A Confocal Laser Scanning Microscopic Study
I l?. ( l. <ì.t
AN IMMUNOHISTOCHEMICAL STUDY OF MARMOSET PERIODONTAL LIGAM ENT MICROVASCU LATU RE
A CONFOCAL LASER SCANNING MICROSCOPIC STUDY
A research project submitted in partial fulfilment of the requirements for the degree of Master of Dental SurgerY
by
JONATHAN F. ASHWORTH B.D.S. (Adel.) *t* it r:ìîj
Orthodontic Unit School of Dentistry Faculty of Health Sciences The University of Adelaide South Australia
1 999 LI g. ?'
TABLE OF CONTENTS
Page No
List of Figures vii List of Tables X List of Abbreviations xi Summary xil Signed Statement XIV Acknowledgements XV
Chapter 1 INTRODUCTION 1
Chapter 2 AIMS AND HYPOTHESES 3
Chapter 3 REVIEW OF THE LITERATURE 4 3.1 The Morphology of the Periodontal Ligament (PDL) 4 3.2 The PDL Microvascular Bed 5 3.2.1 PDL Vascular SuPPIY 5 3.2.2 PDL Vessel Sizes/Classification 6 3.2.3 PDL Vessel Distribution 7 3.2.4 PDL VesselVolume 9 3.2.5 Monkey PDL Microvasculature 10 3.2.6 Functions of the PDL Microvasculature 11 3.2.7 Regulation of PDL Blood Flow 12 3.3 Periendothelial Cells 12 3.3.1 Periendothelial Cells in PDL Microvasculature 13 3.4 Alpha Smooth Muscle Actin 14 3.5 Mechanical Stimulation of PDL Tissues 15 3.5.1 Cellular BiologY 15 3.5.2 Vascular Changes 16 3.6 Angiogenesis 19 3.6.1 Control of Angiogenesis 20 3.6.2 Pericyte lnvolvement in Angiogenesis 21 3.7 Cytokines 22 ilt
3.8 Endothelin-1 23 3.8.1 Biosynthesis 24 3.8.2 Actions 25 3.8.3 Receptors 26 3.8.4 The Physiological Role of Endothelin-1 27 3.9 Endothelin-Related lnvestigations 28 3.9.1 Endothelin-Related lnvestigations in Non-Human Primates 28 3.9.2 lmmunohistochemical Labelling for Endothelin-1 in Dental Tissues 28
3. 1 0 lmmunofluorescence 30 3.10.1 Fluorescence MicroscoPY 32 3.10.2 Confocal Laser Scanning Microscopy 32 3.11 The Marmoset as an ExperimentalAnimal 33 3.11.1 The Marmoset Dentition 33
Chapter 4 MATERIALS AND METHODS 36 4.1 Summary 36 4.2 Ethics Approval 36 4.3 Research Colony 36 4.4 Summary of Pilot Studies 37 4.5 Pilot Study No 1 (RECA-1) 38 4.5.1 The Experimental Animal 38 4.5.2 Tissue Preparation and lmmunohistochemistry 38 4.6 Pilot Study No 2 (ET-1) 39 4.6.1 The Experimental Animal 40 4.6.2 Tissue Preparation and lmmunohistochemistry 40 4.7 Pilotstudy No 3 (QB-END-10) 41 4.8 Pilot Study No 4 (JC-704) 41 4.9 Pilot Study No 5 (cr-SMA) 41 4.10 Main Experiments 42 4.10.1 The Experimental Procedure 42 4.10.2 Loading Device and lntra-Oral Pad 42 4.10.3 Tissue PreParation 44 44 4.10.4 I mmunohistochemistrY 4.10.5 Examination of Mandibular sections and Data collection 47 IV
Chapter 5 FINDINGS 48 5.1 Summary of Pilot StudY Findings 48 5.1.1 RECA-1 48 5.1 .2 Endothelin-1 48 5.1 .3 Suitability of QB-END 10 as a Pan-endothelial lmmunolabel 49 5.1 .4 Suitability of JC-704 as a Pan-endothelial lmmunolabel 49 5.1.5 Suitability of c¿-sMA as a Periendothelial cell lmmunolabel 49 5.2 Pilot Study No 1 (RECA-1) 49 5.3 Pilot Study No 2 (ET-1) 50 5.3.1 Experiment A 50 5.3.2 Experiment B 50 5.3.3 Experiment C 50 5.3.4 Experiment D 51 5.3.5 Experiment E 51 5.3.6 Experiment F 52 5.3.7 Experiment G 52 5.4 Pilot Study No 3 (QB-END 10) 53 5.5 Pilot Study No 4 (JC-704) 53 5.6 Pilot Study No 5 (cr-SMA) 53 5.7 Summary of Findings of Main Experiments 58 5.7.1 Endothelin-1 58 5.7.2 JC-704 58 5.7.3 cr-SMA 58 5.8 Experiment No 1 (ET-1) 58
5.8.1 Experiment 1 58 5.8.2 Experiment 2 59 5.8.3 Experiment 3: unloaded and Loaded Mandibular sections 60 5.9 Experiment No 2 (JC-70A) 60 5.9.1 Vascular MorPhologY 60 5.9.2 Antigen ExPression 61 5.10 Experiment No 3 (a-SMA) 62
Chapter 6 DISCUSSION 77 6.1 Material 77 V
6.2 Load Application 77 6.3 lmmunohistochemical Methodology 78 6.3.1 Fixation 78 6.3.2 lmmunohistochemical Controls 78 6.3.3 Refinement of the Laboratory Methodology 79 6.3.4 Limitations of the Current Laboratory Methodology 80 6.4 lmmunohistochemical Labelling with RECA-1 81 6.5 lmmunohistochemical Labelling for ET-1 81 6.5.1 General Discussion 81 6.5.2 Pilot Study for ET-1 lmmunolabelling 83 6.5.3 Secondary Antibody for Anti-ET-1 85 6.5.4 ET-1 Antibody Cross-Reactivity 85 6.5.5 lmmunolabelling for ET-1 in DentalTissues 85 6.5.6 Production of ET-1 in the MVB of the unloaded and Loaded PDL 87 6.6 lmmunohistochemical Labelling for Endothelium 87 6.6.1 Pilot Study for Pan-endothelial lmmunolabel 87 6.6.2 Endothelial Cells in DentalTissues 88 6.6.2.1 Microvascular Morphology 88 6.6.2.2 Clinical lmPlications 89 6.6.2.3 Antigen ExPression 90 6.7 lmmunohistochemical Labelling for Periendothelial cells 91 6.7.1 Pilot Study for Periendothelial Cell lmmunolabel 91 6].2 Periendothelial Cells in Dental Tissues 91 6.8 Suggestions for Future Research 92 6.8.1 Endothelin-1 92 6.8.2 Morphological Studies 93 6.8.3 Angiogenesis 93
Chapter 7 CONCLUSIONS 95
Chapter 8 APPENDICES 97 8.1 Anaesthetic 97 8.2 Zamboni's Fixative 97 8.3 Phosphate Buffered Saline (PBS) 98 8.4 PBS l30% Sucrose I 0.1% Sodium Azide 98 VI
8.5 Dimethyl Sulphoxide (DMSO) 98 8.6 Coating Slides 99 8.7 Drying Slides with Tissue Sections 99 8.8 Antibody Diluent (0.02M) 99 8.9 Donkey Normal Serum 100 8.10 RECA-1 Antibody 100 8.1 1 Rabbit Anti-Endothelin-1 Antibody 100 8.12 Rabbit Anti-Big Endothelin-1 Antibody 100 8.13 Rabbit Anti-Pre-proendothelin Antibody 101 8.14 Monoclonal Anti-Endothelin Antibodies 101 8.15 Endothelial Cell Marker: QB-END 10 Antibody 101 8.16 Endothelial Cell Marker: Anti-PECAM-1 Antibody 101 8.17 Mouse MonoclonalAnti-Human Alpha-Smooth Muscle Actin Antibody 102 8.18 Cy3rM Conjugated DonkeyAnti-Mouse lgG Antibody 102 8.19 Cy3rM Conjugated Donkey Anti-Rabbit lgG Antibody 102 8.20 CySrM Conjugated Donkey Anti-Rabbit lgG Antibody 103 8.21 Dichlorotriazinyl Amino Fluorescein (DTAF) Antibody 103
I gG 104 8.22 Biolin-S P-Conj ug ated Affin i pu re Donkey Anti-Rabbit 8.23 Cy3rM Conjugated Streptavidin 104 8.24 lmmunohistochemical Labelling Procedure - Soft Tissues. lndirect Technique 105 8.25 lmmunohistochemical Labelling Procedure - Soft Tissues. Streptavidin-Biotin Am plification Techniq ue 105 8.26 lmmunohistochemical Labelling Procedure - Mandibular Sections. lndirect Technique 106
8.27 Chequerboard for Pilot Study No 1 106 8.28 Chequerboards for Pilot Study No 2 107 8.29 Chequerboard for Pilot Study No 3 108 8.30 Chequerboard for Pilot Study No 4 108 8.31 Chequerboard for Pilot Study No 5 108 8.32 Quantification of PECAM-1 Expression in PDL Vessels in Unloaded and Loaded First Molars 109
Chapter 9 REFERENCES 110 vil
LIST OF FIGURES
Figure Subject Page No.
1 The periodontal ligament and surrounding structures 4
2 Luminal volume across PDL circumferential thirds I
3 Blood volume of vessel groups as a percentage of the PDL I 4 Structure and amino acid sequence of endothelins 23
5 Factors that alter endothelin-1 synthesis and the pathway for endothelin-1 generation 25 6 The direct, indirect and streptavidin-biotin immunofluorescent techniques 31 7 Marmoset permanent dentition 34
8 Sagittal sections of marmoset permanent mandibular dentition 35 o Loading device and intra-oral pad 43 10 Marmoset with loading device and intra-oral pad in-situ 46
11 Marmoset mandibular body undergoing sagittal sectioning 46
12 Method of quantifying antigen expression 47 13 lmmunolabelling of RECA-1 antigen in rat kidney 54 14 lmmunolabelling of artery endothelium for Big ET-1 in kidney 54 15 lmmunolabelling of capillary loops for Big ET-1 in glomerulus of kidney 55 16 lmmunolabelling of peritubular capillaries for Pre-pro ET-1 in kidney 55 17 lmmunolabelling of epithelial cells for ET-1 in kidney 56
18 lmmunolabelling of endothelial cells for ET-1 in kidney 56 19 lmmunolabelling of endothelial cells with Jc-704 in stomach 57
20 lmmunolabelling of c¿-SMA in proximal small intestine 57
21 lmmunolabelling of vascular ET-1 within PDL 63 22 lmmunolabelling of postcapillary-sized venule within PDL for ET-1 63 23 lmmunolabelling of collecting venule within PDL for ET-1 64 24 lmmunolabelling of collecting venule bordering PDL and bone for ET-1 64 vlil
25 lmmunolabelling of terminal arteriole within alveolar bone for ET-1 65 26 Pan-endothelial immunolabelling in dentaltissues 66 27 lmmunolabelling of blood vessel traversing cortical plate of tooth socket 66 28 lmmunolabelling of gingival capillary loops 67 29 lntercellular pattern of immunolabelling of endothelial cells in PDL 67 30 Confocal image of pan-endothelial immunolabelling within dentaltissues 68 31 Confocal image of immunolabelled blood vessels traversing cortical plate of tooth sockets 68 32 Magnified confocal image of immunolabelled blood vessels traversing cortical plate of tooth socket 69 33 Confocal image of immunolabelled blood vessels in subapical region of mandible 69 34 Merged confocal image of immunolabelled endothelial cells of collecting venule in apical PDL of mandibular first molar 70 35 Merged confocal image of immunolabelled endothelial cells of collecting venule in PDL 70 36 Confocal images of apical vessels of unloaded molar selected for PECAM-1 quantification 71 37 Confocal images of apical vessels of loaded molar selected for PECAM-1 quantification 72 38 Confocal images of interradicular vessels of unloaded and loaded molars selected for PECAM-1 quantification 73
39 lmmunolabelled o-SMA within periendothelial cells in dental tissues 74 40 lmmunolabelling of cr-SMA within pericyte of collecting vein in alveolar bone showing cell body, nucleus and cytoplasmic processes 74
41 lmmunolabelled o-SMA within pericytes of collecting vein in alveolar bone 75 42 Confocal image of immunolabelled c¿-SMA within vascular smooth muscle cells of terminal arterioles in PDL 76 IX
43 Magnified confocal image of immunolabelled c¿-SMA within vascular smooth muscle cells of terminal arterioles in PDL 76 x
LIST OF TABLES
Table Subject Page No
1 Summary of studies on total vascular volume of the total PDL volume I 2 Summary of pilot studies 37 3 Summary of main experiments 45 XI
LIST OF ABBREVIATIONS
CLSM Confocal laser scanning microscope DMSO Dimethyl sulphoxide DNS Donkey normal serum DTAF Dichlorotriazinyl amino fluorescein ET-1 Endothelin-1 FITC Fluorescein isothiocyanate GCF Gingival crevicular fluid rHc lmmunohistochemistry tL-1 lnterleukin-1 MVB Microvascular bed PBS Phosphate buffered saline PDL Periodontal ligament rGFp Transforming growth factor B VEGF Vascular endothelial growth factor VSMC Vascular smooth muscle cells û,-SMA Alpha smooth muscle actin t1 Central incisor t2 Lateral incisor c Canine PM1 First premolar PM2 Second premolar PM3 Third premolar
M1 First permanent molar M2 Second permanent molar TA Terminal arteriole AVA Arteriovenous anastomosis AC Arterial capillary
P-VC Pericytic venous capillary a-PC Apericytic venous capillary p-PCV Pericytic postcapillary-sized venule A-PCV Apericytic postcapillary-sized venule CV Collecting venule xil
SUMMARY
The biological response to the application of an orthodontic force to a tooth involves (Davidovitch, various cell types of the periodontium assuming altered functional states lgg5). The vasculature of the periodontium undergoes functional and morphological peptide' alterations simultaneously along with other cellular elements' A 21-amino-acid potent vasoactive endothelin-1 (ET-1, Yanagisawa et al', 1988) is a cytokine with endothelium to properties that may play a role in the response of the PDL microvascular tooth loading.
intrusive occlusal load A custom designed device was constructed for the application of an to the buccal segment teeth of four anaesthetised marmosets for 1.5 hours' The investigations, the jaws, contralateral teeth served as controls. For immunohistochemical and fixed and samples of kidney, skeletal thigh muscle and gut, were extirpated immediately after the animals were euthanased'
wafering blade sagittal sections of each mandibular body were prepared using a diamond thick and mounted on a low speed sectioning saw. Each section was 150-200¡rm (lHC) was performed incorporated bone, periodontium and teeth. lmmunohistochemistry established on the undemineralised mandibular sections using laboratory methodologies with pilot studies.
The pilot studies comprised systematic chequerboard tests of indirect cells and immunohistochemical labelling for the identification of ET-1, endothelial muscle and gut' periendothelial cells on cryostat-cut sections of marmoset kidney, skeletal muscle actin (cr-sMA)' The primary antibodies were anti-ET-1, JC-704 and alpha smooth DTAF, FITC, lmmunoglobulin G secondary antibodies conjugated with the fluorochromes Streptavidin-biotin Cy!rur and CySrM were tested for optimum signal-to-noise ratios. not prove as amplification immunolabelled vascular endothelium for ET-1 but did satisfactory as the indirect IHC technique'
to ascertain the location Fluorescence and confocal laser scanning microscopy were used Antibody of immunolabelled vascular ET-1, endothelial and periendothelial cells. the penetration routinely occurred to a depth greater than 85¡rm from either side of mandibular sections' xilt
Endothelin-1 demonstrated an heterogeneous distribution within the microvasculature of the PDL and alveolar bone with the anti-ET-1 primary antibody and the lgG/CySrM secondary antibody at dilutions of 1:50. Vascular ET-1 was located within the walls of terminal arterioles, postcapillary-sized venules and collecting venules. lmmunolabelling for ET-1 occurred within the cytoplasm of endothelial cells and vascular smooth muscle cells. The presence of this potent vasoconstrictor suggests that the microvascular beds of the PDL and alveolar bone might possess contractile properties. The hypothesis that vascular ET-1 expression varies within the PDL in unloaded versus loaded conditions could not be established as insufficient data were generated for quantification. A method for future quantification of antigen expression was examined.
