11
Collagen ReI. Res. Vol. 111988, pp. 11-21 Collagen Remodelling in Rat Periodontal Tissues: Compensation for Precursor Reutilization Confirms Rapid Turnover of Collagen lARO SODEK and lACK M. FERRIER Medical Research Council Group in Periodontal Physiology, Room 4384 Medical Sciences Building, Faculty of Dentistry, University of Toronto, Toronto, Ontario MSS lA8, Canada. Abstract Measurement of collagen turnover is complicated by the reutilization of isotopic precursors used to label the collagen. In an earlier study a novel approach was used to circumvent the problems of precursor recycling and unusually short half-lives were determined for collagen in adult rat periodontal tissues (Sodek, 1977). To verify these results we have used an alternate procedure devised by Poole (1971) in which the decay profile for the radiolabelled protein is corrected in accordance with the decay of the radiolabelled precursor. In this manner real half-lives for mature, neutral salt-insoluble collagen were determined as 3 days in the molar periodontal ligament, 6 days in the continuously erupting incisor ligament and approximately 10 days in the lamina propria of the gingiva, compared to apparent half-lives for these tissues of 6, 12 and approx- imately 20 days, respectively. The values calculated for actual half-lives are, therefore, approximately two-fold faster than values determined without compensating for reutili- zation, a difference that is in agreement with other protein turnover studies in which the effects of precursor reutilization have been measured. Although the real half-lives determined in this study indicate turnover rates for the periodontal tissues that are slightly slower than reported previously, the relative differences between the tissues in the rates of collagen turnover are similar. Moreover, the study confirms the existence of a remarkably high rate of collagen remodelling in these tissues. Key words: gingiva, periodontal ligament, precursor recycling, collagen turnover. Introduction The majority of studies on collagen turnover in various tissues have been carried out using rats. An advantage of this has been that comparisons between tissues can be made. However, different radio labelled precursors have been used, some of which are not sufficiently specific to collagen and require that the collagen be carefully purified. In the earliest studies of Neuberger (Neuberger et aI., 1951; Neuberger and Slack, 1953) e 4 C]- glycine was used as a precursor and the results obtained indicated that collagen turnover in adult rat skin and tendon was very slow but somewhat faster in liver and some bones. In contrast, collagen of rapidly growing animals was found to turn-over at an appreci-

Collagen Remodelling in Rat Periodontal Tissues: Compensation for Precursor Reutilization Confirms Rapid Turnover of Collagen

Embed Size (px)

Citation preview

Collagen ReI. Res. Vol. 111988, pp. 11-21

Collagen Remodelling in Rat Periodontal Tissues: Compensation for Precursor Reutilization Confirms Rapid Turnover of Collagen

lARO SODEK and lACK M. FERRIER

Medical Research Council Group in Periodontal Physiology, Room 4384 Medical Sciences Building, Faculty of Dentistry, University of Toronto, Toronto, Ontario MSS lA8, Canada.

Abstract

Measurement of collagen turnover is complicated by the reutilization of isotopic precursors used to label the collagen. In an earlier study a novel approach was used to circumvent the problems of precursor recycling and unusually short half-lives were determined for collagen in adult rat periodontal tissues (Sodek, 1977). To verify these results we have used an alternate procedure devised by Poole (1971) in which the decay profile for the radiolabelled protein is corrected in accordance with the decay of the radiolabelled precursor. In this manner real half-lives for mature, neutral salt-insoluble collagen were determined as 3 days in the molar periodontal ligament, 6 days in the continuously erupting incisor ligament and approximately 10 days in the lamina propria of the gingiva, compared to apparent half-lives for these tissues of 6, 12 and approx­imately 20 days, respectively. The values calculated for actual half-lives are, therefore, approximately two-fold faster than values determined without compensating for reutili­zation, a difference that is in agreement with other protein turnover studies in which the effects of precursor reutilization have been measured. Although the real half-lives determined in this study indicate turnover rates for the periodontal tissues that are slightly slower than reported previously, the relative differences between the tissues in the rates of collagen turnover are similar. Moreover, the study confirms the existence of a remarkably high rate of collagen remodelling in these tissues.

Key words: gingiva, periodontal ligament, precursor recycling, collagen turnover.

