Cell Tissue Kinet. (1 976) 9,293-302.
P E R I O D O N T A L LIGAMENT CELL K I N E T I C S FOLLOWING O R T H O D O N T I C TOOTH MOVEMENT J. A. YEE,*D. €3. K I M M E ALN D W . S . S . J E E Department of Anatomy, College of Medicine, University of Utah, Salt Lake City, Utah (Received 13 August 1975; revision received 17 November 1975)
ABSTRACT
The population of periodontal ligament (PDL) fibroblasts examined in this study may include osteogenic progenitor cells. PDL fibroblast and osteoblast kinetics in the periodontal ligament of the rat were measured following orthodontic stimulation of bone formation. Both single and multiple injections of tritiated thymidine (3H-TdR) were used. In single injection experiments, the peak percentage of PDL fibroblasts labeled. with 3H-TdR is 15 % at 22 hr post-stimulation. In multiple injection experiments, the total percentage of fibroblasts in the PDL which respond by synthesizing DNA is 50%. 'H-TdR-labeled osteoblasts appear at the same rate as, but with a time delay after, the labeled fibroblasts. Following stimulation, the most likely source of osteoblasts at the bone-forming site is not only fibroblasts which make DNA, divide, then differentiate, but also fibroblasts which either are differentiated to osteoblasts without DNA synthesis and cell division, or are released from G2 block by the orthodontic stimulation. INTRODUCTION The study of bone cell kinetics has been limited by the small number of populations of bone cells which show significant cellular proliferation. Some studies show that the labeling index of osteogenic cells in the periosteum of the mouse tibia declines to 0.7 % by 8 weeks of age (Tonna, 1961). Pre-osteoblasts in the tibia1 periosteum of the young, rapidly growing rabbit (Owen, 1963), or mesenchymal cells in the primary spongiosa of the proximal tibia of growing rats (Kember, 1960; Young, 1962) show a significant labeling index, but are inalterably linked to the complex process of longitudinal bone growth. As the rate of bone elongation slows, the labeling index falls, leaving yet another cell population of low proliferative activity. In other tissues which exhibit low cell proliferation, persons have stimulated the tissue in question prior to studying its cell kinetics. A few of such systems are the partially-hepatectomized liver (Grisham, 1962), the testosterone propionate stimulated seminal vesicle (Morley,
* Present address: Department of Anatomy, School of Medicine, Texas Tech University, Health Sciences Centers, P.O. Box 4569, Lubbock, Texas 79409, U.S.A. Correspondence: Dr Donald B. Kimmel, 2C 110 Medical Center, Department of Anatomy, College of Medicine, University of Utah, Salt Lake City, Utah 84132, U.S.A. 293
294
J. A . Yee, D . B. Kimmel and W.S. S. Jee
Wright & Appleton, 1973), and plucked mouse skin (Hamilton & Potten, 1972). In each of these systems, the result of the stimulation is enhancement of the tissue. In the liver, stimulated cell proliferation results in the regeneration of a great deal of liver tissue. The stimulated cell proliferation in the castrate mouse seminal vesicle results in an increase in the mass of the seminal vesicle. In the plucked skin, cell proliferation in the hair follicles and basal layer of the epidermis results in more rapid hair growth and more rapid turnover of the layers of the skin. The only stimulated system in bone whose cell kinetics had been studied until recently was the fractured long bone in mice (Tonna & Cronkite, 1961). While this is an excellent model for the study of cell kinetics of bone repair following traumatic fracture, it is possible that the cell proliferation response in areas of stimulated bone modeling, not accompanied by a great amount of inflammation and intervening cartilage formation, is different (Frost, 1973). The fibroblasts of the periodontal ligament surrounding teeth are a population of cells with low proliferative activity (Tonna & Weiss, 1972; Roberts & Jee, 1974). Recent work (Roberts & Jee, 1974)has shown that the periodontal ligament (PDL) fibroblasts can be stimulated to proliferate. An elastic is inserted between two molars in the rat maxilla to cause the first molar to move, stretching the fibers of the PDL anterior to the first molar. One result of cell proliferation in this system is cell differentiation to form osteoblasts with consistent bone formation on the alveolar bone surface mesial to the tooth, without a period of cartilage formation. The purposes of this report are to (1) characterize in detail the cell kinetic response of a PDL of arat to orthodontic tooth movement, (2) show that the cells which would seem to have evolved from PDL fibroblasts appear later as differentiated osteoblasts on the alveolar bone surface, and (3) compare this area of stimulated bone modeling to areas in other stimulated tissues in which induced cell proliferation has led to enhancement of that particular tissue. MATERIALS A N D METHODS Experimental animals Male Sprague-Dawley rats weighing 100 g were allowed to acclimate to their cages for 7 days. Water and food (Wayne Lab Blox; Allied Mills Inc., Chicago, Illinois 60606) were available ad libitum. The laboratory room was maintained on a strict 12 hr lightaark cycle. On the eighth day the rats were placed under light ether anesthesia. A piece of elastic with dimensions 2 nun length by 1 mm height by 0-2 mm thickness (trimmed from No. 5-104, Rocky Mountain Co., Denver, Colorado) was secured in the contact area between the first and the second left maxillary molars, using two mosquito hemostats; all of this work was completed between 8 and 10 a.m. The 1 mm height was sufficient to have the elastic in the contact area and out of interference with the rats’ chewing motions. Only rats whose elastics were in place at their sacrifice time were used for data analysis in this experiment. Two experiments were conducted, the first, a pulse label with 3H-thymidine (IH-TdR) at various times post-stimulation, and the second, a multiple label with 3H-TdR study. Experiment 1. In a series of experiments, a number of rats, given in parentheses, were killed at the following intervals post-stimulation of the PDL: 5 (9,9 (9, 11 (9, 17 (7),23 (lo), 28 (S), 29 (5), 35 (4), 37 (4), 41 (4), 47 (4), 49 (4), 53 (4), 59 (4), 65 (4) and 71 (4) hr. Each animal received 1.0 pCi 3H-TdR (New England Nuclear Co., spec-act. = 28.8 Ci/mM) per gram body weight, 1 hr before being killed.