Endothelial cell, pericyte and vascular smooth muscle cell location and morphology was examined with the JC-704 (1:a0) and o-SMA (1:100) primary antibodies respectively in combination with lgG/Cy3rM secondary antibody (1:50). The intercellular distribution of endothelial cell immunolabelling enabled illustration of cellular size and shape. lmmunolabelling of intricate actin fibrils within pericytes and vascular smooth muscle cells provided further insight into periendothelial cell morphology. The confocal laser scanning microscope (CLSM) and CoMOS software (Bio-Rad, version 7.0a) enabled the construction of detailed three-dimensional cellular images by computer stacking of the optical sections.
The results of immunohistochemical investigations of vasoactive peptides such as ET-1 may eventually lead to pharmacological targeting of the PDL microvasculature to assist orthodontic tooth movement. Endothelin-1 may play a significant role during angiogenesis within the periodontium concomitant with orthodontic tooth movement. The administration of agents that influence the function of PECAM-1, the antigen to JC-704, and their effect on the rate of tooth movement might also be investigated. Further research utilising the marmoset experimental and control maxillary sections and the laboratory methodology established in the current study are warranted. XIV
SIGNED STATEMENT
This report contains no material that has been accepted for the award of any other degree or diploma in any other university. To the best of my knowledge and belief, it contains no material previously published except where due reference is made in the text.
I give consent to this copy of my thesis, when deposited in the University Library, being available for loan and photocopying.
Jonathan F. Ashworth
B.D.S. (Adel.) XV
ACKNOWLEDGEMENTS
I would like to thank the following people for their support during the last two years:
Professor M. R. Sims, Visiting Professor, The University of Sydney; Visiting Research Fellow, The University of Adelaide, for his guidance, patience and perseverance with this project, parlicularly during the times in whlch progress was slow.
Associate Professor B. J. Gannon, Department of Anatomy and Histology, Flinders University, for willingly sharing his expertise and for the use of his laboratory.
Professor W. J. Sampson, P. R. Begg Chair in Orthodontics, The University of Adelaide, for his encouragement and suggestions in the preparation of this thesis.
Dr C. W. Dreyer, Senior Lecturer in Orthodontics, The University of Adelaide, for his suggestions and direct laboratory assistance.
Dr. P. Kolesik, Manager, Confocal Facility, Department of Horticulture, Viticulture and Oenology, The University of Adelaide, for operation of the confocal microscope.
Dr. J. Clark, Manager of Animal Services, CSIRO Human Nutrition, for her expert handling of the marmosets.
Bev Manthey, Bone and Joint Research Laboratory, lnstitute of Medical and Veterinary Science, for assistance with the bone-cutting machine and stereomicroscope.
My friends and colleagues, Drs. C. Laparidis and P. Greatrex for their assistance and encouragement during the experiments.
My wife, Donna, for the ongoing love, patience and support she has demonstrated during the course of this research project.
I also wish to gratefully acknowledge the support for this project provided by the Australian Dental Research Fund and the Australian Society of Orthodontists Foundation for Research and Education. i i,. r.i rx \i
Chapter 1 INTRODUCTION orthodontic tooth movement is achieved by the clinical application of continuous tissues of mechanical forces on tooth crowns, which are transmitted to the surrounding the periodontium (sandy, 1gg2). The optimal magnitude of such forces has not been in the order resolved; however, storey (1973) deemed these forces'light'when they were ,heavy' force over a of 250gm, and at around 500gm. The application of an orthodontic prolonged period initiates inflammatory cellular and tissue responses within the subsequent periodontium, which lead to remodelling of the surrounding alveolar bone, and PDL to maintain movement of the tooth (Storey, 1973). lt is the ability of the cells of the during this and repair the ligament tissues and the adjacent alveolar bone and cementum (Berkovitz et al', inflammatory event that enables orthodontic tooth movement to occur 1995). A tooth undergoing orthodontic movement of 1.omm per month requires the PDL, which is complete relocation of the PDL every one to two weeks. Therefore, the orthodontic located between the tooth and bone, acts as the physiological mediator of within the tooth movement (Roberts, 1994). There is also a high degree of cellular activity pDL during physiological "drift" of the teeth, which affords preservation of its width and function (Melcher, 1989).
the Given the perceived mechanical nature of the practice of clinical orthodontics, biological technical aspect of the field, such as appliance type, has dominated over principles. Tissue response, particularly at the cellular and molecular level, has been given lower priority than the technical aspects of orthodontics and, as a result, a clear currently understanding of the biological events concomitant with tooth movement is lacking. Reflecting on research into tooth movement spanning twenty years' Storey stated that (1g73), in endeavouring to determine the ideal forces required to move teeth, "a relatively simple problem in metallurgy had now become a complex biologic response to engineering problem with hormonal overtones." Fortunately, the clinical responses at therapeutic orthodontic forces is most often favourable; however, variable movement' the cellular level sometimes cause undesired root resorption, or slow tooth periodontal A more complete understanding of cellular and molecular events within the alter clinical tissues concomitant with orthodontic tooth movement has the potential to 2 practice. Davidovitch (1991a) suggested that a comprehensive insight into these events would enable assessment and improvement of current clinical practices whilst facilitating avoidance of undesired responses. As a result of biochemical investigations, orthodontists are now able to effect pharmacologic modulation of tooth movement (Grieve et al., 1994). Analysis of the chemical make-up of the gingival crevicular fluid (GCF) may provide a noninvasive means by which the orthodontist could diagnose the biological status of the patient's periodontium. Prostaglandin E and interleukin-1p , two potent mediators of bone resorption, are detectable in the GCF at elevated levels in response to the application of an orthodontic force (Grieve ef al., 1994). Davidovitch (1gg1a) proposed that, via the GCF, an assessment of molecular markers of the intensity of cellular activity might be performed prior to the commencement of treatment. Such diagnostic tools would be of immense value; however, a clearer understanding of these cellular and molecular activities, via biological research, is needed before such ideas will have a significant impact on current clinical practice.
The biological response to orthodontic stimulation involves various cell types of the periodontium assuming altered functional states (Davidovitch, 1995). The vasculature of the PDL and surrounding bone undergoes functional and morphological alterations simultaneously along with other cellular elements of the periodontium. The present investigation was directed towards: 1. establishing a laboratory immunohistochemical technique to enable identification and evaluation of endothelial and contractile periendothelial cells within the microvasculature of the PDL and alveolar bone of marmoset monkeys and; Z. determining if a potent vasoconstrictive 21-amino acid peptide, ET-1, is produced within the microvasculature of the PDL, and if so, to investigate whether its production varies between unloaded, or resting, and loaded, or stimulated conditions within the PDL of marmosets.
The ability to identify ET-1, endothelium and periendothelial cells via immunohistochemistry may provide a means by which further data may be gathered regarding vascular cell behaviour during orthodontic tooth movement. lt is also anticipated that information regarding the vasoconstrictive potential of the microvasculature of the PDL and alveolar bone will contribute to a better understanding of the physiology of the periodontium. 3
Chapter 2 AIMS AND HYPOTHESES
The aims of this exPeriment were: j. To establish immunohistochemical labelling of ET-1 , endothelial cells and periendothelial cells in marmoset monkeys (callithrix iacchus)
2. To i) determine if vascular endothelial cells within the PDL of marmosets produce ET-1' and if so; ii) establish whether vascular endothelial cells within the PDL of marmosets vary in their production of ET-1 in unloaded versus loaded conditions.
3. To examine the morphology of the microvasculature within the paradental tissues of marmosets using immunohistochemical techniques'
The hypothesis of this experiment is that vascular ET-1 expression varies between the unloaded and the loaded PDL of the marmoset. 4
Chapter 3 REVIEW OF THE LITERATURE
3.1 The Morphology of the Periodontal Ligament
The components of the periodontium, the attachment apparatus of a tooth, are the cementum, the PDL, the bone lining the alveolus, and the dentogingival junction (Freeman, 1994). The PDL consists of a gel of ground substance containing cells, blood vessels and nerves, and a fìbrous stroma, which consists mainly of collagen, (predominately types I - 80 o/o, àÍrd lll - 15 %), and small amounts of oxytalan (Berkovitz, lgg9). lt is bordered by the surface of the cementum, the periodontal surface of the alveolarbone, and bythe lamina propria of thegingiva (Melcher, 1989) as represented in Figure 1. The embedded (Sharpey's) fibres within the PDL bind the cementum of the tooth to the bony walls of the alveolar socket. The average width of the PDL is 0.2mm; however, this may vary with position relative to the tooth root, the functional state of the tooth, and age (Berkovitz et a1.,1995). The main cells of the PDL arefibroblasts, sensory nerve cells, endothelial cells, and remnants of the developmental epithelial root sheath of Hertwig, which are identified as the epithelial rests of Malassez (Davidovitch, 1995). Cementoblasts are located on the tooth side of the PDL, and osteoblasts and osteoclasts on the alveolar bone side. Cells of the immune system, including macrophages, mast cells, lymphocytes and polymorphonuclear cells, are also present and enter the PDL via the vascular network (Davidovitch, 1995). The principal function of the PDL is to support the tooth in its socket, whilst permitting it to withstand masticatory forces. Proprioceptors are also present in the PDL to allow correct positioning of the mandible during function (Freeman, 1994).
Ercmd
Drnlim
Gngrc cqnæliw P E crorl liluo R I o Alrclor o bmo Dc¡lol o o N PUP Por¡o¿o.lol Figure 1. The periodontal ligament and ligomnl I I U surrounding structures. From Berkovitz et al. M Cmlum (1ee5). 5
3.2 The PDL Microvascular Bed
The PDL is a highly vascular connective tissue, which reflects the high rate of turnover of its cellular and extracellular elements (Freeman, 1994). The PDL of the mouse has a vascular volume three to seven times that of stereologically-investigated human connective tissues, such as fat, synovial membrane, knee tendon, and synovial capsule (Freezer and Sims, 1987). However, direct comparison is not possible due to the decreased relative vascularity of larger animals. The alveolar third of the mouse PDL and normal mouse muscle possess a similar level of vascularity (Freezer et al., 1987).
3.2.1 PDL Vascular SupPlY ln man, anatomical descriptions indicate that the arterial blood supply to the oral tissues arises from the maxillary artery, which is a terminal branch of the external carotid artery. The inferior alveolar artery gives off intraosseal ascending branches to the medullary cavity, the PDL and the pulps of the teeth. The veins follow the inferior alveolar artery, and drain posteriorly into the pterygoid venous plexus, or anteriorly to the anterior facial vein. The maxillary artery also gives origin to the posterior superior dental artery, which gives off smaller branches. These vessels form a plexus above the maxillary molar teeth to supply their PDL and pulp. The maxillary artery also gives rise to the infra-orbital artery, from which the anterior superior dental artery and sometimes the middle superior dental artery arise. These branches supply the PDL and pulp of the premolar and anterior teeth. Veins drain either anteriorly to the anterior facial vein, or posteriorly to the infraorbitalveins, orthe pterygoid venous plexus (Scott and Dixon, 1978).
The PDL microvascular bed (MVB) is supplied by three sources: i) the gingival vessels, ii) the horizontal transalveolar vessels through perforations in the cribriform plate of the tooth socket wall, iii) the apical vessels, which branch to enter the pulp or the PDL.
The dental tissues of primates are considered to provide an analogue for man in terms of structure and function (Levy, 1971 Wilson and Gardner, 1982). As in man, the arterial supply to the periodontium in Macaque monkeys is derived from the alveolar dental branches that arise from the inferior alveolar artery in the mandible, and the posterior 6 superior alveolar artery in the maxilla (Castelli and Dempster, 1965). The vascular supply to the periodontium of the upper and lower incisors, however, differs from that in man. A terminal branch of the sublingual artery traverses the mandible at the symphysis to supply the periodontium of the mandibular incisors, An ascending terminal branch of the greater palatine artery perforates the bone and provides the vascular supply to the palatal part of the periodontium of the maxillary incisors either side of the incisive foramen. Branches of the facial artery supply the labial part of the periodontium.
3.2.2 PDL Vessel Sizes/Classification ln response to the lack of agreement between topographical studies concerning small vessel terminology, Rhodin (1967 , 1968) provided a classification of microvasculature architecture based on studies performed in vivo, and in srTu after fixation, of the fascial microcirculatory bed of the medial thigh muscle of the rabbit. According to Rhodin's classification, arterioles are small arteries with an internal diameter of between approximately 100¡rm and 50pm, and are subdivided mainly on the basis of the number of smooth muscle layers of the media. Añerioles taper towards terminal añerioles, which have an internal diameter of less than 50pm, and only a single muscle layer. Precapillary sphincters are muscular arrangements that surround the orifices of the smaller branches that arise from terminal arterioles. The entire endothelial cell protrudes towards the lumen of the precapillary sphincter, reducing its internal diameter to about 15pm to 7pm.
Capillaries are essentially endothelial tubes, and are deemed the smallest vessels of the microcirculatory bed, although the dimensions may vary in different organs, tissues and species. lt is generally agreed that true capillaries have a luminal diameter not exceeding 8pm, which is slightly larger than red blood cells. Venous capillaries are continuous with the larger postcapillary venules, and form when a capillary enlarges or converges with another. Postcapillary venules have a luminal diameter of about 8-30pm. Concomitant with increasing luminal diameter, there is a gradual increase in the number of pericytes. Periendothelial cells, including pericytes are reviewed on Page 12. Collecting venules have a luminal diameter of 30-50pm and are surrounded by a complete layer of pericytes in addition to occasional primitive smooth muscle cells. Muscular venules have a luminal diameter of 50-100pm. The periendothelial cells of these vessels are smooth muscle cells, which at times form two layers. Small collecting verns have 100-300pm internal diameters, with a prominent media consisting of continuous layers of smooth muscle cells. 7
(the inner luminal Variations in the criteria used for classification of the microvasculature tube, and the ultrastructural diameter, the number of cell layers around the endothelial gradually, and intermediate stages characteristics of the periendothelial cell layers) occur may be found. The observed vessel diameter may vary due to autoregulatory differences (weideman' mechanisms in the living animal (Rhodin, 1967) and species
1 e84).
of microvascular This nomencrature provided a basis for a universal classification varied according to location architecture; however, it was recognised that this architecture by Rhodin have been within the body (Rhodin, 1967). The following vessels described identified in the PDL (Sims, 1995): (AVA): luminal diameter of i) Terminal Arterioles (TA) and Arteriovenous Anastomoses 11-60Pm. ii)ArterialCapillaries(AC):luminaldiameterofl.StoT.opm. Pericytic (p-Vc) or apericytic iii) Venous Capi¡aries: luminal diameter of 1.5 - 10¡rm. (a-VC). Pericytic (p-PcV) or iv)postcapillary-Sized Venules: luminal diameter of 10-30pm. apericYtic (a-PCV). v)CollectingVenules(CV):luminaldiameterof30-160um. or apericytic at various sites sims (1986), specified that the VC and PCV may be pericytic along the same vessel.
3.2.3 PDL Vessel Distribution are distributed parallel to the The pDL microvasculature forms a network of vessels that PDL, however, varies according long axis of the tooth. Blood vessel distribution within the and proximity to tooth or to species, tooth type, depth from cervical to apical regions, alveolar bone.
incisor (chintavalakorn' 1993) the ln the mouse molar (sims et a1.,1996) and marmoset third, followed by the brood vorume distribution is greatest in the middre circumferential ln the human premolar bone circumferential third, and the tooth circumferential third' bone circumferential third' (Foong, 1993), the blood volume distribution is greatest in the circumferential third (Figure 2)' followed by the middle circumferentialthird, and the tooth I
Eman Emarmoset I mouse o E õf d) s I 0 tooth 1/3 middle 1/3 bone 113 PDL circumferential thirds
Figure 2. Luminal volume across the PDL circumferential thirds for the mouse molar, marmoset incisor, and human premolar. From Sims M. R. (1995).
There is also variation in distribution according to vessel category. In the young mouse molar PDL, 11 per cent of the a-PCV, (the predominant vessel type), occurs in the tooth circumferential third, 23 per cent in the middle third, and 66 per cent in the bone third (Freezer et al., 1987). Other vessel categories differ in their distributions across circumferential thirds.