Introduction

The majority of studies on collagen turnover in various tissues have been carried out using rats. An advantage of this has been that comparisons between tissues can be made. However, different radio labelled precursors have been used, some of which are not sufficiently specific to collagen and require that the collagen be carefully purified. In the earliest studies of Neuberger (Neuberger et aI., 1951; Neuberger and Slack, 1953) e4C]­glycine was used as a precursor and the results obtained indicated that collagen turnover in adult rat skin and tendon was very slow but somewhat faster in liver and some bones. In contrast, collagen of rapidly growing animals was found to turn-over at an appreci-

12 J. Sodek and J. M. Ferrier

ably higher level, consistent with the extensive reconstruction of collagen fibres in growing tissues. Similar results were obtained subsequently from analyses of radiolabel­led hydroxylysine which is more specific to collagen (Kao et ai., 1961a; 1961b). While results of these and other studies (reviewed by Woessner, 1968) are useful for compara­tive purposes, the actual half-lives obtained are highly inaccurate because compensation for the effects of recycling (reutilization) of the radiolabelled precursor was not made.

The significance of precursor reutilization, the extent to which it can effect measure­ment of protein turnover, and approaches that have been used to minimize and avoid the effects of re-utilization have been reviewed in detail (Waterlow et ai., 1978; Zak et ai., 1979). Ideally, a precursor that is not reutilized should be used. For collagen, incorpora­tion of eS02l into hydroxyproline has been used (Jackson and Heininger, 1974; 1975; Molnar et ai., 1986) and has provided important data on collagen turnover in rat skin. However, the cost of the isotope combined with the amount of sample required for analysis has precluded its general use. Attempts to minimize reutilization have been made by administration of large amounts of unlabelled precursor beginning immediately after the injection of isotope (Nimni et ai., 1967; Nissen et ai., 1978). This approach can reduce the effects of recycling considerably if the unlabelled precursor is given for a number of days, but it is not effective for tissues in which collagen is remodelled rapidly, nor does it appear that the effects of massive doses of precursor on protein metabolism have been evaluated. Alternatively, continuous perfusion (Robins, 1979; Palmer et ai., 1980) of radiolabelled precursor and "pool expansion" (Laurent, 1982; McAnulty and Laurent, 1987) have been used to measure collagen synthesis. Turnover can be deter­mined from the fractional synthesis rate even in the presence of growth. These approaches can avoid the problems of reutilization and have the advantage that the specific activity of the precursor can be measured.

From radioautographic studies and the effects of avitaminosis C, it has been suggested that protein turnover, and collagen turnover in particular, is rapid in the periodontium, an organ comprising both hard and soft connective tissues that functions in tooth support (reviewed by Sodek, 1983; 1987). To measure collagen turnover directly in these tissues, a novel approach has been used in which re-utilization of precursor was minimized by analysing the synthesis of both newly synthesized and mature, cross-linked collagen (Sodek, 1976; 1977). From this study, half-lives of mature collagen in molar periodontal ligament, gingiva and alveolar bone of adult rat were calculated at 1, 5 and 6 days, respectively, compared to 15 days in skin. Moreover, in a subsequent study the half-life of collagen in the continuously erupting incisor was estimated at 3 days (Sodek, 1978). In an attempt to validate these findings we have used a different approach, previously described by Poole (1971) for measuring turnover of liver proteins, to calculate real half-lives for mature collagen in the same periodontal tissues. In this method the decay curve for the radioactive precursor is used to generate decay curves for the radioactive proteins to which the experimental decay curves are fitted.

Materials and Methods

Thirty male CBL Wistar rats weighing 200 ± 15 g were given 0.5 f..\,Ci/g body weight eHl-proline (NET-323, New England Nuclear Corp., Boston, MA) in 0.5 ml phosphate­buffered saline (PBS) by a single intraperitoneal injection. At various time intervals over a 24-day period, animals were anaesthetized with ether and approximately 250 f..\,l of blood withdrawn by cardiac puncture. The animals were then killed by cervical disloca-

Collagen Remodelling in Periodontium 13

tion and the mandibles immediately dissected out and placed in ice-cold PBS. The attached gingivae were carefully dissected from the buccal and lingual aspects of the first and second molars under a dissecting microscope using a # 11 scalpel blade. The first and second molars were then extracted using a #21 gauge syringe needle as an elevator, and the periodontal ligament scraped gently from the apical two-thirds of the roots, using a #23 scalpel blade. The incisor teeth were extracted after carefully splitting the associated mandibular and alveolar bone with a scalpel. Ligament was scraped gently from the middle third of each tooth, care being taken to avoid completely the proximal formative and distal degradative ends of the root.