PDL cell kinetics
295
Experiment 2. Multiple injections of 'H-TdR were used. 3H-TdR was injected at 6 hr intervals into every animal. Injections were carried out at 4, 10, 16,22,28,34,40,46,52, 58, 64 and 70 hr post-stimulation with 0.5 pCi 'H-TdR/g body weight. The rats were killed at 5, 11, 17, 23, 29, 35, 41, 47, 53, 59, 65 and 71 hr post-stimulation of the PDL. This meant that some animals received one injection of 'H-TdR ( 5 hr animals), some animals received two injections of 'H-TdR (1 1 hr animals), some received three injections of 'H-TdR (17 hr animals), and so forth, until some animals had received a total of twelve injections of 'HTdR (71 hr animals). Histologic methods and criteria
At autopsy, the rats were killed by decapitation and the maxillas were removed whole and placed in Millonig's phosphate-buffered formalin, pH 7.4. After 2 days they were passed through a series of seven to ten changes of 10 % neutral EDTA (Versene Flakes, Na, EDTA, Technical Grade, Dow Chemical CO.,Midland, Michigan). At various times during this process they were trimmed parasagitally to expose the mesial root of the maxillary first molar from the buccal aspect. When decalcification was completed, after about 10 days, the specimens were embedded in a modified methyl methacrylate embedding medium (Kimmel & Jee, 1975). Three micron thick sections were prepared on the Jung Model-K microtome and affixed to slides coated with 0.5 % gelatin. The plastic was removed from the sections in three serial changes of acetone and the slides were dipped in Kodak's NTB liquid emulsion diluted 1 : 1 with distilled water. The slides were stored in light-tight cans at -18°C for 21 days, then developed in Kodak's D-19, fixed in Kodak's Acid-Fixer (Eastman Kodak Co., Rochester, New York), and stained with Mayer's hematoxylin, then counterstained with eosin. Coverslips were affixed with Permount (Fisher Scientific Co., Chicago, Illinois). The area under examination was the periodontal ligament mesial (anterior) to the mesial root of the maxillary first molar (Fig. 3). The alveolar bone surface mesial to the first molar is normally asite of bone resorption, due to (1) distal tipping of the tooth in response to abrasion in the contact area between the first and second molars and (2) active eruption in response to occlusal abrasion (Roberts, 1975). Beginning at the alveolar crest of bone and continuing apically toward the root of the tooth, five areas, each 220 pm long, were identified (Fig. 1) and counted for the following features: 3H-TdR-labeled fibroblasts, unlabeled fibroblasts, labeled osteoblasts, and unlabeled osteoblasts. The term PDL fibroblast, in this study, designates cells in the connective tissue of the PDL which, at the light microscope level, have round to fusiform nuclei and little visible cytoplasm. It is possible that this grouping of cells includes osteoprogenitor cells (Young, 1962) or preosteoblasts (Owen, 1963), whose morphologic characteristics are not especially different from undifferentiated mesenchymal cells or synthetic fibroblasts (Melcher & Eastoe, 1969). Osteoblasts in this study are cells located next to the alveolar bone surface mesial to the tooth, which have a n eccentric nucleus, basophiliccytoplasm, and prominent Golgi apparatus. Thirty serial sections from each rat maxilla were prepared and placed three per slide. Three sections, separated by at least 30 pm, were examined from each animal; this meant that about 2000 cells from each bone were counted. The numbers were averaged on a per section basis. There was 1100 pm of bone surface counted in each section. All slides were coded with a random number prior to counting procedures in an attempt to
296
J. A. Yee, D.B. Kimmel and W. S. S.Jee
FIG,1. Left maxillag first molar and portion of,the second molar of a Sprague-Dawley rat. The five regions undkr study, each 220 gm square,extend from the crest of the alveolar bone 1100 pm apically. All counts performed in this study were on fibroblasts or osteoblasts lying wholly or partly within the five squares. Bone (B); cementum (C); dentin (D); enamel (E); orthodontic elastic in place between the two teeth (EL); gingiva (G); new bone formed following orthodontic stimulation (N); osteoblasts (0);pulp cavity (P). avoid observer bias. Data collected are expressed as the labeling index (per cent labeled cells 1 hr following a single injection of 'H-TdR) where appropriate, and as the multiple labeling index (per cent labeled cells after more than one injection of 'H-TdR) where appropriate. Statisticalanalysis was carried out by the ANOVA technique (Wolff, 1968). Where necessary, lines were fit to the data using the least squares method. RESULTS Experiment 1(pulse-labelstudy) Only fibroblasts were ever labeled in this study. At 5 hr post-stimulation, the PDL fibroblast labeling index was about 2-5%. After 11 hr it was only about 3.5 %, but by 17 hr it had risen to 7 %. At 23 hr post-stimulation the peak labeling index of 15-5% was observed. The labelling index fell to 7 % by 29 hr, and achieved a plateau level of 2-3 % by 35 hr (Fig. 2). The number of osteoblasts increased linearly with time following stimulation. No 3H-TdRlabeled osteoblasts were ever observed. Experiment 2 (multiple-label study) At 11 hr post-stimulation (two injections of 'H-TdR), the multiple labeling index of fibroblasts was about 4 %. At 17 hr (three injections of "-TdR), the multiple labeling index of fibroblastshad risen to 21 %. It continued to rise until 35 hr post-stimulation(six injections of jH-TdR), when it reached its peak level of about 50 %. It varied only slightly from that level throughout the duration of the experiment, which was a total of 71 hr post-stimulation (twelve injections of 3H-TdR). The slope of the line fit to the rising phase of the fibroblast
PDL cell kinetics
297
15 -
*
0
10
20
30
40
50
60
70
Hwn post-stirnulotion
FIG.2. Labeting index of PDL fibroblasts following orthodontic stimulation. The W-TdR labeling index began about 2 %, rose to a maximum of 15 % at 22 hr, then declined to about 3 % by 65-70 hr post-stimulation.
FIG.3
FIG.4
FIG.3. Periodontal ligament prior to stimulation. Usually, there are a few osteoclasts (arrows) on the alveolar bone surface mesial to the tooth. Alveolar bone (b); cementum (c); dentin (d); fibroblasts (f). Hematoxylin and eosin. x330. FIG.4.Autoradiograph of PDL 35 hr post-stimulation. Many fibroblasts (f) have incorporated 3H-TdR by this time. Hematoxylin and eosin, ~660.