The microvasculature of the PDL in the mouse, marmoset and man consists predominantly of venous elements (Figure 3). I Eman 7 gmarmoset 6 mouse o I E 5 E o 4 oJ À 3 s 2
'l
0 VC PCV AC+TA CV vessel type
Figure 3. Blood volume of vessel groups as a percentage of the PDL for the mouse molar, marmoset incisor, and human premolar. VC: venous capillaries. PCV: postcapillary-sized venules. AC + TA: arterialcapillaries and terminalarteríoles. CV: collecting venules. From Sims M.R. (1995). I
3.2.4 PDL Vessel Volume species, site, The volume of the microvasculature of the PDL will vary according to of function, and age method of tissue preparation for analysis, method of analysis, extent volume as (sims, 1995). Cameron (1995), provided a summary on the total vascular bed species (Table 1)' a percentage of the total PDL volume for different sites in different
percentage of the total PDL Table 1. summary of studies on the total vascular bed volume as a (1995)' volume for various sites in ditferent species. From Cameron J' Stud Vascular Vol PDL Site Method Mouse LM Gould et al. (1977); .7o/o (0.60/o),7 Md first molar (mesial) 7 '3o/o & Melcher (1983) (0.8%) McCulloch Sims (1980) 17% Md first & second molars LM TEM (1984); Sims ef a/. 7.5o/o (1o/o'); 8.5Yo (1'4Yo) Md first molar (mesial) Freezer (1 ee2) Sims 198 10.9% Md first molar LM Rat Moxham ef a/. (1985) 22.1% (5.6%) Md first molar TEM a/. (1989); Clark 20.0%to23.0% Mx first molar aPex TEM Lew ef (1 e86) Sims blished 10.2% lo 24.60/o Mx first molar TEM Monkey LM Wills ef a/.(1976) 0.5% lo 1o/o Md incisors Douvartzidis (1983) 8.3% (0.4%) Md second molar LM Crowe (1988); Weir (1990) 2.6%to 19.7o/o',10.8% Mx incisor aPex TEM Pa e & Sims 1 11.26% 72% Mx incisor TEM Human Götze (1976, 1980). 1.63% to 3.5% Anterior & Premolar teeth LM Sims (1980) 11% Md premolars LM 1993 8.97%to9.52% Md distal TEM F *cited Parlange & ms, 1993 PDL: periodontal igament. maxil mandible. in sE: standard error LM: light microscope. TEM: transmission electron microscope'
bed (MVB) with sims ef a/. (1996) described changes in the volume of the microvascular TA, and CV in aged mice age in the mouse. lncreased luminal volumes in the PDL AVA, The PDL area served by resulted in greater volumes of blood bypassing the capillary bed. that the total number of capillaries decreased in the aged mice. lt was suggested AVA and TA to CV was decreases in vascular efficiency caused by blood shunting from offset by reduced diffusion distances. 10
3.2.5 Monkey PDL Microvasculature ln the periodontium of the maxillary central incisor teeth of Macaque monkeys, Castelli ef a/. (1965) described the entry of arterioles into the periodontal space in the apical region and the adjacent two thirds of the alveolus via perforations in the bony cribriform plate. The arterioles immediately branched to form capillaries with a polyhedric plexiform pattern oriented parallel to the long axis of the tooth. The arterioles that perforated the cribriform plate had a diameter of less than 100pm. Capillaries of 9 to 1Opm in diameter formed a layer close to the cementum, but the exact position of this vascular layer relative to the cementum was not indicated. The venules formed an irregular meshwork closer to the alveolar wall than the capillary layer. The venous plexuses were coarser and less regularly oriented in the apical and cervical thirds of the periodontium than the middle third. Venous drainage was directed towards the dental apex, and through foramina in the alveolar bone to networks within the bone marrow. Venous structures in the apical area anastomosed with those draining the dental pulp to form a venous cap in the marrow space surrounding the apex of the alveolus.
Using a corrosion casting technique, Lee (1987) described the marmoset PDL vasculature in the premolar as consisting of mainly venous elements. These comprised a loose network of occluso-apically oriented collecting venules 30-50pm in diameter that frequently anastomosed. Postcapillary-sized venules 8-30pm in diameter were located at the cervical end and capillaries were limited to the most cervical part of the PDL. There were frequent anastomoses between the PDL vessels and the gingival vasculature, Also, branches traversing the cribriform plate of the tooth socket contributed to the MVB. Terminal arterioles supplied the PDL microvasculature, but only coursed for a short distance within the ligament. Arterioles were scarce in the marmoset premolar PDL. The predominance of venous structures in the marmoset PDL was suppofted by Parlange (1991), who found venous elements to comprise 90.59% of the total vascular pool of the marmoset incisor PDL. ln examining the corrosion casts processed by Lee (1987), Stanley (1993) found the PDL microvasculature of the anterior dental segments of the marmoset to be generally oriented occluso-apically, with frequent intercommunications. Postcapillary-sized and collecting- sized venules were the predominant vessel types in the cervical and middle thirds of the PDL. The apical third was more vascular, with collecting-sized venules being the most 11
common than venous common vessel types. Arterial vessels were also found to be less the apex towards the vessels in the pDLs of the anterior teeth and ran predominantly from from the bone and cervical third. some arterial vessels, however, were found to emerge that vessel type gingiva to supply the cervical third of the PDL. Stanley (1993) suggested studies evaluating these and vascular volume may vary around the tooth socket and that of the other parameters from one aspect of the tooth socket may not be representative sides.
(1993) investigations The method of bone corrosion used for Lee's (1987) and Stanley's network was meant that it could not be ascertained as to whether the PDL vascular the mandibular rocated croser to the tooth or the bone. Douvartzidis (1g83) found in area was located second molar of the marmoset that the greatest percentage of vascular third, then the tooth in the middle circumferential third of the PDL, followed by the bone of the PDL, third. Vertically, the area of greatest vascularity was in the cervical third followed by the apical and middle thirds respectively'
3.2.6 Functions of the PDL Microvasculature in the PDL is to Edwall and Aars (1gg5) stated that the primary role of the vasculature and the PDL tissues, provide exchanges of substrates and metabolites between the blood eruption (Berkovitz, including the dentine. The vasculature may also play a role in tooth and recovery lggo) and sensory functions. The biphasic pattern of tooth intrusion (1966) to suggestthat following the application and removal of an intrusive load led Bien support' The the PDL vasculature provides an haemodynamic mechanism of tooth view is precise mechanism of tooth support, however, is unknown and the contemporary 1995)' Sasano that many components of the PDL are involved (Moxham and Berkovitz, displacement ef a/. (1992) proposed that the tooth has the ability to undergo abrupt blood from pressure following the application of external forces via the movement of of fluid between vascular and zones to tension zones or out of the alveolus. Movement under loading (Moxham ef extravascular components may also influence tooth movement following the application a/., 1995). Ghanges in PDL blood flow and vascular morphology described on Page 16' of prolonged (orthodontic) forces have also been observed and are 12
3.2.7 Regulation of PDL Blood Flow Blood circulation in the PDL in physiological states is influenced powerfully by the sympathetic nervous system, and a reduction in blood flow during stimulation of sympathetic nerves occurs mainly via o-adrenoceptors mediating constriction of precapillary vessels. Adrenergic control of PDL postcapillary resistance vessels occurs through activation of B2-adrenoceptors causing vasodilation (Edwall et al., 1995). Afferent sensory nerve fibres containing substance P and calcitonin gene-related peptide are involved in the transmission of pain, and have been found to be associated with blood vessels in the apical and middle part of the PDL (Kato et al., 1992). Release of these peptides causes neurogenic inflammation involving vasodilation and increased vascular permeability (Edwall ef a/., 1995).
There is conflicting evidence as to the existence of local autoregulatory mechanisms in the PDL circulation. However, if such mechanisms exist, they appear to be overridden by sudden changes in systemic arterial pressure and sympathetic nerve activity (Edwall ef a/., 1995).
3.3 Periendothelial Gells
Rhodin (1967, 1968), in his description of the ultrastructure of the subdermal fascial microvessels of the medial rabbit thigh, included several periendothelial cell types: pericytes, veil cells, primitive smooth muscle cells, and smooth muscle cells. Zimmermann (1923) coined the term "pericyte" and was the first investigatorto carefully describe their structure. They have been variously termed mural cells, adventitial cells, pericapillary cellular investments, Rouget cells and intramural pericytes (Sims, 1986). They are found on the microvessels, particularly the venous elements (Rhodin, 1968). Pericytes possess abundant cytoplasmic processes, which surround the endothelial tube of microvessels. The presence of a basement membrane around a cell that is continuous with the basement membrane of the vascular wall provides a means by which pericyte processes may be distinguished from those of adjacent fibroblasts (Sims, 1986). Dense filament bundles are located near the nucleus and in the cytoplasmic processes and have been reported to contain the proteins o-SMA (Skalli et al., 1989), myosin (LeBeux and Willemot, 1978), and tropomyosin (Joyce et al., 1985). However, Nehls and Drenckhahn (1991) found that pericytes of "true" capillaries lacked cr,-SMA, whilst pericytes of the 13 pre- and postcapillary microvessels expressed this isoform. These authors designated "true" capillaries, or "midcapillaries" as those microvascular segments interposed between pre- and postcapillaries. Pericytes make frequent membranous contacts with the endothelium, but rarely contact each other. The precise functions of pericytes are not clear. They may play a role in microvessel contraction, capillary sprouting during wound healing, structural support of the microvessel wall and the regulation of microvessel macromolecular leakage (Sims, 1986). According to Rhodin (1968), pericytes represent one of the few largely undifferentiated cells in the adult, an observation he rationalised by their association with the capillary bed, a structure which can be stimulated to grow. Rhodin (1968) suggested that as pericytes resemble fibroblasts ultrastructurally, but are rather undifferentiated, they serve as precursors to smooth muscle cells when function demands such differentiation.
Veil cells were described by Rhodin (1968) as fibroblasts with a function in the venous system of laying down collagenous fibres around the microvessels, which, together with the pericytes, contribute to stabilisation of the venous walls. Smooth muscle cells are associated with the arterial and venous vessel walls, are surrounded by a basement membrane, and occur either singularly, or as layers surrounding the endothelium. Primitive smooth muscle cells were seen by Rhodin (1968) as intermediates between pericytes and smooth muscle cells.
Rhodin (1968) described the occurrence of myofilaments within the different cells of the vascular wall as ranging from "none" in the veil cell, "feW' in the endothelium and pericyte, "moderate" in the primative smooth muscle cell, to "abundant" in the mature smooth muscle cell.
3.3.1 Periendothelial Cells in PDL Microvasculature ln the PDL mesial to the mandibular first molar of the mouse, Freezer (1984) found pericytes to be elongate, with little cyoplasmic branching, few contacts with the endothelium, and a basement membrane continuous with the endothelium. They were most commonly identified around periodontal vessels possessing a luminal diameter of 4 to 8¡lm, and once vessels reached a luminal diameter exceeding 20pm, investment with pericytes became incomplete or absent. 14 ln a TEM stereological analysis, Parlange (1991) found few vessels in the marmoset pericytes were ditficult incisor pDL to possess complete pericytic coverings and noted that 4'74% and to distinguish from fibroblasts. Pericytic PCV and apericytic PCV comprised capillaries were 42.51% of the total vascular pool respectively. No pericytic venous identified.
3.4 Alpha Smooth Muscle Actin
Gibson, 1976; Actin is a eukaryotic protein expressed in mammals and birds (Garrels and storti and Rich, 1976). At least six isoforms can be distinguished on the basis of Weber, 1978)' differences in their amino-terminal tryptic peptides (Vandekerckhove and skeletal and Four isoforms occur in muscle tissues: cr-skeletal and o-cardiac actin in cells' Two cardiac muscle cells, and cr- and y-smooth muscle actin in smooth muscle as and forms of non-muscle actin are found in virtually all cells, and are referred to B- of tissue y-cytoplasmic actin. variation in actin isoforms appears to exist on the basis rather than species specificity (Vandekerckhove et al., 1978).
isoform' Skalli ef a/. (19g6) introduced a monoclonal antibody specific for the 0-SMA cells, c¿-SMA was Within microfilamentous bundles of contractile vascular smooth muscle using this antibody identified as the major actin isoform by immunofluorescent techniques (skalli et a:.,1987). Later, skalli ef a/. (19S9) reported that pericytes also contain o-sMA had only been in microfilamentous bundles, and suggested that as this actin isoform properties' previously found in contractile cells, pericytes may also possess contractile "true" vessels were devoid of Nehls et at. (1991) demonstated that pericytes of capillary positive for ü-sMA' ü-sMA, whereas pericytes of pre- and postcapillary vessels were pre- and Therefore, regulation of blood flow by contraction may be a function of have other postcapillary pericytes, while non-muscle pericytes of true capillaries may functions. Pericytes ñìay, therefore, represent an heterogeneous cell population' (1991) did not conform to lnterestingly, use of the term "precapillary vessel" by Nehls et al' Rhodin's almost universal classification of microvascular morphology'
during angiogenesis did Later, Nehls et at. (1gg2) found that young pericytes developing represent the not express c¿-sMA, and suggested that pericytes of true capillaries may to youngest of the capillary network. Thereafter, microvascular networks may expand 15 form pre- and postcapillary segments and the associated pericytes may begin to express c¿-SMA. Nehls et at. (1992) suggested that pericytes of true capillaries and capillary sprouts could be distinguished from other pericytes by demonstrating a failure to express cr-SMA whilst labelling positively for another protein, desmin.
3.5 Mechanical Stimulation of PDL Tissues
3.5.1 Cellular Biology Sandstedt (190a; 1905) was the first investigator to report bone resorption in the areas of pDL compression and bone deposition in the stretched PDL following his experiment involving orthodontic force application for 3 weeks in a dog. This led to the concept that the distribution of pressure and tension within the periodontium determines the pattern of alveolar bone remodelling, and has traditionally, if somewhat simplistically, accounted for orthodontic tooth movement. Many detailed light microscopic investigations of the morphologic tissue changes concomitant with various orthodontic force systems have been undertaken this century. Reitan (1951) observed the proliferation of PDL cells in areas of tension and remodelling on the endosteal sudaces of the alveolar bone. Storey (1973) highlighted the inflammatory nature of the PDL response to heavier, "biodisruptive" orthodontic forces. Whilst not explaining the mechanisms of transduction of physical stimuli to cellular reactions, descriptive light microscopic investigations such as these have provided insight into which tissues warrant further research (Davidovitch, 1991b). Following the work of Sandstedt, Reitan, Storey, and many other investigators, the development of biological research methods such as electron microscopy, autoradiography and immunohistochemistry have enabled research to be conducted at the cellular and molecular levels into the effect of mechanical stress on cellular activity within the periodontium.
Research techniques that have made possible the isolation and culture of monolayers of single cell types, together with methods to alter cell shape, suggested that a relationship exists between cell shape and metabolic activity (Sandy, 1992). Davidovitch (1991b) stated that deformations in cellular shape occur in mechanically stressed periodontal tissues. Osteoblasts, which normally appear flat, become large and round in areas of PDL tension, where they have been found to produce new bone matrix. ln areas of compression, fibroblasts become round, and are associated with enzymic degradation 16 of extracellular matrix. Mechanosensitive cell types such as these, when distorted by load application, undergo a rapid metabolic response (Davidovitch, 1995). For example, membrane channels, which can regulate ion traffic into and out of the cell, may be mechanosensitive (Morris, 1990). Some of these channels may be turned on, whilst others are turned off. Calcium ions are co-factors in the activity of many enzymes, and membrane channels are sites of entry of these ions into cells (Davidovitch, 1991b).
Cellular function may also be altered by mechanical forces causing structural changes to the extracellular matrix (ECM), which envelops membrane receptors, such as integrins. The integrins are proteinaceous molecules which traverse the plasma membrane and connect to the ECM on one side and indirectly to the cytoskeleton on the other (Sandy ef a/., 1993). Cell volumetric changes may be partly regulated by membrane ion channels, and the resultant stretching of the cytoskeleton causes cytoplasmic and nuclear enzymic activation (Sachs, 1988). The cytoskeleton controls cell surface receptors, the secretion of cytokines, ribosome function, and gene transcription (Nathan and Sporn, 1991). Mechanical forces may also elicit piezoelectric potentials (Bassett and Becker, 1962) that activate cells by altering their membrane potential (Davidovitch, 1995). lnteractions between physical forces and cell function have been observed in lymphocytes (Cogoli ef al., 1980), arterial smooth muscle cells (Leung et al., 1976) and endothelial cells (Lansman et al., 1987).