Determination of Blood Proline Specific Radioactivity

Samples of fresh blood, 250 [.11, were mixed with an equal volume of PBS containing 50 Units/ml heparin and 10 nmoles norleucine. The serum proteins were removed by precipitation with 6% (w/v) trichloroacetic acid (TCA). After centrifugation for 5 min at 10,000 g on a Microfuge (Eppendorf, Brinkman Instruments, Toronto, Canada) the supernatant, containing the proline, was collected and extracted three times with equal volumes of diethyl ether to remove the TCA. The aqueous phase remaining was dried under vacuum and then dissolved in 100[.11 0.2M sodium citrate buffer, pH2.2 in preparation for amino acid analysis (see below).

Determination of Hydroxyproline Specific Radioactivity in Mature Collagen

The samples of tissue were immediately transferred to individual 1.5 ml plastic mi­crofuge tubes (Eppendorf) containing 1.0 ml ice-cold PBS. After centrifuging at 10,000 g, the tissues were washed with a further two aliquots of PBS and were then extracted three times with 1.0 ml aliquots of 1 M sodium chloride in 50 mM Tris-HCI, pH 7.4 to remove the newly-synthesized collagen. Each tissue residue was then hydroly­zed in 6 N HCI for 24 hat 110°C in a sealed evacuated hydrolysis tube. After hydrolysis, the tube was opened and the hydrolysate taken to dryness under vacuum and in the presence of sodium hydroxide pellets. The residues were dissolved in 250 [.11 of citrate buffer, pH 2.2, containing 25 nmoles norleucine and 5,000 dpm [14C]-glycine as internal standards. The amino and imino acids in the blood and hydrolyzed tissue samples were separated on a Beckman 121M analyzer system (Beckman Instruments Inc., Palo Alto, CAl, essentially as described previously (Sodek, 1977) with the analytical runs extended to include the determination of norleucine. Quantitation of the recoveries of imino acids and their associated radioactivity were based in each case on the recovery of the respective internal standards and specific radioactivities determined accordingly.

Results

The decay of hydroxyproline specific radioactivity in the salt-insoluble collagen of molar and incisor periodontal ligaments and gingiva over the 24-day period of study is shown in Fig. 1. After the specific radioactivity had peaked, the subsequent decay was reasonably linear. From the linear decay apparent half-lives of 6,12 and approximately 20 days were calculated. If all the incorporation of radiolabelled precursor had occurred

14 J. Sodek and]. M. Ferrier

100,------------------------, A

10

1 100

Q) 8 '0 E c E Q.

~ ~ 10 :~ u <I: (.) ;;:: '0

C1> Q. en 1

100 C

10

1+----------.----------.---~

o 10 20

Time (days)

Fig. 1. Comparison of Decay Profiles for [3H]-proline in Blood and [3H]-hydroxyproline in Rat Periodontal Tissues. The individual hydroxyproline specific radioactiviry values for molar (A) and incisor (B) periodontal ligament and gingiva (C) are shown for each time point (e - e); whereas the proline specific radioactiviry is indicated by a single line joining the averaged data points (. - .). Note the fall in [3Hl-hydroxyproline specific radioactiviry compared to the [3Hl-proline for each tissue.

before the first specific radioactivity measurement, and if subsequently the decay of hydroxyproline specific radioactivity could be described by the differential equation:

dS~(t) = -kSA(t),

where SA(t) is the specific radioactivity of the hydroxyproline in the collagen and k is its rate of turnover, then the half-lives would represent actual half-lives. However, as can be

Collagen Remodelling in Periodontium 15

seen in Fig. 1, the specific radioactivity of the free proline in the blood over the period of study is significant, and in the case of the molar periodontal ligament the hydroxyproline specific radioactivity follows the proline radioactivity decay quite closely. Consequently, any assumption that there is no incorporation of precursor into collagen after the initial labelling period would not be valid. There is, in fact, continuous incorporation, which is occurring at a high level particularly in the molar ligament. In each tissue it is apparent that radiolabelled collagen lost by degradation would be replaced by collagen with an appreciable specific radioactivity.