298
J. A . Yee, D. B. Kimmel and W.S.S.Jee
0 . t . e .
50-
I
0
10
20
I
30 40 50 noun post-stimulation
I
I
60
70
FIG.5. Multiple labeling index of PDL fibroblasts following orthodontic stimulation. The multiple labeling index rose from 4% to 50% by 35 hr (six injections of 'H-TdR) post-stimulation, then plateaued a't that level.
FIG.6 FIG.7 FIG.6. Periodontal ligament 71 hr post-stimulation. New bone (nb) formation has taken place beginning at a growth arrest line (9). Many osteoblastsare present (arrows). Cementum (c). Hematoxylin and eosin, x330. FIG.7. Autoradiograph of periodontal ligament 71 hr post-stimulation. Note 3H-TdR-labeled osteoblasts (ob) and fibroblasts (f) near area of new bone (nb) formation. Hematoxylin and eosin, x660.
PDL cell kinetics
299
multiple labeling curve showed that the rate of increase in the multiple labeling index was 1-7%/hr between 12 and 32 hr post-stimulation (Figs. 4 and 5). The total number of osteoblasts, both 3H-TdR-labeledand unlabeled, adjacent to the alveolar bone was 2 at 5 hr post-stimulation, 9 at 11 hr post-stimulation, 16 at 17 hr post-stimulation, and 22 at 23 hr post-stimulation. This number continued to rise to 42 at 35 hr poststimulation, 57 after 47 hr of stimulation, and 70 after 71 hr of stimulation. The slope of the line in Fig. 8 showed that the rate of increase of osteoblasts was a steady 1.1 osteoblasts/hr (Figs. 6 and 8). The number of labeled osteoblasts was 0 after 5 hr of stimulation, 1 after 11 hr of stimulation, and 9 by 23 hr post-stimulation. It continued to rise linearly to a maximum of 59 at 71 hr post-stimulation. The rate of increase of labeled osteoblasts after 23 hr was 1.0 osteoblasts/hr. The number of unIabeled osteoblasts was 2 after 5 hr, 8 after 11 hr, and 12 after 17 hr of stimulation. It reached a plateau level of 20 by 41 hr post-stimulation (Fig.9).
'
"t
60
:
Hours post-stimulation
FIG. 8. Average number of osteoblasls per section following orthodontic stimulation. The number of osteoblasts per section rose from 0 to 79 by 71 hr post-stimulation.
6o
t Hours post-stimulation
FIG.9. Average number of labeled and unlabeled osteoblasts per section. By 71 hr post-stimulation, labeled osteoblast numbers (---) were increasing rapidly, while unlabeled osteoblast were constant. numbers (-)
300
J. A . Yee, D . B. Kimmel and W . S. S. Jee 0
0
‘0 .
x)
30
40
50
60
70
Hours post-stimulation
FIG.10. Multiple labeling index of osteoblasts following orthodontic stimulation. The multiple labeling index rose from 0 to 60% by 35 hr (six injections of 3H-TdR), then continued to rise more slowly toward 75% by 71 hr (twelve injections of ’H-TdR) post-stimulation.
The multiple labeling index of osteoblasts was 11 % at 11 hr post-stimulation, 31 % by 17 hr post-stimulation, and 43 % after 23 hr of stimulation. After reaching 56 % by 35 hr post-stimulation, the rise in the multiple labeling index became less marked, reaching its maximum of 75% at 71 hr post-stimulation. From 5 to 35 hr the slope of the best-fit line shows that the increase in the multiple labeling index of osteoblasts was 1.9 %/hr. From 35 hr on, the rate of increase was 0-4%/hr (Figs. 7 and 10). DISCUSSION The force applied to the tooth in this study is a tipping force, which results in complex changes in the PDL mesial to the tooth. A more uniform bone formation response could have been provided by a ‘bodily movement’ of the tooth (Grabet, 1972), but such orthodontics is too difficult to accomplish in rats’ mouths. A very exacting way to define the response to this tipping is by histologic observation at serial times following stimulation, as in this study. Despite the theory which would say that tipping results in both formation and resorption on both sides of the tooth, resorption of bone was not observed in any of the five areas examined after stimulation of up to 71 hr duration. Bone formation, however, occurred consistently in the two apical-most areas examined, and occasionally in the third area (Fig. 1). The labeling index study was performed to establish the time sequence of PDL fibroblast proliferative response to this stimulus to the periodontal ligament. In timing, the response of the PDL fibroblasts is quite similar to that of the liver parenchymal cells following partial hepatectomy; however, the magnitude of response is somewhat less than that seen in these same cells or in hotmonally-stimulated cells in the seminal vesicle (Grisham, 1962; Morley et af., 1973). However, it is a much quicker response to an initial insult than the plucking of the mouse skin, where the labeling index is about five times normal for 7 days post-pluck (Hegazy & Fowler, 1973). This stimulus to the PDL causes a complex change probably requiring two different lines of cellular differentiation for its return to the steady state. One
PDL cell kinetics
301
may provide new PDL fibroblasts which will repair stretched and tom fibers of the periodontal ligament, while the other may provide osteoblasts to the alveolar bone surface mesial to the tooth, which will form immature, woven bone. This particular stimulated system seems to be more like the partial hepatectomy and plucking models than the hormone-target tissues systems because it represents a lasting deformation of the tissue to which it must respond with some limited type of repair response. The multiple labeling index study of PDL fibroblasts establishes that the percentage of fibroblasts responding to stimulation of the PDL by synthesizing DNA in this model is about 50 %. It also confirms that the cell cycle time of these cells is at least 34 hr in duration (Roberts, Chase & Jee, 1974). Once again, the stimulated PDL has its own characteristics. In the plucked skin, the percentage of cells responding is 100 % and the cell cycle time, when measured by a multiple-labeling study, is about 37 hr (Hegazy & Fowler, 1973). The number of osteoblasts in the PDL is a measure of the enhancement of the bone formation aspect of this model. The identification of osteoblasts is a particularly good end point, as osteoblasts are morphologically and geographically identifiable. The rate of increase of the multiple labeling index of the PDL