Periodontal cellular activation via mechanical forces is followed by a response stage, which involves the release of signal molecules, such as the cytokines interleukin-1cr and -18, neurotransmitters such as substance P, vasoactive intestinal polypeptide, prostaglandins and growth factors. Localisation of these molecules has been made possible by immunohistochemistry (Davidovitch, 1995). Cytokines and their possible role in orthodontic tooth movement are discussed on Page 22.
3.5.2 Vascular Ghanges It has been suggested that a close relationship exists between the vascular integrity of the PDL and the type of bone resorption (frontal or undermining) that occurs subsequent to the application of an orthodontic force (Gianelly, 1969). The distinction between frontal and undermining bone resorption is made on the basis of the pattern of bone removal. The stimulation of frontal resorption of the alveolar bone, which is usually associated with 17 the application of light forces and rapid orthodontic tooth movement, occurs adjacent to the areas of PDL under pressure. Undermining resorption of the bone occurs in the marrow spaces adjacent to pressure areas of the PDL and is associated with the application of heavy forces and delayed tooth movement. Force induced frontal resorption has been associated with a patent PDL vasculature whereas undermining resorption has been associated with PDL vascular occlusion (Gianelly, 1969).
Histologic studies using electron and light microscopic techniques have revealed acute inflammatory changes in the PDL during the early phases of orthodontic force application, in areas of both pressure and tension (Rygh et al., 1986). Changes in the vasculature of the PDL and alveolar bone during tooth movement have also been observed.
ln areas of relatively light pressure, where frontal resorption is to occur, proliferation of vascular structures, or vascular "activation" takes place (Rygh, 1984). An adequate vascular supply is required for the differentiation of specialised cells involved in tissue remodelling (Brudvik and Rygh, 1995).
When pressure in localised regions of the PDL exceeds the optimum, there are vascular occluding effects. "Hyalinisation" (a light microscopic term) occurs even with the application of small continuous forces and results in the development of areas of aseptic necrotic tissue localised to the PDL (Reitan, 1951). An inflammatory reaction develops at their periphery (Rygh, 1984). Lilja ef a/. (1983) observed dilated blood vessels nearthe
hyalinised zone throughout their experiment where the force application extended from 1 to 30 days. Brudvik and Rygh (1994) reported that these necrotic areas are removed with the simultaneous invasion of macrophage-like multinucleated cells and blood vessels originating from the inflammed peripheral tissues. lt has been proposed by Brudvik and Rygh (1993a; 1993b) that root resorption may be a side effect of the activity of these invading cells. Rygh (1974) suggested that macrophages play an important role in the removal of hyalinised tissue by phagocytic processes. The migration of leukocytes into areas of tension and frontal bone resorption is facilitated by capillary endothelial cells. This ingress of leukocytes, even in the absence of hyalinisation, suggests that these cells influence PDL remodelling (Rygh et a|.,1986).
Following the removal of tissue damaged by compression, invading fibroblast-like cells located behind the initial invading frontier appear to be involved in the processes of tissue 18 removal and synthesis (Rygh, 1gS4). Osteoclasts on the endosteal surface remove bone Therefore, and, following the removal of hyalinised tissue, tooth movement begins again. orthodontic tooth movement frequently involves intermittant processes of tissue damage, inflammation and rePair.
Rygh ef a/. (1g86) observed that there is also increased vascularisation in areas of tension, most commonly in the middle of the PDL and towards the alveolar bone. Rygh of and Brudvik (1gg5) reported that as the volume of blood vessels increased, the volume of collagen fibres running from tooth to alveolar bone reduced, possibly via the actions extracellular collagenases produced by macrophage-fibroblast interaction.
on first Murrell ef a/. (1g96) found that the application and removal of orthodontic force the PDL blood maxillary molar tooth of rats produced changes in the number and density of vessels. Vascular distribution and density increased after the application of orthodontic position, force, decreased following the removal of force, increased during relapse of tooth and normalised after an interval equivalent to the duration of orthodontic force application' These investigators suggested that as normalisation occurred following an interval equivalent to the duration of orthodontic force application, the PDL microvasculature may play a role in modulating interstitial tissue pressures, resulting in the production of tissue forces and thereby causing relapse of orthodontically moved teeth'
Tang and Sims (1992), in a TEM investigation, observed changes in the number and periodontal distribution of tissue channels (TC) in the normal and tensioned rat molar ligament. The water-rich sol phase ground substance in connective tissues was (the described as comprising a system of TC that form a fine, loosely connected network TC ultracirculation), which conducts the bulk-flow of fluid throughout the tissues. These investigators concluded that their experiment provided evidence for substantial fluid circulation between the MVB and the connective tissue compartment via the TC of TC ultracirculation. Orthodontic tension resulted in a marked increase in the number extending from the PDL MVB.
in local blood Matsuo et at. (1gg7) also found that orthodontic forces caused alterations pressure and vascular distribution in the PDL of the second premolar of the dog' After network on one day of orthodontic force application, an area of the periodontal vascular "exposed", or devoid the pressure side became occluded and the alveolar bone appeared 19 of an overlying PDL vascular network, when compared with other areas of bone that had an overlying network of patent vessels. After three days of force application, capillary loops appeared on the periphery of the avascular area. After 14 days, undermining resorption was observed subjacent to the exposed area of bone, along with the proliferation of vessels in the marrow space and the capillary loops in the PDL. Vessel proliferation finally led to the disappearance of the avascular area. These morphological associations between the proliferating vascular network and resorption of the alveolar bone would fuñher demonstrate the fundamental role the MVB of the periodontium plays in orthodontic tooth movement.
These investigations have provided descriptions of morphological changes observed at the light and electron microscopic level in the MVB of the PDL following mechanical stimulation. Advances in laboratory techniques have also enabled investigation of the dynamic aspects of blood circulation in the oral tissues. Kvinnsland et al. (1989) used fluorescent microspheres to visualise and assess PDL and pulpal blood flow in the rat following the application of a mesially directed force of 30-509m to the maxillary molar for 5 days. They reported a substantial increase in blood flow rate in the PDL on the mesial and, to a lesser extent, distal aspects of the experimental tooth roots compared with the contralateral control molar. lt was hypothesised that an increase in blood flow is necessary following an orthodontic stimulus to support the increased cellular activity concomitant with bony appositional and resorptive processes. An increase in blood flow rate is dependent on vasodilation and an absence of gross leakage, both of which are basic vascular responses to inflammation. Kvinnsland et al. (1989) suggested that at this specific stage of orthodontic treatment in rats, i.e. 5 days, there was a general vasodilation to allow an increased blood flow rate, but that vascular responses to orthodontic stimuli were dynamic, and possibly differed before and after this period compared to those found in their investigation.
3.6 Angiogenesis
The foregoing investigations demonstrated the predominance of vascular proliferation in the mechanically stimulated periodontium in areas of pressure and tension, and during the processes of frontal and undermining resorption. Furthermore, inflammatory processes together with the removal of hyalinised areas, appear to be fundamental mechanisms 20 underlying orthodontic tooth movement. Angiogenesis, or neovascularisation, is the term applied to the rapid ingrowth of new capillary blood vessels from pre-existing vessels, and is a physiological event that is involved in inflammation and tissue regeneration (Polverini, 1995). lncreased nutritional requirements of regenerating tissues necessitate such capillary proliferation. Blood vessel proliferation is also essential to other physiological and pathological processes. Angiogenesis is associated with bone mineralisation, preceding endochondral ossification and bone fracture repair (Trueta, 1963). Folkman (1971) suggested that solid tumours are dependent on angiogenesis for sustained growth. Therefore, considerable interest exists in developing the therapeutic potential of inhibition of tumour neovacularisation. The formation of new blood vessels is also frequently observed during embryogenesis and is termed vasculogenesis (Risau ef a/., 1988). lnitially, angiogenesis involves disruption of contacts between endothelial cells, pericytes, and smooth muscle cells, and migration of endothelial cells through the partially disintegrated basement membrane and extracellular matrix. Stimulators of angiogenesis, including specific growth factors and cytokines, are secreted by a variety of cells, and induce endothelial cells to divide and form tubular structures. Maturation then involves the re-establishment of the basement membrane (Polverini, 1995). Newly formed capillaries remain according to metabolic demands, then regress as tissue regeneration is completed, and the inflammatory stimulus subsides.
3.6.1 Gontrol of Angiogenes¡s The regulation of angiogenesis is complex and not completely understood (Höckel et al., 1993). However, macrophages are thought to play a key role in inflammatory and tumour angiogenesis as they are: i) ubiquitous in all tissues and may be recruited from blood- borne monocytes during inflammation, ii) functionally heterogeneous and can be activated to stimulate angiogenic activity, iii) able to secrete products involved in both the stimulation and down-regulation of angiogenesis (Sunderkötter et al., 1994). Activated macrophages may influence angiogenesis by secreting pro-angiogenic factors including proteases, growth factors [acidic and basic fibroblast growth factors, vascular endothelial growth factor (VEGF), transforming growth factors (TGFcr, TGFp) and platelet derived growth factorl, interleukin-1 (lL-1), interleukin-6, interleukin-8, substance P and prostaglandins. Macrophages may also secrete angiostatic factors including interferons and thrombospondin-1 (Sunderkötter et al., 1994). Vessel formation requires both increased production of pro-angiogenic factors and downregulation of circulating 21
control of angiogenesis may angiogenic inhibitors (e.g. thrombospondin and angiostatin). stimulators (suggesting local be imposed by the relatively short half-lives of angiogenic (wong et al', 1996)' Significant activity) and the longer half-lives of angiogenic inhibitors substances can be potential benefits may in future be derived if selected angiogenic 1993)' leading to utilised to stimulate "therapeutic angiogenesis" (Höckel et al'' into the PDL, along with enhancement of wound healing. The ingress of macrophages stimulated periodontium' angiogenesis, is an histologic feature of the orthodontically to suggest that foreskin using an in vitromodel, Lewis (1996) found preliminary evidence large vein endothelial cells are highly microvascular endothelial cells and human umbilical of the vessel network formed sensitive to tensile forces, displaying enhanced maturation on the basis that vascular in matrigel under such conditions' From his findings, and late phase of angiogenesis network formation in matrigel constitutes a model for the forces may also play a (Haralabopoulos et a:.,1994), he argued that tensile orthodontic role in vessel angiogenesis.
3.6.2Pericyte lnvolvement in Angiogenesis as to the role of pericytes in currenily, there is a lack of consensus in the literature that the formation of a basement capillary sprouting. Blood and Zetter (1990) concluded pericytes is generally associated with the membrane and investment of capillaries with (1992) identified pericytes at the end of capillary proliferation. However, Nehls ef a/. evidence for an early role of leading edges of capillary sprouts, providing circumstantial pericytes bridging the gaps between pericytes in angiogenesis. They frequently observed guide for proliferating endothelial cells opposing endothelial sprouts, possibly providing a et al' (1968) to fuse and form continuous capillary loops. conversely, speiser retinal capillaries undergoing the documented the prior loss of pericytes associated with retinopathy. Furthermore, Sato neovascularisation that characterises proliferative diabetic have an inhibitory effect on endothelial cell et at. (1990) demonstrated that pericytes an activated form of TGFp' proliferation under in-vitro conditions, a process mediated by of pericytes may be permissive Kuwabara and cogan (1963) suggested that the absence play an important' but as yet not for angiogenesis. Therefore, pericytes would appear to completely defined role in angiogenesis'
Schorefa/.(1995)reportedphenotypicchangesofpericytesincultureconsistentwith the mineralisation of extracellular differentiation along the osteogenic pathway, including 22 matrix in areas of high cell density. lt was suggested that as angiogenesis precedes mineralisation, and on the basis of these observations of pericyte behaviour in vitro, lhal these cells contribute to the mineralisation process.
3.7 Cytokines
Greater knowledge of molecular behaviours within the MVB and other elements of the paradental tissues following mechanical stimulation may be seen as a progression towards a more complete understanding of orthodontic tooth movement, augmenting the histologic information provided by past investigations. Cytokines, growth factors, colony stimulating factors and cell adhesion molecules, which together possess a myriad of actions, have been implicated in the web of intimate molecular events concomitant with cell-mediated remodelling of the mechanically stressed periodontium. Leukocytes, osteoblasts, fibroblasts, epithelial cells, endothelial cells and platelets are capable of synthesizing many of these local factors (Davidovitch, 1995). This review will in part focus on one group of such local factors, the cytokines, particularly the cytokine ET-1.
Neurotransmitters, endocrine hormones, autacoids, and cytokines make up the four major classes of soluble intercellular signalling molecules (Nathan et al., 1991). Cytokines are secreted regulatory proteins that control cellular growth, differentiation and function (Nicola, 1994). The molecular detail of over 100 cytokines has been described. These include growth factors, colony stimulating factors, interleukins, lymphokines, monokines, and interferons. Cytokines relay information from producer cells, which act as biological sensors, to adjacent responsive cells, to elicit an appropriate biological response. Responder cells destroy cytokines by receptor mediated endocytosis. Unlike hormones, cytokines are rarely found in the circulation, but rather are produced from cells that are widespread throughout the body and have a localised action. Sandy (1992) indicated that the production of cytokines by mechanically deformed tissues of the periodontium, and their role in cellular changes associated with tooth movement, is a quintessential area in contemporary orthodontic research. lnterleukin-1 was the first cytokine implicated as a local regulating factor during orthodontic tooth movement. lt has two molecular forms, cr and B, which have similar biological actions. Monocytes are major producers of lL-1, although it is also released by endothelial and othercelltypes. According to Dinarello (1988), its ascribed biological role is thought to involve the mediation of a variety of events associated with host defence 23
(1988) reported mechanisms, inflammation, and autoimmunity' Davidovitch et al' application of a elevated levels of lL-1cr and lL-1B in cat periodontal tissues following the vitro lhat human tipping mechanical force to the teeth. Saito ef a/. (1991) demonstrated rn the pDL fibroblasts responded to the administration of lL-1p by significantly increasing production of prostaglandin E (PGE), a potent stimulator of bone resorption. Fibroblasts with may, therefore, influence bone remodelling by synthesizing PGE, which interacts was neighbouring bone cells. lncreased etficiency of orthodontic tooth movement (Yamasaki et al', 1984) demonstrated in monkeys (yamasaki et ar., 1982) and humans of oral when pGEl was administered locally by injection. Furthermore, administration in a rate of indomethacin, a potent inhibitor of PGE synthesis, Íor 21days in cats, resulted bone resorption tooth movement which was half that of controls. This implicated PGE in during orthodontic tooth movement (chumbley and Tuncay, 1986). 3.8 Endothelin-1
sequenced by Endothelin, a 21 amino acid peptide, is a cytokine that was isolated and endothelin-2 Yanagisawa et al. in 1988. Three isoforms are known (Figure a); ET-1, bridges (ET-2), and endothelin-3 (ET-3). Each isoform contains two intra-chain disulphide amino acids, linking paired cysteine amino acid residues. ET-2 differs from ET-1 by two and ET-3 differs by six amino acids.