Because of the high degree of precursor reutilization the differential equation shown above was modified to compensate for the effect of reutilization as follows:

dS~(t) = k[SB(t)-SA(t)],

where SB(t) is the specific radioactivity of the free precursor. This equation was numeri­cally integrated using values for SB(t) obtained by linear interpolation from the experi­mentally-determined decay curve shown in Fig.2a. From this equation a series of theoretical curves were generated with various k (turnover) values for each tissue, as shown in Fig. 2b-d. The solid lines in each case represent curves generated from two k values that provide a close fit to the experimental data for periodontal ligament, incisor periodontal ligament and gingiva respectively. The best fit values translate into half-lives of 3,5.7 and 10 days and turnover times (the time interval equivalent to the time taken to completely replace all the collagen within the tissue) of 4.3,8.1 and 14.2 days, respec­tively. It should be noted that the k values and half-lives describe turnover if the system is at physiological steady-state, with the rate of synthesis equal to the rate of degradation. If the system is not at steady-state, the k values will apply to synthesis. This is because it is the rate of synthesis for a given pool (and not the rate of degradation) that determines the rate at which the specific radioactivity of that pool changes.

From the decay profiles of collagen in the various tissues shown in Fig. 2, it is also evident that the maximal specific radioactivity for hydroxyproline is 42 dpm/nmole in the molar periodontal ligament; whereas the maximal value for incisor ligament is 28 dpm/nmole and gingiva approximately 17 dpm/nmole. Further, the maximum specific radioactivity is reached in 1-2 days in the molar periodontal ligament, approxi­mately 4 days in the incisor periodontal ligament, and approximately 8 days in the gingiva. Notably, the relative differences in these parameters approximate the differences in turnover rates for collagen in these tissues and is consistent with the observation of Poole (1971 ) on liver proteins.

In an exact mathematical calculation, based on the preceding equation, the peak of the SA profile must fall on the SB profile. Since our numerical integration was done at intervals of 1 day, our calculated SB peaks came within 0.5 days of the interpolated curve for SB. In cases such as the incisor periodontal ligament, where the actual SA peak is considerably further out in time from the SB curve, the apparent discrepancy can be explained mathematically by assuming the existence of an intermediate pool between SB and SA, with the specific radioactivity of this intermediate pool falling off more slowly with time than SB. The existence of intermediary pools from which precursor is taken for protein synthesis has been demonstrated in a number of studies that are discussed by Waterlow et al. (1978).

16 J. Sodek and J. M. Ferrier

100~-----------------------. r-------------------------. A B

10 Q) (5 E c: E 0. ~ :: :~ 100 ti C D c(

.g '(3 Q) 0. en

10

1+----------.-----------.----~ ~----~-----.----~----~--~

o 10 20 0 10 20

Time (days)

Fig. 2. Semi-logarithmic Plot of [JH1-proline and [3Hl-hydroxyproline Data and Calculated Decay Curves. A) Specific radioactivity of free [JH1-proline in blood as a function of time. Error bars indicate standard error of the mean. B-D) Specific radioactivity of protein-bound eHl-hydroxyproline in molar periodontal ligament, incisor ligament and gingiva (.). Error bars on the measured values indicate standard error of the men. The calculated values (0 , L::.) were obtained for synthesis rate of k = O.200/day (t = 3.S days) and k = O.2S0/day (t = 2.8 days), respectively, in molar ligament; for synthesis rate of O.IOO/day (t = 6.9 days) and k = O.IOO/day (t = 3.5 days), respectively, in incisor ligament; and for synthesis rate of O.075/day (t = 9.2 days) and k = O.IOO/day (t = 6.9 days), respectively, for gingiva.

Discussion

The turnover of collagen in connective tissues provides insights into the dynamic aspects of connective tissue function. Although the need for extensive and rapid re­modelling of collagen in tissues that are growing or undergoing repair is readily appreci­ated, collagen in tissues of mature, adult animals has been considered to be essentially inert (Kivirriko, 1971). This view has evolved from numerous studies on collagen turnover in which the problems associated with precursor reutilization were either not appreciated or were ignored. Precursor reutilization is currently recognized as the major source of errors in the determination of protein turnover and has been the prime reason for the development of specialized procedures for studying protein turnover (reviewed

Collagen Remodelling in Periodontium 17

by Waterlow et aI., 1978; Zak et aI., 1979). The effect of precursor reutilization is more profound on proteins such as collagen, that have relatively low rates of turnover compared to cellular proteins.

In common with other amino acid precursors, reutilization of proline occurs because of a number of factors. Proline is efficiently resorbed in the kidney and little of it is deaminated or converted into other metabolites, so that radiolabelled proline remains available for protein synthesis over relatively long time periods. The recycling of proline is further accentuated by the continual return to the blood of radiolabelled proline that has been incorporated into a variety of proteins with different turnover rates that are being degraded. Even more efficient reutilization can occur if the radiolabelled proteins are degraded within the same cells that are synthesizing collagen.