fibroblasts was 1.7 %/hr between 12 and 32 hr post-stimulation. In this interval, the rate of increase of the multiple labeling index of osteoblasts was 1-9%/hr. The similarities in these rates of labeling increase are indirect evidence that multipotent PDL fibroblasts are the source of osteoblasts forming bone in this model. It does not seem as likely that the stimulation could have evoked such a response in another population of cells not as closely allied with the PDL as the fibroblasts which reside there nativeIy. It is possible that if PDL fibroblasts which were osteogenic in nature could be identified, an even closer correlation between the labeling of PDL osteoblasts and a portion of the PDL fibroblasts could be established. Unfortunately, this distinction could not be made at the light microscope level. It is significant to note that few if any unlabeled osteoblasts evolve after 34 hr of stimulation. There may be two pathways for osteoblast differentiation following stimulation of the PDL, one direct, and one indirect. It is possible that the unlabeled osteoblasts came from cells which went through an S phase which was so short in duration that no 3H-TdR from the serial injections was available, or that they came from cells which were proliferating before the serial injections of 3H-TdR began. However, in choosing the spacing for the serial injections, the work of Roberts et al. (1974) and Kimmel (unpublished), which found that S phase for the PDL fibroblasts after stimulation was about 8 hr, was considered. With injections every 6 hr, it is considered unlikely that a significant number of cells were 'missed' during their S phases. While it remains possible that some cells which had proliferated shortly prior to stimulation were the source of the new osteoblasts, it is as likely that a significant portion of the osteoblasts which initially evolved in this stimulation (through the first 28 hr) differentiate from cells which do not go through an S phase immediately (less than 12 hr) prior to differentiation. Considering the similarities in function and ultrastructure of fibroblasts and osteoblasts, it seems reasonable to speculate that a PDL fibroblast could differentiate directly to an osteoblast. There is also some possibility that the unlabeled osteoblasts arise from fibroblasts which are in G,block (Gelfant & Smith, 1972), then are released by the stimulation to divide, then differentiate. The methods used in this study do not permit a distinction to be made between these two possibilities. The PDL fibroblasts of this study, which may include cells of both osteogenic and fibroblastic potential, are multipotent, reflecting the varied function of the PDL. For example, in
J . A. Yee, D. B. Kimmel and W.S. S. Jee
302
the PDL surrounding the continuously-erupting mouse incisor, the end result of PDL fibroblast proliferative and differentiation activity is the eruption of the tooth (Zajicek, 1974). In the current paper the ability of the PDL fibroblast to proliferate and differentiatein response to a tipping of the tooth is described. The PDL fibroblast populationis apparentlythe primary source of osteoblasts which form bone which stabilizes the tooth in its new position. It (the PDL fibroblast) is probably the main cell concerned with all movements and maintenance of the teeth. ACKNOWLEDGMENT
This work was supported by U.S.E.R.D.A. E( 11-1)-119 and NIH Grants DE-151 and GM958. REFERENCES
FROST, H.M. (1973) Bone Modeling und Skeletal Modeling Errors, p. 73. C. C. Thomas, Springfield, Illinois. GELFANT,S. & SMITH,J.G. (1972) Aging: noncycling cells an explanation. Science, 178,357. GRABER, T.M. (1972) Orthodontics:Principles and Prucrice, 3rd edn, p. 502. Saunders, Philadelphia. GRISHAM,J.W. (1962) A morphologic study of deoxyribonucleic acid synthesis and cell proliferation in regenerating rat liver; autoradiography with thymidine-H3. Cancer Res. 22,842. HAMILTON, E. & EN. C.W. (1972) Influence of hair plucking on the turnover time of the epidermal basal layer. Cell Tissue Kiner. 5, 505. HEGAZY, M.A.H. & FOWLER, J.F. (1973) Cell population kinetics of plucked and unplucked mouse skin. I. Unirradiatedskin.. Cell Tissue Kiner. 6,17. KEMBER, N.F.(1960) Cell division in endochondral ossification. A study of cell proliferation in rat bones by the method of tritiated thymidine autoradiography. J. Bone Jr Surg. 42-B.824. KIMMEL,D.B. & JEE.W.S.S. (1975) A rapid plastic embedding technique for preparation of three-micron thick sections of decalcified hard tissue. Stain Technol.50.83. MELCHER, A.H. & EASTOE,J.E. (1969) The connective tissue of the periodontium. Biology ofrhe Periodontiurn fed. by A. H. Melcher and W. H. Bowen), p. 167. Academic Press,New York. MORLEY, A.R., WRIGHT,N.A. & APPLETON.D. (1973) Cell proliferation in the castrate mouse seminal vesicle in response to testosterone propionate. I. Experimental observations. Cell Tissue Kinet. 6, 239. OWEN. M. (1963) Cell population kinetics of an osteogenic tissue. J. Cell Biol.19.19. ROBERTS. W.E.(1975) Cell kinetic nature and diurnal periodicity of the rat periodontal ligament. Arch. Oru Biol. 20,465. ROBERTS, W.E., CHASE. D.C.& JEE, W.S.S. (1974) Counts of labeled mitoses in the orthodontically-stimulated periodontal ligament in the rat. Arch. Oral Biol. 19,665. ROBERTS, W.E. &JEE.W.S.S. (1974) Cell kinetics oforthodontically-stimulated and non-stimulated periodontal ligament in the rat. Arch. Oral Biol. 19,17. TONNA, E.A. (1961) The cellular component of the skeletal system studied autoradiographically with tritiated thymidine (HTdR) during growth and aging. J. biophys. biochem. Cyrol. 9,813. TONNA, E.A. & CRONKITE, E.P. (1961) Cellular response to fracture studied with tritiated thymidine. J. Bone Jr Surg. &A, 352. TONNA, E.A. & WEIS. R. (1972) The cell proliferative activity of paradontal tissues in aging mice. Arch. Oral Biol. 17,969. WOLFF,C.M. (1968) Principles of Biometry. Van Nostrand, Princeton, New Jersey. YOUNG,R.W. (1962) Cell proliferation and specialization during endochondral osteogenesis. J. Cell Biol. 14, 357.
ZAJICEK,G . (1974) Fibroblast cell kinetics in the periodontal ligament of the mouse. Cell Tissue Kiner. 7,479.