Ser Leu Ser cys Ser cys Endothelin-l Met N (Human/Pocine/ Dog/Rat) Asp Lys Leu Asp lle lle Tyr Glu cys Val Tyr Phe cys His c
Ser Endothelin-2 Ser N cys Ser cys (Human)
Asp
Lys Leu Asp lle lle Tyr Glu cys Val Tyr Phe cys His c
Endothelin-3 N cys cys (Human/Rat)
Asp Lys Leu Asp lle lle Tyr Glu cys Val Tyr cys His c
Nicola N' A' (1994) Figure 4. Structure and amino acid sequence of endothelins' From 24
3.8.1 Biosynthesis Endothelins are produced by endothelial cells, neuronal cells, and numerous other cells (Nicola, 1994). The gene encoding ET-1 can be found in a wide variety of tissues, including heart, lung, brain, kidney, pancrees and spleen (Webb, 1997). ET-1 is the predominant isotype produced by endothelium. According to Haynes and Webb (1993), vascular smooth muscle cells also produce ET-1 in vitro, but at a rate 100 times less than that of endothelium. Each isoform of endothelin has a separate gene encoding a specific precursor (lnoue et al., 1989). ET-1 is produced from Pre-proendothelin-1 (212 amino acids), which is cleaved by a dibasic-pair specific endoprotease to form Big endothelin-1, or Proendothelin-1 (38 amino acids), the precursor to ET-1 (Figure 5). This cleavage occurs at the Lys-52-Arg-53 and the Lys-91-Arg-92 sites (Yanagisawa et al., 1988). Big ET-1 undergoes enzymic cleavage at the Trp-21-Yal-22 site by the action of endothelin converting enzymes (ECE-1 and ECE-2), which are metalloproteases. ECE-1 is considered the most physiologically important ECE, and exists in two isoforms; ECE-1a and ECE-1b. lntracellular ECE-1a is expressed in the Golgi apparatus of ET-1 producing cells. Responder cells to ET-1, such as vascular smooth muscle cells, can convert extracellular Big ET-1 to ET-1 by the expression of extracellular ECE-1b (Warner et al.,
1 e96).
Production of ET-1 by endothelial cells (Figure 5) may be stimulated by hormones (adrenaline, thrombin, angiotensin ll, arginine-vasopressin), growth factors (TGFP), lL-1, ET-3, physical stimuli (hypoxia and altered vascular shear stress), free radicals and endotoxin. lnhibition of ET-1 production is caused by endothelium-derived nitric oxide promoting the production of cyclic GMP or cyclic AMP (Haynes ef a/., 1993).
Little is known as to whether ET-1 is stored as an intracellular pool for rapid release. However, regulation of endothelin synthesis is thought to occur primarily at the level of gene transcription on a de noyo basis in response to endothelial cell stimulation (Webb, 1997). 25
Xenobiotics Physico-chemical 423187 stimuli Peptides Cyclosporin Hypoxia Blood Cytokines Shear stress (low) components tL-1 Osmolarity Thrombin TGFp Glucose Endotoxin Oxidized LDL Endothelin + + + +
I lnhibition Hormones + Promotor Prostacyclin Adrenaline ET-1 gene Nitric oxide Angiotensin ll ANP,BNP,CNP Vasopressin Heparin lnsulin Shear stress Cortisol Pre-pro ET-1 mRNA (high) Nucleus
Pre-pro ET-1 Endoprotease t Big ET-1 Endothelin- Endothelial converting cell enzyme * ET-1
ET.I
Figure 5. Factors that alter ET-1 synthesis and the pathway for ET-1 generation. lL-1: interleukin-1. TGFp: transforming growth factor B. LDL: low-density lipoprotein. ANP, BNP, CNP: atrial, brain and ctype natriuretic peptides. From Webb (1997, from an original design from Professor P. Vanhoutte).
3.8.2 Actions Haynes (1995) listed the following actions of ET-1; (1) it is a potent, long acting -ii vasoconstrictor, (2) it has inotropic and mitogenic properties, (3) it influences salt and water homeostasis, (4) it stimulates the generation of renin, angiotensin ll, aldosterone and adrenaline. Chakravarthy et al. (1992) demonstrated contraction of pericytes originating from bovine retinal capillaries in vitro in response to ET-1. ln addition to directly affecting vascular tone via stimulation of vascular smooth muscle, Webb (1997) suggested that ET-1 might indirectly enhance vascular tone by augmenting vasoconstriction to other agents, such as angiotensin ll, noradrenaline and serotonin; and 26
(1992) stated that by enhancing sympathetic nervous system activity. chakravarthy et al. the ET-1 is the most potent constrictor of large and small vessels yet identified, although 1000 times vessel sizes were not defined. Neuropeptide Y and norepinephrine are 100 to play less potent than ET-1 in producing vasoconstriction (Ohlen et al., 1989). lt may a role in the production of pain or hyperalgesia either directly, or through synergystic mechanisms with prostaglandins (Ferreira et al., 1989)'
be ET-1 undergoes bi-directional release from endothelial cells, but its effect appears to greatest at the abluminal side of the vascular endothelium i.e., at the endothelial cell/smooth muscle cell/pericyte interface (Pohl and Busse, 1989). Therefore, according to webb, (1997) it appears to be primarily a locally acting peptide. Furthermore, yanagisawa et al. (1ggg) and Haynes et al. (1991) reported that concentrations of ET-1 that found in that cause vascular contraction in vitro or in vivo are approximately tenfold short at less the circulation. Dupuis ef a/. (1g96) indicated that its half-life in the blood is than 5 minutes.
(VSMC), and binds VEGF is produced predominantly in the vascular smooth muscle cells ef a/' to specific receptors on endothelial cells, leading to their proliferation. Pedram (1gg7) found that ET-1 and ET-3 stimulated the synthesis of VEGF by cultured human VSMC 3 to 4 fold. Moreover, endothelial cell proliferation was stimulated by VEGF produced by ET-incubated vsMC. Tissue hypoxia is recognised as a stimulus for VEGF production by VSMC and for ET production by endothelial cells. Therefore, Pedram ef a/' (19g7) suggested that hypoxia may stimulate VEGF production both directly via VSMC and indirectly via ET secretion.
3.8.3 Receptors webb (1997) reported that the endothelins act on two receptor subtypes, ET¡ and ETs, which are characterised on the basis of their pharmacology. ETn binds ET-1 Maguire preferentially, and ETs has equal affinity for all three endothelins. Davenport and (1gg4), suggested that vasoconstriction induced by ET-1 involves activation of the ETn and small receptor, which predominates on vascular smooth muscle cells in both large and kidney, blood vessels. Messenger RNA for ET4 can be detected in the aorta, heart, in endothelial but not in the liver or endothelial cells. The ETe receptor may be detected and aorta' and vascular smooth muscle cells, predominately in the brain, lung, kidney, actions of Activation of ETe receptors has a moderating effect on the vasoconstrictive 27
ET-1 through the production of the vasodilator substances prostacyclin and nitric oxide (Webb, 1997).
These receptors are linked to phospholipase C, which leads to the formation of inositol triphosphate (lP3). Webb (1997) reported that stores of intracellular Ca'* are released by the stimulation of lPg and this is important for the initiation of contraction, as the intracellular contractile apparatus is activated by a rise in Ca'* concentrations. Endothelins also decrease membrane potentials of VSMC, causing influx of extracellular Ca'* and a more sustained increase in intracellulat Ca'* concentrations.
BQ-123 and BQ-788, respectively, are selective antagonists at the ETn and ETs receptors, and TAK-044isa combined ET¡and ETs antagonist. These peptides have been used to perform human pharmacology studies investigating the physiological role of ET-1.
3.8.4 The Physiological Role of Endothelin-1 lnfusion of Big ET-1 in the brachial artery in man has been shown to cause forearm vasoconstriction. However, infusion of the ECE antagonist phosphoramidon, prevents vasoconstriction in the presence of Big ET-1 (Haynes et al., 1993). Moreover, infusion of phosphoramidon results in progressive forearm vasodilation, suggesting that production of ET-1 by ECE situated in endothelial cells is important to the maintenance of basal vascular tone.
The specific ETn receptor antagonist, BQ-123, prevents vasoconstriction to administered ET-1, and when administered alone, causes vasodilation, suggesting that ET-1 maintains vascular tone through the activation of ETa receptors (Haynes et al., 1993).
According to Webb (1997), the L-arginine/nitric oxide system, sympathetic nervous system, and the 'endothelin system' are the only mediator systems known to maintain basal vascular tone. Regulation of the endothelin system is most likely to occur at a local level, due to the nature of its biosynthesis, actions and dissociation of its receptor complex. 28
3.9 Endothelin-Related lnvestigations
3.9.1 Endothelin-Related lnvestigations in Non-Human Primates Studies using non-human primates have been performed to determine the effects of endothelin on regional blood flows (Clozel and Clozel, 1989), to provide a primate model of chronic optic nerve ischemia (Orgul et al., 1996), to investigate the possible role of endothelin in such disease states as atherosclerosis (Lopez et al., 1990), cyclosporin A- induced hypertension (Bartholomeuz et al., 1996) and subarachnoid haemorrhage (Hino et al., 1996), and to investigate the distribution of endothelin in various tissues (Millar ef al., 1995; Matsumoto et al., 1997). A review of the literature demonstrates that no published investigations have been performed to determine the occurrence of ET-1 in the dental tissues of non-human primates.
3.9.2 lmmunohistochemical Labelling for Endothelin-1 in Dental Tissues Casasco et al. (1991) identified "endothelin-like immunoreactivity" in human tooth germ and mature dental pulp. The ET-1 antiserum used for the investigation was found to cross-react with ET-2 and ET-3 and, therefore, immunoreactive material was referred to as being "endothelin-like immunoreactive". Such immunoreactivity was found only within the cytoplasm of endothelial cells of arteries, veins and "small thin vessels" (Casasco ef al., 1991). Neither pulp cells nor neural cells displayed immunoreactivity, and it was suggested that the vascular endothelium might be the only source of endothelin in human dental tissues. lt was hypothesised that endothelin may play roles in the regulation of local pulpal blood pressure and/or blood flow and, when released by vascular endothelium, in the regulation of tooth development, due to its mitogenic properties.
Casasco et al. (1992) demonstrated endothelin-like reactivity in inflamed human pulp tissue, but only within the vascular endothelium. There appeared to be a lower intensity of immunolabelling in sections of inflamed pulp compared with normal pulp, suggesting possible changes in endothelin biosynthesis and/or degradation with inflammation. However, this result may have been due to a loss of antigenicity following delays caused by the difficulties associated with acquiring inflamed pulp tissue compared with healthy pulp tissue. 29
infusion of l nmol over 5 The use of Laser Doppler flowmetry revealed that intra-arterial artery in dogs produced a minutes of ET-1 into the mental branch of the inferior alveolar 1992)' and profound decrease in canine tooth pulpal blood flow (Gilbert et al'' pulpal microvasculature' This demonstrated the existence of endothelin receptors in the but not reversed' effect was attenuated by the Ca2* channel blocker nifedipine,
for right microscopic and shiohama et at. (1gg5) used immunohistochemicar techniques loaded and unloaded electron microscopic examination of ET-1 distribution in the achieved following the periodontium of wistar rats. Experimental tooth movement was maxillary molars' a method insertion of an elastic band between the first and second teeth served as controls. introduced by wardo and Rothbratt (1g54). The contra-raterar 6 hours to 14 days' However' The duration of orthodontic force application was between component of the the specific duration of force application for the immunohistochemical experiment was not rePorted.
PDL for ET-1 was "detected compared with the control side, immunolabelling in the However, the method of intensely" (shiohama et al., 1995) on the experimental side' Furthermore' in the quantifying the intensity of immunofluorescence was not described' "more intensely" PDL of experimental teeth, immunolabelling was detected (Shiohamaetal.,1995)onthetensionsidecomparedwiththepressureside.lncontrast immunoreactivity was limited to to Casasco et at. (1g91), who reported that endothelin-like dental pulp, shiohama et al' vascular endothelial cells of human tooth germ and mature fibroblasts and (1995) found ET-1 immunoreactivity was associated with osteoclasts, endothelial cells of the PDL.
cultured endothelial networks Lewis (1996) reported that enhanced expression of ET-1 by and low level tensional loading' occurred when they were exposed to compressive loading HesuggestedthatsuchachangeinET-lsynthesiswasconsistentwithabloodvessel vasoconstriction under the action of inflammatory response in which there is initial arterial An upregulation of ET-1 at 12 smooth muscle cells and vasodilation of the capillary bed. be due to stimulation of ET-1 hours after the application of tensile forces was thought to ET-1 synthesis continued to synthesis by lL-1B. Despite a fall in the initial lL-18levels, of lL-1B with the lL-1cr rise after 18to 24 hours and was attributed to the replacement He found that there was form, which has a stronger stimulatory effect on ET-1 synthesis. in the lower ranges greater enhancement or suppression of synthesis of these cytokines 30 of load application in either direction, and suggested that this finding was consistent with the widely held view that lighter orthodontic forces move teeth faster.
Further investigation into the occurrence of ET-1 in the resting and mechanically stimulated periodontium is indicated. Quantification of immunolabelling for ET-1 in the pDL microvasculature might provide further insight into the physiology of the periodontium and the possible roles of its MVB and of ET-1 in orthodontic tooth movement.
3.1 0 lmmunofluorescence
The immunofluorescence technique of antigen localisation utilises the specific reactivity of antibody with antigen, where one of the reactants, usually antibody, is labelled with a fluorochrome, or fluorescent dye, to enable observation of the antibody-antigen reaction (Hayashi, 1983). lt was first described by Coons in 1941 (Rost, 1992), and has applications in microbiology, immunology, pathology, histology and almost every biomedical field in which it is useful to identify and locate antigens in tissues and cells (Hayashi, 1g83). Thefluorochrome may be applied tothe antigen using one of a number of methods (Figure 6).
The Direct Method: The fluorochrome is tagged to the antibody, and when tissue containing the corresponding antigen is stained with the fluorochrome-labelled antibody, only those cells with the antigen will fluoresce under a fluorescence microscope. Conjugation of the fluorochrome to the antibody must not hinder the antibody reactivity. The labelled antibody may be applied to antigens located within the cytoplasm, or on the cell membrane. lt may also be applied not only to single antigens, but also to any antigenic substance, such as an antigen-antibody complex preformed in vivo.
The lndirect Method: An unlabelled (primary) antibody is reacted with a known or unknown antigen. The antigen-antibody complex is then labelled with a secondary antibody that has been tagged by a fluorochrome. The indirect method of immunofluorescence is 5-10 times more sensitive than the direct method (Hayashi, 1983).
