork A (1955) Facial growth in man, studied with the aid of metallic implants . Acta Odont Scand 13: 9-34 Bollen AM, Makinen KK, Makinen PL, Carlson DS (1989) Collagenolytic and phosphatase activity in the rat mandible after functional protrusion. Archs Oral BioI 34: 267-273 Copray JCVM, Janssen HWB, Duterloo HS (1985) Effect of compressive forces on phosphatase activity in mandibular condylar cartilage of the rat in vitro . J Anat 140: 479-489 Hoyte DAN (1956) A histological study of bone growth using Alizarin red S. J Anat 90: 585 Isotupa K, Carlson DS , Makinen KK (1992) Influence of asymmetric occlusal relationships and decreased maxillary width on the growth of the facial skeleton in the guinea pig . Anat Anz 174: 447-451 Makinen KK, Makinen PL (1978) Purification and characterization of two human erythrocyte arylamidases preferentially hydrolysing N-terminal arginine or lysine residues. Biochem J 175: 1051-1067 MassIer M, Schour I (1951) The growth pattern of the cranial vault in the albino rat as measured by vital staining with Alizarin red S. Anat Rec 110: 83-101 Nielsen R, Magnusson BC (1981) Enzyme histochemical studies of induced heterotopic cartilage and bone formation in guinea pigs with special reference to acid phosphatase. Scand J Dent Res 89 : 491-498 Reddi AH, Huggins C (1972) Biochemical sequences in the transformation of normal fibroblasts in adolescent rats. Proc Nat! Acad Sci 69: 1601-1605 Seppa HE, Jarvinen M (1978) Rat skin main neutral protease: purification and properties. J Invest Dermatol 70: 84-89 Stutzman J, Petrovic A (1984) Human alveolar bone tum-over rate. A quantitative study of spontaneous and therapeutically-induced variations. In : McNamara, Jr. JA , Ribbons KA (eds) Malocclusion and Periodontium. Craniofacial Growth Series, Center of Human Growth and Development, Ann Arbor, Michigan , pp 185-212 Accepted October 30, 1991
446