P E R I O D O N T A L LIGAMENT CELL K I N E T I C S FOLLOWING O R T H O D O N T I C TOOTH MOVEMENT J. A. YEE,*D. €3. K I M M E ALN D W . S . S . J E E Department of Anatomy, College of Medicine, University of Utah, Salt Lake City, Utah (Received 13 August 1975; revision received 17 November 1975)
ABSTRACT
The population of periodontal ligament (PDL) fibroblasts examined in this study may include osteogenic progenitor cells. PDL fibroblast and osteoblast kinetics in the periodontal ligament of the rat were measured following orthodontic stimulation of bone formation. Both single and multiple injections of tritiated thymidine (3H-TdR) were used. In single injection experiments, the peak percentage of PDL fibroblasts labeled. with 3H-TdR is 15 % at 22 hr post-stimulation. In multiple injection experiments, the total percentage of fibroblasts in the PDL which respond by synthesizing DNA is 50%. 'H-TdR-labeled osteoblasts appear at the same rate as, but with a time delay after, the labeled fibroblasts. Following stimulation, the most likely source of osteoblasts at the bone-forming site is not only fibroblasts which make DNA, divide, then differentiate, but also fibroblasts which either are differentiated to osteoblasts without DNA synthesis and cell division, or are released from G2 block by the orthodontic stimulation. INTRODUCTION The study of bone cell kinetics has been limited by the small number of populations of bone cells which show significant cellular proliferation. Some studies show that the labeling index of osteogenic cells in the periosteum of the mouse tibia declines to 0.7 % by 8 weeks of age (Tonna, 1961). Pre-osteoblasts in the tibia1 periosteum of the young, rapidly growing rabbit (Owen, 1963), or mesenchymal cells in the primary spongiosa of the proximal tibia of growing rats (Kember, 1960; Young, 1962) show a significant labeling index, but are inalterably linked to the complex process of longitudinal bone growth. As the rate of bone elongation slows, the labeling index falls, leaving yet another cell population of low proliferative activity. In other tissues which exhibit low cell proliferation, persons have stimulated the tissue in question prior to studying its cell kinetics. A few of such systems are the partially-hepatectomized liver (Grisham, 1962), the testosterone propionate stimulated seminal vesicle (Morley,
* Present address: Department of Anatomy, School of Medicine, Texas Tech University, Health Sciences Centers, P.O. Box 4569, Lubbock, Texas 79409, U.S.A. Correspondence: Dr Donald B. Kimmel, 2C 110 Medical Center, Department of Anatomy, College of Medicine, University of Utah, Salt Lake City, Utah 84132, U.S.A. 293
294
J. A . Yee, D . B. Kimmel and W.S. S. Jee
Wright & Appleton, 1973), and plucked mouse skin (Hamilton & Potten, 1972). In each of these systems, the result of the stimulation is enhancement of the tissue. In the liver, stimulated cell proliferation results in the regeneration of a great deal of liver tissue. The stimulated cell proliferation in the castrate mouse seminal vesicle results in an increase in the mass of the seminal vesicle. In the plucked skin, cell proliferation in the hair follicles and basal layer of the epidermis results in more rapid hair growth and more rapid turnover of the layers of the skin. The only stimulated system in bone whose cell kinetics had been studied until recently was the fractured long bone in mice (Tonna & Cronkite, 1961). While this is an excellent model for the study of cell kinetics of bone repair following traumatic fracture, it is possible that the cell proliferation response in areas of stimulated bone modeling, not accompanied by a great amount of inflammation and intervening cartilage formation, is different (Frost, 1973). The fibroblasts of the periodontal ligament surrounding teeth are a population of cells with low proliferative activity (Tonna & Weiss, 1972; Roberts & Jee, 1974). Recent work (Roberts & Jee, 1974)has shown that the periodontal ligament (PDL) fibroblasts can be stimulated to proliferate. An elastic is inserted between two molars in the rat maxilla to cause the first molar to move, stretching the fibers of the PDL anterior to the first molar. One result of cell proliferation in this system is cell differentiation to form osteoblasts with consistent bone formation on the alveolar bone surface mesial to the tooth, without a period of cartilage formation. The purposes of this report are to (1) characterize in detail the cell kinetic response of a PDL of arat to orthodontic tooth movement, (2) show that the cells which would seem to have evolved from PDL fibroblasts appear later as differentiated osteoblasts on the alveolar bone surface, and (3) compare this area of stimulated bone modeling to areas in other stimulated tissues in which induced cell proliferation has led to enhancement of that particular tissue. MATERIALS A N D METHODS Experimental animals Male Sprague-Dawley rats weighing 100 g were allowed to acclimate to their cages for 7 days. Water and food (Wayne Lab Blox; Allied Mills Inc., Chicago, Illinois 60606) were available ad libitum. The laboratory room was maintained on a strict 12 hr lightaark cycle. On the eighth day the rats were placed under light ether anesthesia. A piece of elastic with dimensions 2 nun length by 1 mm height by 0-2 mm thickness (trimmed from No. 5-104, Rocky Mountain Co., Denver, Colorado) was secured in the contact area between the first and the second left maxillary molars, using two mosquito hemostats; all of this work was completed between 8 and 10 a.m. The 1 mm height was sufficient to have the elastic in the contact area and out of interference with the rats’ chewing motions. Only rats whose elastics were in place at their sacrifice time were used for data analysis in this experiment. Two experiments were conducted, the first, a pulse label with 3H-thymidine (IH-TdR) at various times post-stimulation, and the second, a multiple label with 3H-TdR study. Experiment 1. In a series of experiments, a number of rats, given in parentheses, were killed at the following intervals post-stimulation of the PDL: 5 (9,9 (9, 11 (9, 17 (7),23 (lo), 28 (S), 29 (5), 35 (4), 37 (4), 41 (4), 47 (4), 49 (4), 53 (4), 59 (4), 65 (4) and 71 (4) hr. Each animal received 1.0 pCi 3H-TdR (New England Nuclear Co., spec-act. = 28.8 Ci/mM) per gram body weight, 1 hr before being killed.