Other immunohistochemical methods of antigen location are the peroxidase- antiperoxidase (PAP) method, the alkaline phosphatase-antiperoxidase (APAAP) method and the avidin- or streptavidin-biotin methods. These immunohistochemical techniques have afforded greater sensitivity than the direct and indirect methods, as more fluorochrometagged molecules are bound to each antigenic site. 31
Antigen Biotinylated secondary ^.^ antibody Primary Antibody /( )K >K- Fluorochrome A)K Streptavidin- - )K >K Direct lndirect Technique Technique )K )K. à< )K Streptavidin- biotin Technique Figure 6. The direct, indirect and streptavidin-biotin immunofluorescent techniques. 32 3.1 0.1 Fluorescence MicroscoPY Fluorescence microscopy functions on the basis that when materials absorb a sufficient amount of radiation energy, the component molecules become excited, and electrons in the outer orbits become altered. This energy is lost by collision of neighbouring molecules within the material, or by emission of photons, or light, as the molecules return to their original stable state. The emitted light has a longer wavelength than the original absorption light, and this emission is termed fluorescence (Hayashi, 1983). ln the fluorescence microscope, illumination of the specimen occurs from a light source rich in short wavelengths. An excitation filter is placed between the lamp and the specimen and functions to provide the excitation beam corresponding to the maximum absorption of the fluorophore. The fluorophore is that part of the molecule responsible for fluorescence. Specificity of the technique may be increased by confining the wavelengths of light used for excitation to a narrow band, as light at other wavelengths not corresponding to an absorption peak may excite other fluorophores which may happen to be present (Rost, 1ee2). A barrier, or suppression filter, is placed between the specimen and the eye, and excludes direct rays of the excitation light and rays of lower wavelength than those emitted by the fluorochrome from the observation system of the microscope. For example, the fluorochrome fluorescein isothiocyanate (FITC) may be used with an excitation filter which can select a wavelength of 495nm, and a barrier filter which passes only 525nm fluorescence. 3.10.2 Gonfocal Laser Scanning Microscopy Confocal laser scanning microscopy has been described as 'CAT scanning for cells' (Johnson, 1994). lt is a fluorescence microscope technique that uses computer processing to reconstruct a series of images of an immunolabelled tissue structure to form a three-dimensional image. Conventional fluorescence microscopy is limited by the fact that light above and below the focal plane enters the eyepiece, which occassionally results in blurring. This can make localisation and high resolution photography of immunolabelled structures difficult. The CLSM only gathers light emitted from a tissue structure that is within the focat plane of the lens. A laser provides a source of light that is focussed onto the tissue structure by an objective lens. The fluorescence produced by the immunolabelled sample passes through the same lens to a light sensitive detector, which measures the light intensity. An aperture in front of the detector is positioned so 33 that only light from the focal point within the sample is allowed to pass through to the detector. The laser light is scanned across the specimen and the information is digitised for generation of an image on a computer monitor. This enables collection of 'optical (Johnson, sections, for accurate localisation of structures and construction of 3-D images 1 ee4). 3.11The Marmoset as an Exper¡mental Animal Marmosets are South American primates of the suborder Anthropoidea (Johnston et al., Kinzey, 1g7O) and, together with tamarins, form the subfamily Cattitrichinae (Sussman and 1gg4). subdivision of the cattitrichidae family is based on the morphology of the lower incisor and canine teeth (Johnston ef at., 1970). The lower incisors of the marmoset are narrow, elongate, and reach the occlusal level of the canines, whereas those of the genera tamarins are spatulate and shorter than the canine teeth. There are two of The species marmosets , Callithrix, of which there are seven species, and Cebuella. Cattithrix jacchus (cotton-ear, or white tufted-ear marmosets) are located in the wild in eastern Brazil (Sussman et al., 1984). Marmosets have been favoured primates for use in biomedical research due to their small size (200-500gm); relatively low purchase cost; low cost of housing and feeding compared with other primates; good breeding capacity in captivity; and short gestation period (approximately 140 days). Furthermore, they usually give birth to biovular twins (Johnston et al., 1g7O). Weaning is complete by 3 months of age, physical maturity is attained at 12 to 18 months, and their estimated life span is 15 years. 3.11.1 The Marmoset Dentition Marmosets are omnivorous, feeding on three primary types of food items; insects, fruits, flowers and nectar, and plant exudates. They use their relatively unusual lower dentition the to gouge holes and induce the flow of gum in trees and vines. The enamel is thick on of labial side of the lower incisors, and lacking on the lingual side, to enable maintenance a chisel-like edge (Rosenberger, 1978). The dental formula of the primary dentition is:- lncisors 212 Canines: 1/1; Molars: 3/3, and the dental formula of the permanent dentition is:- lncisors: 212; Canines: 111; permanent Premolars: 3/3; Molars: 212 (Goss, 1984. Figure 7). The presence of two molars in each quadrant contrasts the three molars found in other New World monkeys (James, 1980). 34 a) v - .., ;!såi:: t,{ i-l \'^ t'.# ^rt { c) b) - - I '?', íl <Ë 4&(j FigureT.Marmosetpermanentdentition.a).Buccalviewofdentition,maxillaandmandible. scale bar = 5mm' c)' occlusal view of scale bar = 8mm. b). occlusal view of dentate maxilla M400 stereomicroscope (switzerland). dentate mandibre. scare bar = 5mm. wird Heerbrugg 35 Reitan and Kvam (1971) observed variable tissue responses between the human, rat, dog and monkey during orthodontic tooth movement due to distinct physiological and anatomical differences. lt was recommended that caution be exercised in directly extrapolating findings from animal experiments to man. lt has been suggested, however, that the periodontal ligament of the marmoset (Figure 8), is histologically analogous to man, providing one of the more reliable animal models for orthodontic biological research [Dreizen et al. (1967), Levy and Bernick (1968a, b), Bernick and Levy (1968a, b) Levy ef al. (1970), Skougaard et al. (1970), Levy (1971), Levy et al. (1972), Cohn (1972), Grant ef al. (1972), Page ef al. (1974), Levy (1976), Bernick ef al. (1977) and Douvartzidis (1983)1, a) b) PM3 MI Nl2 Figure L Photographs of sagittal section of permanent mandibular dentition of the marmoset. a). Lower left third premolar, first and second molars. Scale bar = 1mm. b). Lower left first molar. Arrowhead indicates PDL. Scale bar = 1mm. Wild Heerbrugg M400 stereomicroscope (Switzerland). 36 Chapter 4 MATERIALS AND METHODS 4.1 Summary Buccal segment teeth of four anaesthetised marmosets were subjected to an occlusal load of 120-2OOgm for 1.5 hours, with contra-lateral teeth serving as controls. Dental and periodontal tissues, and samples of kidney, skeletal muscle, and gut were extirpated immediately after the animals were euthanased. 2. A series of pilot studies were performed for the purposes of establishing immunohistochemical labelling techniques using various primary and secondary antibodies. 3. The distribution of vascular endothelium, pericytes and smooth muscle cells, and the expression of the cytokine endothelin-1 within the paradental tissues was investigated using immunohistochemical techniques. 4.2 Ethics Approval permission for the experiment to be performed was granted by the Ethics Committee of The University of Adelaide; Approval No: S/011/97 4.3 Research Colony The experimental animals for this investigation were obtained from.the research colony of Cattithrix jacchus housed within the Division of Human Nutrition, Commonwealth Scientific and lndustrial Research Organisation (CSIRO), Adelaide, Australia. This colony was established from a breeding nucleus of 15 pairs imported from the United Kingdom in 1978 (Mclntosh and Looker, 1982). The marmosets were housed in "social" cages in family groups. They were fed a high protein, low fat diet, with high levels of Vitamin C and D, in the form of a commercially prepared extruded New World primate ration (Ridley Agri Products, Queensland), high protein cat biscuits (WhiskcettesrM and Go-CatrM), fresh 37 fruit, and vegetables. Pentavite supplements (Fauldings, Underdale, South Australia) were supplied weekly, and crickets, meal worms and baby rats were supplied when available. Water was provided ad libitum' 4.4 Summary of Pilot Studies A description of the materials and methods of the pilot studies follows, and a summary is provided in Table 2. All antibodies and donkey normal serum (DNS) used in this parts of investigation, with the exception of JC-704, were aliquoted and diluted with equal prior distilled water and glycerol according to manufacturer's recommendations to being frozen and stored. This was then considered the "neat" state. All antibody and serum dilutions described in this report were derived from the neat state. "Experimental A distinction must be made in the use of the term "control" in this study. control,' refers to the unloaded contralateral teeth in the animal. "lmmunohistochemical of control" refers to positive or negative control tissue sections used to assist evaluation binding results and exclude immunofluorescence that may have resulted from non-specific or cross-reactivity. ln all of the pilot studies, two negative control sections were incorporated. These sections were processed in an identical fashion to the experimental for sections, with the exception that 10% DNS diluted in antibody diluent was substituted the primary and the secondary antibody in one of the control sections, and only the secondary antibody was applied to the other control section. Table 2. Summary of Pilot studies. Pilot Antibody Routine Purpose Seconda RECA-1 lgG Cy3 1. Test for presence of RECA-1 antigen in marmoset. 1 (MoxRt) (Dox Mo) 2. Test laboratorY technique' DTAF lgG 1. Test for presence of ET-1 in marmoset tissues. AntiET-1 (Dox Rb) or; 2 Establish correct antibody working dilutions. (Rbx Hu) Rb 2. 1 Test for suitabilitY of QB-END 10 as Pan- lgG Cy3 QB-END 1O endothelial label in marmosets. 3 (Dox (Mox Hu) Mo) 2. Establish correct worki dilutions 1. Test for suitability of JC-704 as pan-endothelial JC-704 lgG Cy3 labelin marmosets. 4 (Mox Hu) (DoxMo) 2. Establish correct dilutions. 1 Test suitability of u-SMA as label of lgG Cy3 cr-SMA periendotheli cells in marmosets. 5 (DoxMo) (Mox Hu) 2 Establish correct worki d utions Mo: mouse. Rt: rat. Do: donkey Rb: rabbit. Hu: human. 38 4.5 Pilot Study No 1 (RECA-1) lmmunohistochemical staining using RECA-1 on marmoset tissues constituted an appropriate pilot study for two reasons: 1. A review of the literature demonstrated that immunohistochemical staining using the RECA-1 antibody has not been attempted on primate tissues. 2. lmmunolabelling with RECA-1 in rat tissues could be performed "in parallel" as a positive controlto the marmoset tissues. 4.5.1 The Experimental Animal Tissues were obtained for pilot study No 1 from a marmoset that had died at the CSIRO facility. An animal which had been euthanased and frozen soon after was favoured for use in the pilot study, as it was reasoned that there was less chance that autolysis had occurred. 4.5.2 Tissue Preparation and I m m u noh istochem istry The selected marmoset for this pilot study was a male of 1 year and 7 months, which had been injured fighting, and 2 days later euthanased and frozen. The cadaver was defrosted, and was found to be in good condition. lt was noted that rigor mortis had not set in. Samples of skeletal thigh muscle, kidney, and gut were obtained and placed in Zamboni's fixative at 4"C for 30 hours. The tissues were then transferred into 0.01M phosphate buffered saline (PBS), pH 7.4, at 4'C. Four days later, the picric acid was removed from the tissues by the application of four 1S-minute rinses of 80% ethanol. The tissues were cleared with a 4% solution of dimethyl sulphoxide (DMSO; Merck Pty. Ltd., Kilysth, Victoria, Australia) with three, 10 minute rinses. The DMSO was diluted to 4% in 0.01 M PBS, pH 7 .4 in cases of RECA-1 immunolabelling, as this was originally thought to reduce "background staining", when compared with those tissues cleared with higher DMSO concentrations. (Thiswas laterfound notto bethe case). Finally,0.01M PBS, pH 7 .4, was used to apply three, 10 minute rinses to the tissues before they were placed in a solution of PBS/30% sucrose and 0.01% sodium azide at 4"C overnight. As the tissues initially floated in the storage medium, they were placed on a bench shaker until they sank, in order to prevent surface dehydration. 39 thickness of samples of the tissues were prepared and undenruent cryostat sectioning at a (Sigma BioSciencesrM, 1g¡rm. The sections were placed on poly-L-ornithine coated slides phosphorus St Louis, Missouri). The slides were dried in a vacuum dessicator containing pentoxide for 30 minutes. The sections were blocked for 30 minutes with 10% DNS (Jackson lmmunoResearch Laboratories, lnc., west Grove, Pennsylvania, usA), diluted in 0.01M antibody diluent, in order to eliminate background staining caused by cross- reaction of the secondary antibody with endogenous immunoglobulins in the specimens. a series of For lHC, the tissue sections were incubated overnight in a humid box using purchased via Jomar dilutions of the primary antibody, mouse anti-rat RECA-1 (Serotec, the slides Diagnostics, Magill, south Australia) after the excess DNS was drained off placed PBS (Appendix 8.27). The slides were dipped in 0.01M PBS, pH7.4, then in a using the bath for three, 10 minute intervals. The sections were incubated for t hour secondary antibody, donkey anti-mouse lgG conjugated with CyJru (Jackson dilution of lmmunoResearch Laboratories, lnc., West Grove, Pennsylvania, USA), at a 1:100. The sections were again rinsed with 0.01M PBS, pH7'4, in the same manner' paper mounted' The slides were dried around the sections with high-grade blotting and New The mountant was a prepared mixture of Mowiol 4-88 (Calbiochem, Alexandria, South Wales, Australia), glycerol and 2'5o/o DABCO (Sigma)' from a ral, Tissue sections of skeletal muscle, kidney, and gut, which were extirpated were immunolabelled with RECA-1 using the technique described above' BX50' Tokyo' The sections were viewed under a fluorescence microscope (Olympus computer and an NIH Japan), and images recorded using an Apple Maclntosh Quadra lmage programme; version 1.61 (American National lnstitute of Health, Bethesda, (N'4.) 0.16; 10X, Maryland, usA). objective lenses used were4X, numerical aperture N.A. 0.40; 20X, N.A. 0.70 and 40X, N.A. 0.85' 4.6 Pilot StudY No 2 (ET-1) the experimental This pilot study was performed using soft tissue sections derived from produce reliable results animals in order to establish a laboratory technique that would antibody and the and to determine the correct working dilutions of both the anti-ET-1 appropriate secondary antibodY. 40 4.6.1 The Experimental Animal Details of the experimental procedure prior to sacrifice of the animals are in Section 4.10.1. Samples of kidney, liver, gut and stomach were extirpated 5 minutes post-mortem and placed in Zamboni's fixative at 4'C for 24 hours. The picric acid was removed from the tissues by the application of four, 15 minute rinses of 80% ethanol. The tissues were cleared with 100% DMSO with three, 10 minute rinses. Finally, 0.01M PBS, pH 7.4,was used to apply three, 10 minute rinses to the tissues before they were placed in a solution of PBS/30% sucrose and 0.01% sodium azide at 4"C. The tissue samples were placed on a bench shaker until they submerged in the storage medium. 4.6.2 Tissue Preparation and lmmunohistochemistry Frozen sections of kidney, skeletal muscle and gut were cut at 14pm using a cryostat and placed on poly-L-ornithine coated slides (Sigma). The slides were dried in a vacuum dessicator, containing phosphorus pentoxide, for 30 minutes. The sections were blocked with 10% DNS diluted in antibody diluent for 30 minutes in order to reduce the level of background staining. Excess DNS was drained off the slides before applying the primary antibody. lmmunohistochemistry involved incubation with rabbit anti-ET-1 antibody (Peninsula Laboratories, lnc., Belmont, California, USA) using a chequerboard test of varying concentrations (Appendix 8.28a). The sections were incubated in a humid box for 24 hours, then dipped in 0.01M PBS, pH7.4, and placed in a PBS bath for three, 10 minute intervals. The secondary antibody, donkey anti-rabbit lgG, conjugated with dichlorotriazinyl amino fluorescein (DTAF; Jackson lmmunoResearch Laboratories, Inc., West Grove, Pennsylvania, USA.), was applied at varying dilutions according to the chequerboard for t hour. Dilutions of antibody were prepared using 10% DNS in antibody diluent. The sections were again rinsed with 0.01M PBS, pH7.4, in the same manner. The slides were dried around the sections with high-grade blotting paper and mounted in Mowiol4-88 (Calbiochem); glycerol and DABCO (Sigma). The sections were viewed under a fluorescence microscope (Olympus), and images were recorded using an Apple Maclntosh computer and NIH lmage programme' 41 4.7 Pitot Study No 3 (QB-END r0) Pilot studies 3 and 4 were performed in order to identify a suitable pan-endothelial immunolabel for use on marmoset tissues, For the remainder of the pilot studies, tissues were sourced in a manner identical to that of pilot study No 2. Tissue sections of kidney, skeletal muscle and gut were prepared for IHC using the protocol described above. lmmunohistochemistry involved incubation in a humid box overnight with the endothelial cell marker, QB-END 10 (Novocastra Laboratories Ltd, Newcastle upon Tyne, UK) at various dilutions (Appendix 8.29). The primary antibody was drained from the sections by the application of high grade filter paper to the edges of the slides. Rinsing was performed three times for 10 minutes by pipetting 10% DNS diluted in antibody diluent on to the sections. The sections were drained between each rinse using filter paper. They were incubated in the secondary antibody, donkey anti- mouse lgG, conjugated with Cy3rM (Jackson), at a dilution of 1:100 for t hour. The sections were drained in the same manner as described above and rinsed three times using 0.02M PBS, pH 7.4. Finally, the sections were mounted in 20o/o buffered glycerol, pH 8.6, in preference to the Mowiol used earlier (Page 80). The tissues were viewed under a fluorescence microscope (Olympus). 