That reutilization of radiolabelled proline is affecting the decay of radiolabelled hydroxyproline in this study is clearly seen from the similarities in the specific radioac­tivities of the precursor and the collagen hydroxyproline, particularly in the periodontal ligament. Apparent half-lives, determined from the data given in Fig. 1 are similar to several other studies in which reutilization of precursor was not considered (Orlowski, 1978; Taverne et aI., 1986). By compensating for the effects of precursor reutilization, using the approach of Poole (1971), the calculated half-lives are reduced significantly. These half-lives confirm the rapid turnover of collagen in periodontal tissues, indicated in previous studies (Sodek, 1976; 1977; 1978). Although the actual half-lives are not quite as fast as determined in these earlier studies, the current values probably represent minimal turnover rates since compensation for precursor is calculated from the blood values rather than from the tissue values. The tissue values are likely to be higher than plasma during the phase of declining blood specific radioactivity (Reddi, 1986). Further, intracellular recycling of labelled proline can occur through the extensive phagocytic degradation of collagen, and probably other matrix proteins, by fibroblastic cells in these tissues (Svoboda et aI., 1981; Beertsen and Everts, 1977). Ideally, to provide a more accurate evaluation of recycling, the specific radioactivity of proline tRNA should be measured for each tissue. However, currently this has not been possible because of the small amounts of periodontal tissue that can be dissected.

The hydroxyproline decay profiles obtained for these tissues are similar to those obtained by Poole (1971) for [3HJ-leucine labelled liver proteins. As shown in his study, the maximal specific radioactivity obtained and the time period required to reach the maximum could be related to the turnover rate. Although there are differences between these studies in that leucine is an essential amino acid whereas proline is non-essential, it was shown in Poole's study that proteins with an apparent half-life of 3-4 days have real half-lives of 0.1-2 days; this is similar to the differences found here for molar ligament. Poole's study also points to a progressive decrease in the accuracy of calculating real half­lives as the apparent half-life increases beyond 4 days. Thus, even the compensated value for collagen turnover in the gingiva, and to some degree the incisor ligament, must be considered only as a good estimate. However, the relative rates of turnover in the different tissues are consistent, not only with previous studies, but also with the relative specific radioactivity maxima and the time taken for the maxima to be reached.

A recent study by Imberman et al. (1986) has also shown high turnover of collagen in rat periodontal tissues. However, despite attempts to avoid recycling of precursor by using a procedure of "pool expansion" whereby the radiolabelled precursor is adminis­tered with a large excess of cold precursor to produce a stable influx of isotope similar to the continuous perfusion method, the half-lives are closer to values obtained without consideration of re-utilization. The discrepancies with our results may reflect problems

2 Collagen 811

18 ]. Sodek and]. M. Ferrier

in applying the pool expansion approach to measuring collagen turnover in periodontal tissues. The method was originally devised to measure collagen synthesis rates (Laurent, 1982; McAnulty and Laurent, 1987). Since a significant amount of the collagen syn­thesized is degraded both intracellularly and extracellularly soon after synthesis, the rate of synthesis of collagen incorporated intoJibrils may not have been measured accurately. This will translate into inaccuracies in the calculation of matrix collagen half-lives. Further, to compensate for growth it has been assumed that the periodontal tissues are growing at a similar rate to skin. This is not valid since the periodontal ligament, once formed, does not increase appreciably in size (McCulloch and Melcher, 1983).

Studies comparing [1~02)' as a non-recycling precursor, with [3H)-proline have shown that reutilization more than doubles the half-life of rat skin collagen (Jackson and Heininger, 1974), as appears to occur in this study and in the studies of Poole (1971). In a more recent study using [180 2) to label hydroxyproline in rat skin collagen, collagen insoluble in 0.5 M acetic acid was found to be stable whereas 45% of the soluble collagen was degraded (Molnar et at, 1986). This indicates that not all of the collagen in growing tissues is remodelled, as has been suggested previously from the studies of Etherington and Bailey (1982). Although there is very little acid-soluble collagen in periodontal tissues (Sodek, 1977) it is conceivable that within these tissues there are collagen fiber systems with different rates of remodelling.