PDL cell kinetics
295
Experiment 2. Multiple injections of 'H-TdR were used. 3H-TdR was injected at 6 hr intervals into every animal. Injections were carried out at 4, 10, 16,22,28,34,40,46,52, 58, 64 and 70 hr post-stimulation with 0.5 pCi 'H-TdR/g body weight. The rats were killed at 5, 11, 17, 23, 29, 35, 41, 47, 53, 59, 65 and 71 hr post-stimulation of the PDL. This meant that some animals received one injection of 'H-TdR ( 5 hr animals), some animals received two injections of 'H-TdR (1 1 hr animals), some received three injections of 'H-TdR (17 hr animals), and so forth, until some animals had received a total of twelve injections of 'HTdR (71 hr animals). Histologic methods and criteria
At autopsy, the rats were killed by decapitation and the maxillas were removed whole and placed in Millonig's phosphate-buffered formalin, pH 7.4. After 2 days they were passed through a series of seven to ten changes of 10 % neutral EDTA (Versene Flakes, Na, EDTA, Technical Grade, Dow Chemical CO.,Midland, Michigan). At various times during this process they were trimmed parasagitally to expose the mesial root of the maxillary first molar from the buccal aspect. When decalcification was completed, after about 10 days, the specimens were embedded in a modified methyl methacrylate embedding medium (Kimmel & Jee, 1975). Three micron thick sections were prepared on the Jung Model-K microtome and affixed to slides coated with 0.5 % gelatin. The plastic was removed from the sections in three serial changes of acetone and the slides were dipped in Kodak's NTB liquid emulsion diluted 1 : 1 with distilled water. The slides were stored in light-tight cans at -18°C for 21 days, then developed in Kodak's D-19, fixed in Kodak's Acid-Fixer (Eastman Kodak Co., Rochester, New York), and stained with Mayer's hematoxylin, then counterstained with eosin. Coverslips were affixed with Permount (Fisher Scientific Co., Chicago, Illinois). The area under examination was the periodontal ligament mesial (anterior) to the mesial root of the maxillary first molar (Fig. 3). The alveolar bone surface mesial to the first molar is normally asite of bone resorption, due to (1) distal tipping of the tooth in response to abrasion in the contact area between the first and second molars and (2) active eruption in response to occlusal abrasion (Roberts, 1975). Beginning at the alveolar crest of bone and continuing apically toward the root of the tooth, five areas, each 220 pm long, were identified (Fig. 1) and counted for the following features: 3H-TdR-labeled fibroblasts, unlabeled fibroblasts, labeled osteoblasts, and unlabeled osteoblasts. The term PDL fibroblast, in this study, designates cells in the connective tissue of the PDL which, at the light microscope level, have round to fusiform nuclei and little visible cytoplasm. It is possible that this grouping of cells includes osteoprogenitor cells (Young, 1962) or preosteoblasts (Owen, 1963), whose morphologic characteristics are not especially different from undifferentiated mesenchymal cells or synthetic fibroblasts (Melcher & Eastoe, 1969). Osteoblasts in this study are cells located next to the alveolar bone surface mesial to the tooth, which have a n eccentric nucleus, basophiliccytoplasm, and prominent Golgi apparatus. Thirty serial sections from each rat maxilla were prepared and placed three per slide. Three sections, separated by at least 30 pm, were examined from each animal; this meant that about 2000 cells from each bone were counted. The numbers were averaged on a per section basis. There was 1100 pm of bone surface counted in each section. All slides were coded with a random number prior to counting procedures in an attempt to
296
J. A. Yee, D.B. Kimmel and W. S. S.Jee
FIG,1. Left maxillag first molar and portion of,the second molar of a Sprague-Dawley rat. The five regions undkr study, each 220 gm square,extend from the crest of the alveolar bone 1100 pm apically. All counts performed in this study were on fibroblasts or osteoblasts lying wholly or partly within the five squares. Bone (B); cementum (C); dentin (D); enamel (E); orthodontic elastic in place between the two teeth (EL); gingiva (G); new bone formed following orthodontic stimulation (N); osteoblasts (0);pulp cavity (P). avoid observer bias. Data collected are expressed as the labeling index (per cent labeled cells 1 hr following a single injection of 'H-TdR) where appropriate, and as the multiple labeling index (per cent labeled cells after more than one injection of 'H-TdR) where appropriate. Statisticalanalysis was carried out by the ANOVA technique (Wolff, 1968). Where necessary, lines were fit to the data using the least squares method. RESULTS Experiment 1(pulse-labelstudy) Only fibroblasts were ever labeled in this study. At 5 hr post-stimulation, the PDL fibroblast labeling index was about 2-5%. After 11 hr it was only about 3.5 %, but by 17 hr it had risen to 7 %. At 23 hr post-stimulation the peak labeling index of 15-5% was observed. The labelling index fell to 7 % by 29 hr, and achieved a plateau level of 2-3 % by 35 hr (Fig. 2). The number of osteoblasts increased linearly with time following stimulation. No 3H-TdRlabeled osteoblasts were ever observed. Experiment 2 (multiple-label study) At 11 hr post-stimulation (two injections of 'H-TdR), the multiple labeling index of fibroblasts was about 4 %. At 17 hr (three injections of "-TdR), the multiple labeling index of fibroblastshad risen to 21 %. It continued to rise until 35 hr post-stimulation(six injections of jH-TdR), when it reached its peak level of about 50 %. It varied only slightly from that level throughout the duration of the experiment, which was a total of 71 hr post-stimulation (twelve injections of 3H-TdR). The slope of the line fit to the rising phase of the fibroblast
PDL cell kinetics
297
15 -
*
0
10
20
30
40
50
60
70
Hwn post-stirnulotion
FIG.2. Labeting index of PDL fibroblasts following orthodontic stimulation. The W-TdR labeling index began about 2 %, rose to a maximum of 15 % at 22 hr, then declined to about 3 % by 65-70 hr post-stimulation.
FIG.3
FIG.4
FIG.3. Periodontal ligament prior to stimulation. Usually, there are a few osteoclasts (arrows) on the alveolar bone surface mesial to the tooth. Alveolar bone (b); cementum (c); dentin (d); fibroblasts (f). Hematoxylin and eosin. x330. FIG.4.Autoradiograph of PDL 35 hr post-stimulation. Many fibroblasts (f) have incorporated 3H-TdR by this time. Hematoxylin and eosin, ~660.