4.8 Pilot Study No 4 (JG-704) Frozen sections of marmoset gut were prepared and underwent the same IHC procedure as above. However, the tissues were incubated using the mouse monoclonal anti-human antibody to Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1 or CD31), JC-704 (Dako Australia Pty. Ltd., Botany, NSW, Australia) at various dilutions (Appendix 8.30). The secondary antibody was donkey anti-mouse lgG conjugated with Cy3ru (Jackson), at a dilution of 1:100. The sections were viewed and images recorded using the protocol of the previous pilot studies. 4.9 Pilot Study No 5 (o-SMA) Frozen sections of marmoset gut were prepared and IHC performed according to the method of pilot study No 3. The primary antibody was mouse monoclonal anti-human 42 or-SMA antibody (Maxim Biotech, lnc., San Francisco, California, USA) at various dilutions (Appendix 8.31). The secondary antibody, donkey anti-mouse lgG, conjugated with CyJru (Jackson), was applied at a dilution of 1:100. The sections were viewed under the fluorescence microscope, and images recorded using the computer. 4.10 Main Experiments 4.10.1 The Experimental Procedure Four adult male marmoset monkeys were used for the current study. Sections of buccal segment teeth and their surrounding structures, as well as samples of kidney, liver, skeletal muscle, gut and stomach were obtained for lHC. The animals were anaesthetised at the CSIRO housing facility using 0.6m1 of Saffan (pitman-Moore Australia Ltd. North Ryde, New South Wales, Australia) administered intramuscularly. Anaesthesia was maintained using supplementary increments (0.5m1) of intramuscular anaesthetic as required during the course of the experiment. The corneal and plantar reflexes were checked regularly to determine the depth of anaesthesia. The same procedure was performed on each animal. An intra-oral pad was placed randomly between the left or right upper and lower third premolars and first and second molars. A loading device was placed around the animal's head in order to ensure that an occlusally-directed load was apptied to the teeth contacting the intra-oral pad (Figure 9). The load was applied for 1.S hours, during which time anaesthesia was maintained (Figure 10). The unloaded contra-lateral teeth served as experimental controls. A final o.gml dose of Saffan was administered 5 minutes before the animal was sacrificed by guillotining. The head was immediately placed in Zamboni's fixative at 4'C with the intra- oral pad and loading device in situ. The buccal mucoperiosteum was elevated from the mandibular bone to enable rapid penetration of the fixative to the periodontal tissues. 4.10.2 Loading Device and lntra-Oral Pad A loading device (Figure g) was custom-designed by one of the author's supervisors, professor M.R. Sims, for the purposes of a series of research projects conducted in the laboratory. Several devices were constructed from 0.040" diameter stainless steel tubing 43 and wire to dimensions appropriate for the head of an adult marmoset. Proprietary 5/16" orthodontic elastics, (TP Orthodontics, La Pofte, lndiana, USA) each of which have a designated force value of 2oz, were used to activate the devices. Four elastics were placed on each device; two on each side. After each device was positioned on a marmoset, an indelible mark was placed on one of the plungers to indicate its degree of activation. After tissue fixation was complete, the devices were removed from the animals, and the load applied during the experiment was measured by "re-activation" of each device. A Gorrex gauge (Haag-Steit AG, Liebfeld-Bern, Switzerland) measured the load required to displace the plunger back to the original position as indicated by the indelible mark. Each device delivered between 120-2OOgm of force. The force transmitted to the teeth was not quantified. However, some dissipation of this force by the soft tissues of the head would have occurred. LOADING INTRA-ORAL DEVICE PAD Hook for e/asfl'cs Rubber pad Plunger Plastic handle O¡Thodontic e/asfics Metal tubing I I Figure 9. Diagrammatic representation of loading device and intra-oral pad used for application of occlusally-directed load to experimental teeth. 44 4.10.3 Tissue Preparation After overnight fixation, the maxilla and mandible were resected, and the remaining soft tissues removed from the bone. The whole mandible and maxilla were rinsed several times in 0.01M PBS, pH7.4, and placed in a storage solution of PBS/30% sucrose and O.O1o/o sodium azide at 4'C. The maxillae remained in storage solution for future research. The mandible was later hemi-sectioned, and the left and right bodies were individually mounted on Perspex blocks using LoctiterM adhesive in preparation for sectioning. Sagittal sections approximately 150¡rm thick encompassing the bone, periodontium, and teeth of each mandibular body were cut using a diamond wafering blade mounted on a low speed sectioning saw (lsometrM Buehler, Lake Bluff, lllinois, USA. Figure 11). The lubricant used during sectioning was 0.01M PBS, pH 7 .4, at 4'C. The sections were placed sequentially in graded ethanol solutions (70o/o,80%, 90% and 100%) for 5 minute intervals in order to remove the picric acid. They were placed in 100% DMSO for 5 minutes, and rinsed in 0.01 M PBS, pH 7 .4, for three, 10 minute intervals, and finally stored in PBS/30% sucrose and 0.01% sodium azide at 4"C. The sections were later divided into 'anterior' and 'posterior' segments between the 2nd and 3'd premolars using a scalpel blade. 4.10.4 I m m u noh istochem istry The jaw sections were placed in small plastic vials and blocked in 500p1 of 10% DNS diluted in antibody diluent for two hours. The vials were placed on a bench shaker during the blocking and incubation periods. The excess DNS was drained off the sections, and the relevant primary antibody was applied (Table 3). The antibody was diluted in antibody diluent containing 10% DNS. As the hard-tissue jaw sections were approximately 150pm thick, the preferred concentration of primary and secondary antibodies, which had been determined during the pilot studies, was doubled (Table 3). The penetration of the antibodies through tissue sections is concentration driven. Therefore, a higher concentration was required for the thicker hard tissue jaw sections as compared to the cryostalcut soft tissue sections. The sections were incubated in primary antibody solution at room temperature for 4 days. 45 The sections were washed in 10% DNS diluted in antibody diluent for three, 30 minute intervals. lncubation with the appropriate secondary antibody was performed for 2 days in a manner identical to that of the primary antibody with the exception that the vials were covered with aluminium foil to prevent fading of the fluorophore. Finally, the sections were washed in antibody diluent without DNS for three, 30 minute intervals. DNS was no longer required as the sections were not further exposed to any antibodies. The sections were mounted on large coverslips in 2oo/o bulÍered glycerol, pH 8.6. To enable the sections to be viewed from either side using the CLSM, a coverslip was placed on each side. A glass slide was too thick to enable penetration of the confocal laser beam. Therefore, viewing would have been limited to one side of the section. The sections and glycerol were sealed by the application of nail varnish along the edges of the coverslips. Table 3. Summary of main experiments. Antibody Routine/D ilution Experiment Purpose Prima 1. Functional implications (FM + CLSM). lgG Cy5 Anti-ET-1 2. Check for alteration in ET-1 expression (Dox Rb) 1 (Rbx Hu) concurrent with tooth loading (CLSM). 1:50 1:50 3. Future research aPPlications. 1. Description of microvascular morphology JC-704 lgG Cy3 (FM + CLSM). 2 (Mox Hu) (Dox Mo) 2. Check for alteration in PECAM-1 expression 1"40 1:50 concurrent with tooth loading (CLSM). 3. Future research aPPlications' c¿-SMA lgG Cy3 1. Description of microvascular morphology 3 (Mox Hu) (Dox Mo) (FM + CLSM). 1:100 1:50 2. Future research aPPlications. Mo: mouse. Do: donkey Rb: rabbit. Hu: human. FM: fluorescence microscope. CLSM: confocal laser scanning microscope. 46 Figure '10. Photograph of anaesthetised marmoset with loading device and intra-oral pad in-situ. P = intra-oral pad LD = loading device Figure 11 Photograph of marmoset mandibular body (M) undergoing sagittal sectioning using low speed saw (lsometrM Buehler, Lake Bluff, lllinois, USA). P = perspex block B = rotating diamond blade 47 4.10.S Examination of Mandibular Sections and Data Gollection All mandibular sections were initially viewed under a fluorescence microscope to allow orientation and to evaluate the immunolabelling procedure. Where more detailed examination of vascular morphology or quantification of antigen expression was required, the sections were examined thereafter using a Bio-Rad MRC-1000 Confocal Laser Scanning Microscope System (Hemel Hempstead, Herts., U.K.) and CoMOS image analysis software (Bio-Rad, version 7.0a). The maximum power output of the Krypton- Argon laser was 15mW. A detailed image of an immunolabelled vascular structure for morphological analysis was constructed by "stacking" serial images captured in 1pm ,,slices,, in the Z-axis of the tissue section. Generation time of each optical slice was according approximately one second. lmmunolabelled antigen expression was measured to average pixel brightness per unit area using a 0 to 256 greyscale. Statistical comparison of antigen expression in the microvasculature of the marmoset periodontium during loaded (experimental) and unloaded (control) conditions required the the concurrent incubation of the appropriate mandibular sections from each animal under same laboratory conditions. Variability in the intensity of immunolabelling for reasons quantification in other than antigen expression was further accounted for by performing terms of "signal-to-noise" ratios aS opposed to signal intensity per se' Vessels of the of Same morphology Were compared. For each Vessel, 20 areas of the vessel wall between 3 and 10¡rm, were analysed for average pixel brightness, together with an "noise" (Figure 12). equivalent area adjacent to the vessel for assessment of background "Noisg": average PDL pixel brightness/unit t-----r T----1 l-----r l-----r Tr i t¡ t¡ l¡ I area of background L----.: !-----.: L----.,: l-----.i Vesselwall "Signal": average Ì pixel brightness/unit area of vessel wall Vesse/ Iumen Blood vesse/ by measurement of Figure 12. Method of quantifying antìgen expression within blood vessel wall ,,signal-to-noise" ratios using coMos image analysis software (Bio-Rad)' 48 Chapter 5 FINDINGS 5,1 Summary of Pilot Study Findings 5.1.1 RECA-í The use of the RECA-1 antibody for endothelial cell immunolabelling in marmoset soft tissues produced a negative result. However, immunolabelling of endothelium with this antibody occurred in rat tissues. 5.1.2 Endothelin-1 Attempts at immunolabelling for ET-1 in marmoset kidney, skeletal muscle and gut vascular endothelial cells were at first inconclusive. Therefore, numerous pilot studies were required. A. Tested primary antibody rabbit anti-ET-1 (Peninsula) and secondary antibody IgG/DTAF. Appendix 8.28a. Results inconclusive. Repeated with same outcome. B. Tested primary antibody rabbit anti-ET-1 (Peninsula) and secondary antibody lgG/Cy3rM. Appendix 8.28b. Higher concentrations of primary antibody used. Sections received extra washes of graded ethanol solutions prior to incubation in an effort to remove any remaining traces of picric acid. 100o/o DMSO also applied to ensure adequate antibody penetration. Results inconclusive. C. Tested primary antibody rabbit anti-Big ET-1 (Peninsula) and secondary antibody lgG/Cy3rM. Appendix 8.28b. Positive immunolabelling of endothelial cells of some renal vascular structures. D. Concurrent trial of rabbit anti-Big ET-1 and anti-ET-1 (Peninsula) primary antibodies on separate sections followed by lgG/Cy3rM secondary antibody. Appendix 8.28c. This experiment was designed for the use of each primary antibody under the same laboratory conditions. Positive immunolabelling for Big ET-1 in renal vascular endothelium occurred. lmmunolabelling for ET-1 was observed in multinucleated tissue cells and occasional epithelial cells of the proximal convoluted tubules of renal nephrons, but not in vascular structures, 49 E. Tested primary antibody rabbit anti-Pre-pro ET-1 (Peninsula) and secondary antibody lgG/Cy3rM. Appendix 8.28d. Produced immunolabelling in some endothelial cells of renal peritubular caPillaries. F. Tested another proprietary brand of two anti-ET-1 primary antibodies (Yamasa), one specific for the N-terminal, the other for the Cterminal of ET-1 and used lgG/Cy3rM secondary antibody. Appendix 8.28e. Positive immunolabelling was observed within occasional epithelial cells of the proximal convoluted tubules of renal nephrons' However, no immunolabelling of endothelial cells was demonstrated. G. Streptavidin-biotin amplification technique tested. lmmunolabelling of endothelium of several larger renal blood vessels occurred, but with higher levels of background staining. S.1.3 Suitability of QB-END 10 as a Pan-endothelial lmmunolabel lmmunolabelling of endothelial cells did not occur using the QB-END 10 primary antibody and lgG/Cy3rM secondary antibody combination. 5.1,4 Suitability of JC-7OA as a Pan-endothelial lmmunolabel pan-endothelial immunolabelling of marmoset gastric vasculature using the JC-704 primary antibody and lgG/Cy3rM secondary antibody combination was observed. 5.1.S Suitability of cr-SMA as a Periendothelial Gell lmmunolabel lmmunolabelling of periendothelial cells of marmoset small intestine using the a-SMA primary antibody and lgG/Cy3rM secondary antibody was demonstrated. 5.2 Pilot Study No 1 (RECA'1) lmmunolabelling of endothelial cells with RECA-1 was not found after incubation in any of the various dilutions of primary antibody for sections of skeletal muscle, kidney or gut of the marmoset, Positive immunolabelling of endothelium occurred, however, in those tissues extirpated from the rat (Figure 13), which suggested that the laboratory methodology had been performed correctly. 50 5.3 Pilot Study No 2 (ET-1) 5,3.1 Experiment A Results of the attempted immunolabelling for ET-1 were at first inconclusive. The pilot study was repeated according to the same chequerboard (Appendix 8.28a) using sections of marmoset stomach only, again with unsatisfactory results. Attempts were made to determine if modification of the laboratory methodology was required to successfully label vascular ET-1 in marmoset tissues. 5.3.2 Experiment B It was thought possible that there may have been traces of picric acid remaining in the tissues preventing binding of the anti-ET-1 antibody to the ET-1 antigen. The manufacturers of the anti-ET-1 antibody, Peninsula Laboratories lnc., had cautioned that ET-1 is a delicate peptide, and that picric acid may constitute a barrier to successful immunolabelling. To remove any traces of picric acid that may have remained after washing the tissue in graded alcohols, 70% ethanol was pipetted directly on to cryostat- cut sections of marmoset kidney for 30 minutes. The 70o/o ethanol was replaced with 80o/o,9Oo/o and finally, 100o/o ethanol at 1S-minute intervals. To ensure adequate penetration of the antibodies through the tissues during incubation, lOOo/o DMSO was applied to the sections for 15 minutes. The secondary antibody, donkey anti-rabbit lgG, conjugated with Cy3rM, had recently become available, and was substituted for donkey anti-rabbit DTAF, as used previously, since it was found in other pilot studies to produce a strong fluorescent signal at a dilution of 1:100. The same laboratory protocol was otherwise followed, with the exception that higher concentrations of primary antibody were tested (Appendix 8.28b). The results were again inconclusive. 5.3.3 Experiment C The same laboratory protocol and chequerboard were followed using sections of marmoset kidney whilst substituting rabbit anti-Big ET-1 (Peninsula) as the primary antibody (Appendix 8.28b). The control slides revealed no immunofluorescence, apart from autofluorescence of a granular material in the cuboidal epithelial cells lining the convoluted tubules of the kidney, lmmunolabelling occurred in the endothelial cells of some, but not all, of the arterioles, veins and peritubular capillaries (Figure 14). The most satisfactory "signal-to-noise" ratio occurred at a primary antibody dilution of 1:50. Where 51 immunolabelling did occur within a vessel, all of the endothelial cells were not necessarily involved. Glomerular capillary loops also displayed immunolabelling for Big ET-1 (Figure 15). This labelling was not consistent throughout the sections and varied within each glomerulus. Occasional multinucleated tissue cells also displayed positive immunolabelling for Big ET-1. These cells were considered to be macrophages and were typically grouped within tissues adjacent to blood vessels. 5.3.4 Experiment D Following successful immunolabelling for Big ET-1 in the marmoset kidney, another pilot study was designed to attain labelling for vascular ET-1. A series of kidney sections were prepared and underwent the additional steps described in Experiment B. They were incubated in either rabbit anti-ET-1 or rabbit anti-Big ET-1 antibodies (Appendix 8.28c). ln processing the sections concurrently, the influence of variation in laboratory conditions could be largely discounted if immunolabelling for Big ET-1 was successful. The sections incubated in anti-Big ET-1 demonstrated immunolabelling that was comparable to those of the previous experiment. The sections incubated in anti-ET-1 did not display immunolabelling of any vascular structures. However, the multinucleated cells that had been found to be immunoreactive to anti-Big ET-1, were also immunolabelled by the anti-ET-1 antibody. These cells did not fluoresce in any of the control slides. Low levels of fluorescence were also noted in some epithelial cells lining the proximal convoluted tubules of renal nephrons. The implications of these findings are discussed in Chapter 6. 