The unusually rapid remodelling of collagen in periodontal tissue is not consistent with the suggestion that collagens in various tissues are remodelled at comparable rates (Nissen et at, 1978). However, the results are consistent with the well-known effects of scurvy and the high incidence of phagosomes containing collagen fibrils in periodontal fibroblasts. Phagocytosis of collagen by fibroblastic cells, which appears to be the major pathway for collagen breakdown during physiological remodelling, is also prominent at sites of wound healing (Ten Cate and Freeman, 1974). Further, stereo logical analyses of the amount of internalized collagen in fibroblasts of periodontal tissues and skin of rats has been shown to be directly related to the relative rates of collagen turnover in these tissues, and also to the cell density of the respective tissues (Svoboda et at, 1980). Thus, the higher rate of collagen remodelling in periodontal ligament is likely to be due, in part, to the high cell density. The rapid remodelling in periodontal ligament is also consistent with other biochemical properties of this tissue (reviewed by Shuttleworth and Smalley, 1984) such as the nature of collagen cross-links (Pearson et at, 1975) and the relatively high (15-20%) proportion of type III collagen (Sodek and Limeback, 1979), features which are characteristic of rapidly growing and healing tissues. Interestingly, periodon­tal ligament also contains relatively high amounts of osteonectin (Wasi et at, 1984), a protein that appears to be identical to SPARC protein and a 43 Kda endothelial cell glycoprotein (Mason et aI., 1986a; Domenicucci et at, 1987) which are found in rapidly remodelling tissues (Mason et aI., 1986b).

The significance of the rapid turnover of collagen in periodontal ligament is not clear (Sodek, 1983; 1987). The lower turnover in the incisor ligament compared to the molar ligament has been suggested to be due to the combination of a rapidly remodelling collagen in the intermediate plexus, which allows movement of the continuously erupt­ing tooth, with the tooth-associated and bone-associated collagen on either side being more stable (Sodek, 1978). Notably, the tooth-associated collagen also undergoes "life­time" turnover whereby collagen synthesized at the formative end of the tooth is ultimately degraded at the incisal end as the tooth erupts. This contribution to collagen remodelling has been avoided in this study by analyzing only the middle third of the ligament. That tooth eruption contributes to the rapid turnover is supported by the

Collagen Remodelling in Periodontium 19

increase in collagen turnover in the molar ligament during "activated eruption" (Kanoza et aI., 1980), but an inherent high rate of turnover is also indicated from the maintenance of a high rate of collagen turnover in non-functional teeth of dogs (Plecash and Bentley, 1982) and rats (Sodek, unpublished data). This inherent high rate of collagen remodel­ling may be important physiologically in the appositional alignment of teeth during normal function and undoubtedly is of importance in orthodontic tooth movement.

The high rate of collagen remodelling in periodontal and other dental tissues has provided a unique opportunity to study several aspects of collagen biosynthesis in vivo. From these studies it has been shown that crosslin king can occur within 30 min of collagen synthesis; that types I and III collagens, but not type V collagen, are remodelled at comparable rates; and that type III collagen precursors are incorporated into cross­linked fibrils (Sodek and Limeback, 1979; Limeback and Sodek, 1979). Subsequently, similar evidence was presented for the cross-linking of type I collagen precursors in a study of the biosynthesis of a1(l)-trimer in vivo (Sodek and Mandell, 1982). The formation of fibrils containing procollagens and procollagen intermediates was subse­quently demonstrated in embryonic skin by electron microscopy using antibodies to procollagen peptides (Fleischmajer et aI., 1981).

Acknowledgements

The authors wish to acknowledge the assistance of R.j.j. Kanoza, S. M. Mandell and L. Kelleher in this project, and also thank Mrs. E. Krissilas and Ms. M. Visconti for preparing the manuscript.

References

Beertsen, W. and Everts, V.: The site of collagen remodelling in the periodontal ligament of the mouse incisor. Anal. Rec. 189: 479-498,1977.

Domenicucci, c., Hofmann, T., Isenman, D., Wasi, S. and Sodek,J.: Purification and properties of osteonectin extracted from fetal porcine calvariae. Submitted for publication.

Etherington, D. J. and Bailey, A. J. : Metabolism of rat skin collagen. The effect of arrested development and subsequent catch-up growth on the stability of the connective tissue fibers . Collagen Rei. Res. 2: 507-522, 1982.

Fleischmajer, R., Timpl, R., Tuderman, L., Raisher, L., Weistner, M., Pedish, J. S. and Graves, P. N. : Ultrastructural identification of amino propeptides of type I and III collagens in human skin. Proc. Natl. Acad. Sci. USA 78: 7360-7364, 1981.

Imberman, M., Ramamurthy, N., Golub, L. and Schneir, M.: An assessment of collagen half-life in rat periodontal tissues: application of the pool expansion approach. J. Periodontal Res. 21: 396-402,1986.

jackson, S. H. and Heininger, J. A.: A study of collagen reutilization using an 180 2 labeling technique. Clinica Chemica A cta 51: 163-171, 1974.