298
J. A . Yee, D. B. Kimmel and W.S.S.Jee
0 . t . e .
50-
I
0
10
20
I
30 40 50 noun post-stimulation
I
I
60
70
FIG.5. Multiple labeling index of PDL fibroblasts following orthodontic stimulation. The multiple labeling index rose from 4% to 50% by 35 hr (six injections of 'H-TdR) post-stimulation, then plateaued a't that level.
FIG.6 FIG.7 FIG.6. Periodontal ligament 71 hr post-stimulation. New bone (nb) formation has taken place beginning at a growth arrest line (9). Many osteoblastsare present (arrows). Cementum (c). Hematoxylin and eosin, x330. FIG.7. Autoradiograph of periodontal ligament 71 hr post-stimulation. Note 3H-TdR-labeled osteoblasts (ob) and fibroblasts (f) near area of new bone (nb) formation. Hematoxylin and eosin, x660.
PDL cell kinetics
299
multiple labeling curve showed that the rate of increase in the multiple labeling index was 1-7%/hr between 12 and 32 hr post-stimulation (Figs. 4 and 5). The total number of osteoblasts, both 3H-TdR-labeledand unlabeled, adjacent to the alveolar bone was 2 at 5 hr post-stimulation, 9 at 11 hr post-stimulation, 16 at 17 hr post-stimulation, and 22 at 23 hr post-stimulation. This number continued to rise to 42 at 35 hr poststimulation, 57 after 47 hr of stimulation, and 70 after 71 hr of stimulation. The slope of the line in Fig. 8 showed that the rate of increase of osteoblasts was a steady 1.1 osteoblasts/hr (Figs. 6 and 8). The number of labeled osteoblasts was 0 after 5 hr of stimulation, 1 after 11 hr of stimulation, and 9 by 23 hr post-stimulation. It continued to rise linearly to a maximum of 59 at 71 hr post-stimulation. The rate of increase of labeled osteoblasts after 23 hr was 1.0 osteoblasts/hr. The number of unIabeled osteoblasts was 2 after 5 hr, 8 after 11 hr, and 12 after 17 hr of stimulation. It reached a plateau level of 20 by 41 hr post-stimulation (Fig.9).
'
"t
60
:
Hours post-stimulation
FIG. 8. Average number of osteoblasls per section following orthodontic stimulation. The number of osteoblasts per section rose from 0 to 79 by 71 hr post-stimulation.
6o
t Hours post-stimulation
FIG.9. Average number of labeled and unlabeled osteoblasts per section. By 71 hr post-stimulation, labeled osteoblast numbers (---) were increasing rapidly, while unlabeled osteoblast were constant. numbers (-)
300
J. A . Yee, D . B. Kimmel and W . S. S. Jee 0
0
‘0 .
x)
30
40
50
60
70
Hours post-stimulation
FIG.10. Multiple labeling index of osteoblasts following orthodontic stimulation. The multiple labeling index rose from 0 to 60% by 35 hr (six injections of 3H-TdR), then continued to rise more slowly toward 75% by 71 hr (twelve injections of ’H-TdR) post-stimulation.
The multiple labeling index of osteoblasts was 11 % at 11 hr post-stimulation, 31 % by 17 hr post-stimulation, and 43 % after 23 hr of stimulation. After reaching 56 % by 35 hr post-stimulation, the rise in the multiple labeling index became less marked, reaching its maximum of 75% at 71 hr post-stimulation. From 5 to 35 hr the slope of the best-fit line shows that the increase in the multiple labeling index of osteoblasts was 1.9 %/hr. From 35 hr on, the rate of increase was 0-4%/hr (Figs. 7 and 10). DISCUSSION The force applied to the tooth in this study is a tipping force, which results in complex changes in the PDL mesial to the tooth. A more uniform bone formation response could have been provided by a ‘bodily movement’ of the tooth (Grabet, 1972), but such orthodontics is too difficult to accomplish in rats’ mouths. A very exacting way to define the response to this tipping is by histologic observation at serial times following stimulation, as in this study. Despite the theory which would say that tipping results in both formation and resorption on both sides of the tooth, resorption of bone was not observed in any of the five areas examined after stimulation of up to 71 hr duration. Bone formation, however, occurred consistently in the two apical-most areas examined, and occasionally in the third area (Fig. 1). The labeling index study was performed to establish the time sequence of PDL fibroblast proliferative response to this stimulus to the periodontal ligament. In timing, the response of the PDL fibroblasts is quite similar to that of the liver parenchymal cells following partial hepatectomy; however, the magnitude of response is somewhat less than that seen in these same cells or in hotmonally-stimulated cells in the seminal vesicle (Grisham, 1962; Morley et af., 1973). However, it is a much quicker response to an initial insult than the plucking of the mouse skin, where the labeling index is about five times normal for 7 days post-pluck (Hegazy & Fowler, 1973). This stimulus to the PDL causes a complex change probably requiring two different lines of cellular differentiation for its return to the steady state. One
PDL cell kinetics
301
may provide new PDL fibroblasts which will repair stretched and tom fibers of the periodontal ligament, while the other may provide osteoblasts to the alveolar bone surface mesial to the tooth, which will form immature, woven bone. This particular stimulated system seems to be more like the partial hepatectomy and plucking models than the hormone-target tissues systems because it represents a lasting deformation of the tissue to which it must respond with some limited type of repair response. The multiple labeling index study of PDL fibroblasts establishes that the percentage of fibroblasts responding to stimulation of the PDL by synthesizing DNA in this model is about 50 %. It also confirms that the cell cycle time of these cells is at least 34 hr in duration (Roberts, Chase & Jee, 1974). Once again, the stimulated PDL has its own characteristics. In the plucked skin, the percentage of cells responding is 100 % and the cell cycle time, when measured by a multiple-labeling study, is about 37 hr (Hegazy & Fowler, 1973). The number of osteoblasts in the PDL is a measure of the enhancement of the bone formation aspect of this model. The identification of osteoblasts is a particularly good end point, as osteoblasts are morphologically and geographically identifiable. The rate of increase of the multiple labeling index of the PDL