Jackson, S. H. and Heininger, J. A.: Proline recycling during collagen metabolism as determined by concurrent 180 2_ and 3H-labeling. Biochim. Biophys. Acta 381: 359-367, 1975 .

Kanoza, R. J. J., Kelleher, L. , Sodek, J. and Melcher, A. H.: A biochemical analysis of the effect of hypofunction on collagen metabolism in the rat molar periodontal ligament. Arch. Oral Bioi. 22: 655-665,1980.

Kao, K.-Y. T., Hilker, D. M. and McGavack, T. H.: Connective Tissue IV. Synthesis and turnover of protein in tissues of rats. Proc. Soc. Exp. Bioi. Med. 106: 121-124, 1961a.

Kao, K.-Y. T., Hilker, D. M. and McGavack, T. H.: Connective Tissue V. Comparison of synthesis and turnover of collagen and elastin in tissues of rat at several ages. Proc. Soc. Exp. Bioi. Med. 106:335-338,1961b.

20 J. Sodek and}. M. Ferrier

Kivirriko, K. 1.: Urinary excretion of hydroxyproline in health and disease. Int. Rev. Conn. Tissue Res. 5: 93-163,1971.

Laurent, G. J.: Rates of collagen synthesis in lung, skin and muscle obtained in vivo by a simplified method using [3H] proline. Biochem. J. 206: 535-544,1982.

Laurent, G. J., Sparrow, M. P., Bates, P. C. and Millward, D. J.: Turnover of muscle protein in the fowl; collagen content and turnover in cardiac and skeletal muscles of the adult fowl and the changes during stretch-induced growth. Biochem. J. 176: 419-427,1978.

Limeback, H. F. and Sodek, J. : Procollagen synthesis and processing in periodontal ligament in vivo and in vitro. A comparative study using slab-gel fluorography. Eur. J. Biochem. 100: 541-550, 1979.

Limeback, H. F., Sodek, J. and Brunette, D. M.: Nature of collagens synthesized by monkey periodontal ligament fibroblasts in vitro. Biochem. J. 170: 63-71, 1978.

Mason, 1. ]., Murphy, P., Munke, M., Francke, U., Elliot, R. W. and Hogan, B. L. M.: Developmen­tal and transformation-sensitive expression of the SPARC gene on mouse chromosome 11. EMBO}. 5: 1831-1837, 1986a.

Mason, 1. J., Taylor, A., Williams, ] . G. , Sage, H. and Hogan, B. L. M. : Evidence from molecular cloning that SPARC, a major product of mouse embryo parietal endoderm, is related to an endothelial cell 'culture shock' glycoprotein of M, 43,000. EMBO J. 5: 1465-1472, 1986b.

McAnulty, R. J. and Laurent, G. J.: Collagen synthesis and degradation in vivo. Evidence for rapid rates of collagen turnover with extensive degradation of newly-synthesized collagen in tissue of the adult rat. Collagen Rei. Res. 1: 93-104, 1987.

McCulloch, C. A. G. and Melcher, A. H.: Cell density and cell generation in the periodontal ligament of mice. Am. J. Anat. 167: 43-58, 1983.

Molnar,]. A., Alpert, N., Burk,J. F. and Young, V. R.: Synrhesis and degradation rates of collagens in vivo in whole skin of rats studied with 180 1 labelling. Biochem. J. 240: 431-435 , 1986.

Neuberger, A., Perrone, J. C. and Slack, H . G. B.: The relative metabolic inertia of tendon collagen in the rat. Biochem.}. 49: 199-204, 1951.

Neuberger, A. and Slack, H. G. B. : The metabolism of collagen from liver, bone, skin and tendon in the normal rat. Biochem. }. 53: 47-52, 1953.

Nimni, M. E., de Guia, E. and Bavetta, L. A.: Synthesis and turnover of collagen precursors in the rabbit skin. Biochem. J. 102: 143-149, 1967.

Nissen, R., Cardinale, G. J. and Udenfriend, S.: Increased turnover of arterial collagen in hyperten­sive rats. Proc. Natl. Acad. Sci. USA 75: 451-453,1978.

Orlowski, W. A. : Biochemical studies of collagen turnover in rat incisor periodontal ligament. Arch. Oral Bioi. 23: 1163-11 68, 1978.

Palmer, R. M., Robins, S. P. and Lobley, G. R.: Measurement of the synthesis rates of collagens and total protein in rat muscle. Biochem. J. 192: 631-636, 1980.

Pearson, C. H., Wohllebe, M., Carmichael, D. J. and Chovelon, A.: An investigation of the collagen, glycosaminoglycan and insoluble glycoprotein components at different stages of tissue develop­ment. Connect. Tissue Res. 3: 195-206,1975.

Plecash, J. M. and Bentley, ]. P.: Crosslink analysis as an indicator of collagen turnover in periodontal ligament from functioning and non-functioning teeth in the dog. Arch. Oral Bioi. 27: 463-468,1982.

Poole, B.: The kinetics of the disappearance of labeled leucine from the free leucine pool of rat liver and its effects on the apparent turnover of catalase and other hepatic proteins. J. BioI. Chem. 246: 6587-6591,1971.

Reddi, A. S.: Metabolism of glomerular basement membrane in short- and long-term streptozoto­cin diabetic rats. Arch. Int. Phys. et Biochemie 94: 205-218, 1986.

Robins, S. P.: Metabolism of rabbit skin collagen. Differences in the apparent turnover rates of type I and type III collagen precursors determined by constant intravenous infusion of labelled amino acids. Biochem.J.181: 75-82, 1979.

Shuttleworth, C. A. and Smalley, J. W.: Periodontal ligament. Int. Rev. Connect. Tissue Res. 10: 211-247,1983.

21

Sodek, J.: A new approach to assessing collagen turnover by using a micro assay . A highly efficient and rapid turnover of collagen in rat periodontal tissues. Biochem. J. 160: 243-246, 1976.

Sodek, J.: A comparison of the rates of synthesis and turnover of collagen and non-collagen proteins in adult rat periodontal tissues and skin using a microassay. Arch. Oral Bioi. 22: 655-665, 1977.

Sodek, J.: A comparison of collagen and non-collagenous protein metabolism in rat molar and incisor periodontal ligaments. Arch. Oral Bioi. 23: 977-982,1978.

Sodek, J. and Limeback, H. F.: Comparison of the rates of synthesis, conversion, and maturation of type I and type III collagens in rat periodontal tissues. J. Bioi. Chem. 254: 10496-10502, 1979.

Sodek, J.: Periodontal ligament: metabolism. In: Handbook of Experimental Aspects of Oral Biochemistry, ed. by Lazzari, E. P., CRC Press Inc. , Boca Raton, FL, 1983, pp. 183-193.

Sodek, J. and Mandell, S. M.: Collagen metabolism in rat incisor predentine in vivo: synthesis and maturation of type I, 0'1 (I) trimer, and type V collagens Biochemistry 21: 2011-2015, 1982.

Sodek, J.: Collagen turnover in periodontal ligament. In: Biology of Tooth Movement, ed. by Norron, A. L. and Burstone, C. J. CRC Press Inc., Boca Raton, FL, In Press.

Svoboda, E. L. A., Shiga, A. and Deporter, D. A.: A stereological analysis of collagen phagocytosis by fibroblasts in three soft connective tissues with differing rates of collagen turnover. Anat. Rec. 199:473-480,1981.

Taverne, A. A. R., Lemmens, I. G. and Tonio, G. J. M.: Lathyrogens and the role of collagen in the eruption of rat incisors. Arch. Oral Bioi. 31: 127-133, 1986.

Ten Cate, A. R. and Freeman, E.: Collagen remodelling by fibroblasts in wound repair. Preliminary observations. Anat. Rec. 183: 543-546,1974.

Wasi, S., Otsuka, K., Yao, K.-L. , Sodek, J. and Termine, J. D.: An osteonectin-like protein in porcine periodontal ligament and its synthesis by periodontal ligament fibroblasts. Can. ] . Biochem. Cell Bioi. 62: 470-478, 1984.

Waterlow, J. c., Garlick, P.] . and Millward, D. J.: Protein turnover in mammalian tissues and in the whole body. North-Holland, Amsterdam, 1978.

Woessner, J. F. In: Treatise on Collagen, ed. by Gould, R. S., Academic Press, London, 1968, pp. 253-329.

Zak, R., Martin, A. F. and Blough, R.: Assessment of protein turnover by use of radiolabelled tracers. Physiol. Rev. 59: 407-447, 1979.

Dr. Jaro Sodek, MRC Group in Periodontal Physiology, Faculty of Dentistry, Room 4384 Medical Sciences Building, University of Toronto, Toronto, Ontario M5S 1A8, Canada.