fibroblasts was 1.7 %/hr between 12 and 32 hr post-stimulation. In this interval, the rate of increase of the multiple labeling index of osteoblasts was 1-9%/hr. The similarities in these rates of labeling increase are indirect evidence that multipotent PDL fibroblasts are the source of osteoblasts forming bone in this model. It does not seem as likely that the stimulation could have evoked such a response in another population of cells not as closely allied with the PDL as the fibroblasts which reside there nativeIy. It is possible that if PDL fibroblasts which were osteogenic in nature could be identified, an even closer correlation between the labeling of PDL osteoblasts and a portion of the PDL fibroblasts could be established. Unfortunately, this distinction could not be made at the light microscope level. It is significant to note that few if any unlabeled osteoblasts evolve after 34 hr of stimulation. There may be two pathways for osteoblast differentiation following stimulation of the PDL, one direct, and one indirect. It is possible that the unlabeled osteoblasts came from cells which went through an S phase which was so short in duration that no 3H-TdR from the serial injections was available, or that they came from cells which were proliferating before the serial injections of 3H-TdR began. However, in choosing the spacing for the serial injections, the work of Roberts et al. (1974) and Kimmel (unpublished), which found that S phase for the PDL fibroblasts after stimulation was about 8 hr, was considered. With injections every 6 hr, it is considered unlikely that a significant number of cells were 'missed' during their S phases. While it remains possible that some cells which had proliferated shortly prior to stimulation were the source of the new osteoblasts, it is as likely that a significant portion of the osteoblasts which initially evolved in this stimulation (through the first 28 hr) differentiate from cells which do not go through an S phase immediately (less than 12 hr) prior to differentiation. Considering the similarities in function and ultrastructure of fibroblasts and osteoblasts, it seems reasonable to speculate that a PDL fibroblast could differentiate directly to an osteoblast. There is also some possibility that the unlabeled osteoblasts arise from fibroblasts which are in G,block (Gelfant & Smith, 1972), then are released by the stimulation to divide, then differentiate. The methods used in this study do not permit a distinction to be made between these two possibilities. The PDL fibroblasts of this study, which may include cells of both osteogenic and fibroblastic potential, are multipotent, reflecting the varied function of the PDL. For example, in
J . A. Yee, D. B. Kimmel and W.S. S. Jee
302
the PDL surrounding the continuously-erupting mouse incisor, the end result of PDL fibroblast proliferative and differentiation activity is the eruption of the tooth (Zajicek, 1974). In the current paper the ability of the PDL fibroblast to proliferate and differentiatein response to a tipping of the tooth is described. The PDL fibroblast populationis apparentlythe primary source of osteoblasts which form bone which stabilizes the tooth in its new position. It (the PDL fibroblast) is probably the main cell concerned with all movements and maintenance of the teeth. ACKNOWLEDGMENT
This work was supported by U.S.E.R.D.A. E( 11-1)-119 and NIH Grants DE-151 and GM958. REFERENCES
FROST, H.M. (1973) Bone Modeling und Skeletal Modeling Errors, p. 73. C. C. Thomas, Springfield, Illinois. GELFANT,S. & SMITH,J.G. (1972) Aging: noncycling cells an explanation. Science, 178,357. GRABER, T.M. (1972) Orthodontics:Principles and Prucrice, 3rd edn, p. 502. Saunders, Philadelphia. GRISHAM,J.W. (1962) A morphologic study of deoxyribonucleic acid synthesis and cell proliferation in regenerating rat liver; autoradiography with thymidine-H3. Cancer Res. 22,842. HAMILTON, E. & EN. C.W. (1972) Influence of hair plucking on the turnover time of the epidermal basal layer. Cell Tissue Kiner. 5, 505. HEGAZY, M.A.H. & FOWLER, J.F. (1973) Cell population kinetics of plucked and unplucked mouse skin. I. Unirradiatedskin.. Cell Tissue Kiner. 6,17. KEMBER, N.F.(1960) Cell division in endochondral ossification. A study of cell proliferation in rat bones by the method of tritiated thymidine autoradiography. J. Bone Jr Surg. 42-B.824. KIMMEL,D.B. & JEE.W.S.S. (1975) A rapid plastic embedding technique for preparation of three-micron thick sections of decalcified hard tissue. Stain Technol.50.83. MELCHER, A.H. & EASTOE,J.E. (1969) The connective tissue of the periodontium. Biology ofrhe Periodontiurn fed. by A. H. Melcher and W. H. Bowen), p. 167. Academic Press,New York. MORLEY, A.R., WRIGHT,N.A. & APPLETON.D. (1973) Cell proliferation in the castrate mouse seminal vesicle in response to testosterone propionate. I. Experimental observations. Cell Tissue Kinet. 6, 239. OWEN. M. (1963) Cell population kinetics of an osteogenic tissue. J. Cell Biol.19.19. ROBERTS. W.E.(1975) Cell kinetic nature and diurnal periodicity of the rat periodontal ligament. Arch. Oru Biol. 20,465. ROBERTS, W.E., CHASE. D.C.& JEE, W.S.S. (1974) Counts of labeled mitoses in the orthodontically-stimulated periodontal ligament in the rat. Arch. Oral Biol. 19,665. ROBERTS, W.E. &JEE.W.S.S. (1974) Cell kinetics oforthodontically-stimulated and non-stimulated periodontal ligament in the rat. Arch. Oral Biol. 19,17. TONNA, E.A. (1961) The cellular component of the skeletal system studied autoradiographically with tritiated thymidine (HTdR) during growth and aging. J. biophys. biochem. Cyrol. 9,813. TONNA, E.A. & CRONKITE, E.P. (1961) Cellular response to fracture studied with tritiated thymidine. J. Bone Jr Surg. &A, 352. TONNA, E.A. & WEIS. R. (1972) The cell proliferative activity of paradontal tissues in aging mice. Arch. Oral Biol. 17,969. WOLFF,C.M. (1968) Principles of Biometry. Van Nostrand, Princeton, New Jersey. YOUNG,R.W. (1962) Cell proliferation and specialization during endochondral osteogenesis. J. Cell Biol. 14, 357.
ZAJICEK,G . (1974) Fibroblast cell kinetics in the periodontal ligament of the mouse. Cell Tissue Kiner. 7,479.