5.3.5 Experiment E As immunolabelling for Big ET-1 was observed in segments of the vascular endothelium of the marmoset kidney, it was deemed appropriate to investigate the distribution of Pre- pro ET-1. The tissue sections were incubated in rabbit anti-Pre-pro ET-1 antibody (Peninsula) followed by the secondary antibody, donkey anti-rabbit lgG, conjugated with Cy3tt (Appendix 8.28d). The same protocol was otherwise used. lmmunolabelling was observed in the multinucleated tissue cells, in occasional endothelial cells of the peritubular capillaries (Figure 16) and in some epithelial cells lining the proximal convoluted tubules. 52 5.3.6 Experiment F Further investigation into the establishment of a reliable laboratory protocol for immunolabelling vascular ET-1 involved testing another proprietary brand of two anti-ET-1 antibodies on marmoset kidney sections. These were monoclonal anti ET-1 antibodies raised in mouse (Yamasa, purchased via Sapphire Bioscience, Alexandria, NSW), one specific for the Nterminal of ET-1, the other specific for the C-terminal (Appendix 8.28e). Donkey anti-mouse lgG conjugated with Cy3ru (Jackson) served as the secondary antibody. The sections demonstrated striking immunolabelling of the cytoplasm and brush border of the epithelial cells lining the proximal convoluted tubules. However, no vascular elements of the kidney demonstrated immunofluorescence (Figure 17). lmmunolabelling of the epithelial cells was not universal, but rather occurred in clusters throughout the cortex. lsolated collecting ducts within the renal medulla also demonstrated cytoplasmic imm unolabelling. 5.3.7 Experiment G From findings of the foregoing experiments, it was reasoned that ET-1 might be present in such small amounts that its detection may require the incorporation of a signal amplification procedure, such as the streptavidin-biotin technique, into the methodology. After overnight incubation in rabbit anti-ET-1 antibody (Peninsula), sections of marmoset kidney were washed three times for 1O minutes in 10% DNS and incubated for 2 hours in donkey anti-rabbit biotin (Jackson) at a dilution of 1:50. The sections were then washed in the same manner, and incubated in Cy3rM-conjugated streptavidin (Jackson) for 1.5 hours. The sections were washed and mounted in the same manner as previous experiments. The IHC control sections did not display immunofluorescence. The experimental sections displayed higher levels of background labelling than the conventional indirect technique used to this point. This made interpretation of the results more difficult. The multinucleated cells labelled for ET-1 as previously. Endothelial cells of several larger blood vessels of the cortex displayed immunolabelling (Figure 18). No such labelling was seen in the renal glomeruli or the peritubular capillaries. 53 5.4 Pilot Study No 3 (QB-END 10) lmmunolabelling of endothelial cells was not observed using the QB-END 10 endothelial cell marker on skeletal muscle, kidney or gut of the marmoset. 5.5 Pilot Study No 4 (JC-704) Specific immunolabelling of endothelial cells with the JC-704 antibody was observed in tissue sections of marmoset stomach with little background staining (Figure 19). No labelling occurred in the control sections. The most satisfactory immunolabelling occurred at a primary antibody dilution of 1:80. This antibody was deemed a satisfactory pan-endothelial marker in marmoset stomach. 5.6 Pilot Study No 5 (a-SMA) positive immunolabelling for o-SMA was found in tissue sections of marmoset proximal small intestine (Figure 20). The best "signal-to-noise" ratio occurred at a primary antibody dilution of 1:200. No positive immunolabelling was observed on the control slides. The cr-SMA antibody constituted a satisfactory immunolabelfor periendothial cells in marmoset gut. 54 Figure 13. lmmunolabelling of RECA-1 antigen in rat kidney, demonstrating peritubular capillaries. Secondary antibody: lgG/Cy3. Scale bar = 500Pm. Figure 14. lmmunolabelling of artery endothelium for Big ET-1 in marmoset kidney Secondary antibody: lgG/Cy3. Scale bar = 500pm' 55 Figure 15. lmmunolabelling of capillary loops for Big ET-1 in glomerulus of marmoset kidney Secondary antibody: lgG/Cy3. Scale bar = 500Pm. Figure 16. lmmunolabelling of peritubular capillaries for Pre-pro ET-1 in marmoset kidney Secondary antibody: lgG/Cy3. Scale bar = 5001.¡m. 56 Figure 17. lmmunolabelling of basal cytoplasm and brush border of epithelial cells of proximal convoluted tubules for ET-1 in marmoset kidney (arrows). Note lack of labelling of glomerulus (G), artery (A) and peritubular capillaries. Secondary antibody: lgG/Cy3. Scale bar = 500pm. Figure 18. lmmunolabelling of endothelial cells of artery for ET-1 in marmoset kidney using streptadivin-biotin amplification. Tertiary antibody: Streptavidin/Cy3. Scale bar = 5001.1m. 57 Figure 19 lmmunolabelling with JC-7OA antibody (PECAM-1 antigen) of endothelial cells of capillaries surrounding tubular glands of gastic mucosa in marmoset (solid arrows). Open arrows denote immunolabelled arterioles. Secondary antibody: lgG/Cy3. Scale bar = 500pm' Figure 20. lmmunolabelling of a-SMA within rugae of proximal small intestine of marmoset. Labelled structures are likely to be pericytes. Secondary antibody: lgG/Cy3. Scale bar = 1000pm. 58 5.7 Summary of Findings of Main Experiments 5.7.1 Endothelin-l Experiments were performed to establish immunolabelling of vascular ET-1 in marmoset jaw sections. Anti-ET-1 and anti-Big ET-1 primary antibodies (Peninsula) were used. Experiment 1. Tested immunolabelling of vascular ET-1 and Big ET-1 in jaw sections using lgG/FITC secondary antibody. Results inconclusive. Experiment 2. lmmunolabelled vascular ET-1 in jaw sections using lgG/Cy5rM secondary antibody. Experiment 3. lmmunolabelling for ET-1 in unloaded and loaded jaw sections using methodology established in Experiment 2. ET-1 expression within MVB of the PDL was studied under the fluorescence microscope and CLSM. 5.7.2 JC-704 Pan-endothelial immunolabelling in marmoset jaw sections was established and served two purposes: L Vascular morphology in the paradentaltissues was examined. 2. Quantification of PECAM-1 antigen expression was investigated using the CLSM and CoMOS software. 5.7.3 cr-SMA Positive immunolabelling of periendothelial cells in marmoset jaw sections was established. Periendothelial cell morphology was examined using the fluorescence microscope and CLSM. 5.8 Exper¡ment No f (ET-1) 5.8.1 Experiment I After immunolabelling for Big ET-1 in marmoset kidney sections, a jaw section was incubated in anti-Big ET-1 (Peninsula). Two sections served as controls via the substitution of the primary antibody with 10% DNS. One control section was incubated in DNS only, the other in DNS, followed by the secondary antibody. A fourth section was incubated in anti-ET-1 (Peninsula). Both primary antibodies were applied at a dilution of 59 1 in 50. Finally, the sections were incubated in the secondary antibody donkey anti-rabbit lgG/FlTC. This secondary antibody was used instead of donkey anti-rabbit lgG/Cy3rM, as the latter was not available at the time. This experiment constituted a trial of the IHC methodology using jaw sections. The sections used were derived from the extreme buccal or lingual aspects of the mandibular bodies and contained only small fragments of dental and periodontal tissue. Sections from these areas of each mandibular body were used either for trial experiments or as IHC controls. The IHC control sections demonstrated no immunofluorescence and appeared generally dull under the fluorescence microscope. The section incubated in the anti-ET-1 antibody did not demonstrate immunolabelling. This was not unexpected following the results of the pilot studies. However, immunolabelling for Big ET-1 was also unsuccessful. Big ET-1 had been successfully labelled in marmoset kidney sections using donkey anti- rabbit lgG/Cy3rM. A factor that may have contributed to the negative result in the jaw sections was the use of the secondary antibody, |gG/FITC (Page 85)' 5.8.2 Experiment 2 Although the pilot studies had indicated that ET-1 is present in marmoset kidney at any one time in only minute amounts, one of the author's supervisors, Professor M. R. Sims, had demonstrated that in rat kidney it was possible to label this protein using the indirect IHC technique. This was made possible by use of the secondary antibody, donkey anti- rabbit lgG/CySrM. Prior to this, ET-1 had only been immunolabelled in the current study using the streptavidin-biotin amplification technique. When the lgG/Cy5rM secondary antibody was used, tissues were viewed (via a video-linked monitor) using an infrared filter on the fluorescence microscope. The level of background staining was significantly reduced, and this facilitated detection of true immunolabelling. On the strength of this finding in rat tissue, another section of marmoset jaw was incubated with rabbit anti-ET-1 antibody (Peninsula), followed by donkey anti-rabbit lgG/Cy5rM (Jackson). Both antibodies were used at a dilution of 1:50. lmmunolabelling of endothelial cells of several of the visible blood vessels within the PDL was observed. The appearance of the PDL using this secondary antibody and filter combination was notably different when compared with previous experiments performed during the course of this 60 study. There was a significantly reduced level of background fluorescence, which enabled better visualisation of the immunolabelled microvasculature. 5.8.3 Experiment 3: Unloaded and Loaded Mandibular Sections Examination of ET-1 expression in the periodontal microvaculature in unloaded and loaded conditions proceeded using the above methodology. lmmunofluorescence was not displayed by any of the IHC controls. ln the unloaded and loaded mandibular sections, immunolabelling for ET-1 was observed within portions of the microvascular endothelium of the PDL (Figures 21, 22 and 23), pulp and bone (Figure 24). lmmunolabelled vessels were identified in the apical, middle and crestal thirds of the PDL of the three teeth within the mandibular sections (PM3, M1 and M2). However, immunofluorescence was not observed in all the visible blood vessels. Only the cytoplasm of the vascular cells was labelled. Therefore, the unlabelled nuclei were well defined (Figures 22, 23 and 25). The vascular smooth muscle cell nuclei of an immunolabelled terminal arteriole in the bone projected abluminally (Figure 25). Quantification allowing comparison of ET-1 expression in unloaded and loaded conditions was not undertaken and is discussed in Chapter 6. 5.9 Exper¡ment No 2 (JG-704) 5.9.1 Vascular Morphology Pan-endothelial immunolabelling of the dental and paradental microvasculature of the marmoset was achieved using the JC-704 antibody at a dilution of 1:40 and the lgG/Cy3rM secondary antibody at 1:50. Vascular endothelium was clearly labelled within pulpal tissues, and throughout the periodontium of the teeth (Figure 26). Larger blood vessels within the medullary bone that were continuous with those of the periodontium were observed. Branches of these blood vessels were seen to traverse the cribriform plate of tooth sockets and enter the PDL, where they branched into smaller vessels (Figure 27). Some of these branches were seen to course within the PDL parallel to the tooth root. Capillary hairpin loops were identified in the crevicular and col regions of the gingivae, the latter forming connections with the vasculature of the underlying alveolar bone (Figure 28). 61 The intercellular distribution of immunofluorescence meant that it was possible to delineate the outline of individual endothelial cells at higher magnifications using the fluorescence microscope (Figure 29). Overall vascular distribution within the dental and paradental tissues was also well demonstrated when the sections were viewed under the CLSM at lower magnifications (Figures 30 - 32). The vascular component of the neurovascular bundle within the inferior alveolar canal was also well labelled (Figure 33). Morphological detail of the endothelial component of the vasculature was even greater when merged images were constructed using the CLSM and CoMOS software (Figures 34 and 35). Following these experiments, it was concluded that the laboratory methodology used to identify the MVB of the dental and paradental tissues within sections of marmoset mandible via IHC was successful. 5.9.2 Anti gen ExPression The JC-704 and lgG/Cy3rM immunolabelling technique had proved reliable, and provided the first opportunity to implement quantification of antigen expression as detailed on page 47. At this time, a methodology for immunolabelling vascular ET-1 in the jaw sections had not been established. An unloaded and a loaded section from the same animal were concurrenly processed for endothelial cell immunolabelling according to the same protocol as the previous experiment. Using the CLSM, a vessel with a luminal diameter of between 10 and 20pm (postcapillary venule size) in the apical area of the mesial and distal roots (Figures 36 and 37) and in the interradicular area (Figure 38) of the first molar from the unloaded and the loaded sections were selected for quantification of pECAM-1 expression (the antigen to JC-704). As the load had been applied from an occlusal direction, it was reasoned that these areas of the PDL constituted sites of greatest pressure concentration, and therefore might display most variability in PECAM-1 expression. Data generated for pECAM-1 expression are provided in Appendix 8.32. Due to the fact that these data were derived from only one "set" of jaw sections from one animal, statistical tests of significance were not indicated. However, these data and the implications of this method of quantification of microvascular immunolabelling are discussed in ChaPter 6. 62 5.10 Experiment No 3 (cr-SMA) lmmunolabelling with the anti-cr-SMA antibody in the marmoset periodontium enabled visualisation of periendothelial cells using either the fluorescence microscope (Figure 39) or the CLSM. At high magnification under the fluorescence microscope, it was possible to study periendothelial cell morphology and recognise those features consistent with pericytes (Figures 40 and 41) and vascular smooth muscle cells. Generally, cells positive for this antibody in the pulpal tissues displayed morphological characteristics consistent with vascular smooth muscle cells. lmmunolabelled cells in the alveolar bone resembled either pericytes or vascular smooth muscle cells. Periendothelial cells within the PDL were more difficult to detect than in the pulp and alveolar bone. Both cell types were observed in the PDL, but more resembled pericytes than vascular smooth muscle cells. The "merging" facility of the CoMOS software enabled construction of highly detailed images of immunolabelled periendothelial cells. The most striking demonstration of such detail was that of the periendothelial cells associated with two blood vessels within the PDL on the labial side of a mandibular lateral incisor (Figures 42 and 43). These vessels coursed parallel to each other in an occluso-apical direction and possessed a maximum luminal diameter of approximately 22pm. This measurement was derived from one optical section of the merged image. The ultrastructural characteristics of the periendothelial cells were consistent with those of vascular smooth muscle cells and this, together with the luminal dimensions, suggested that these vessels were terminal arterioles. ldentification of periendothelial cells in the vasculature of marmoset jaws using IHC according to the aforementioned protocol proved satisfactory. 63 Figure 21 CLSM image of endothelial cells immunolabelled for ET-'l within PDL of interradicular area of mandibular first permanent molar of marmoset. B = Bone. MR = Mesial root. DR = Distal root Secondary antibody: lgG/Cy5. Scale bar = 65Um. Figure 22. CLSM image of endothelial cells of postcapillary-srzed venule immunolabelled for ET-1 within PDL of distal root apex of mandibular second permanent molar of marmoset Arrows indicate endothelial cell nuclei B = Bone. Secondary antibody: lgG/CyS. Scale bar = 50pm 64 Figure 23 CLSM image of endothelial cells of collecting venule immunolabelled for ET-1 in apical PDL of distal root of mandibular first permanent molar of marmoset. Arrows indicate well- defined nuclei of endothelial cells. Only the endothelial cell cytoplasm is labelled. B = Bone. Secondary antibody: lgG/Cy5. Scale bar = 20pm. Figure 24. CLSM image of endothelial cells immunolabelled for ET-1 within collecting venule bordering PDL and bone (B) in apical area of distal root of mandibular first permanent molar of marmoset Rectangle denotes area presented in Figure 25. Secondary antibody: lgG/Cy5 Scale bar = 150pm. 65 Figure 25. CLSM image of terminal arteriole immunolabelled for ET-1 within bone apical to mandibular first permanent molar of marmoset. Arrows indicate nuclei of vascular smooth muscle cells projecting abluminally. Secondary antibody: lgG/Cy5. Scale bar = 30pm. 66 I Figure 26 Fluorescence microscope view of sagittal section of mandibular canine and first premolar of marmoset demonstrating immunolabelling of endothelium within pulp (P), gingivae (G), bone (B) and PDL using JC-704 antibody. Secondary antibody: lgG/Cy3 Scale bar = 2000pm I I I Figure 27 Fluorescence microscope view of blood vessels traversing mesial cortical plate of socket of mandibular second premolar between medullary bone (B) and PDL of marmoset using JC-704 antibody Secondary antibody: lgG/Cy3 Scale bar = '1000pm. 67 . : ::.:tl Figure 28 Fluorescence microscope view of capillary hairpin loops in crevicular (CR) and col areas between mandibular canine and first premolar of marmoset using JC-704 antibody Arrows denote communications of loops with vasculature of underlying alveolar bone Secondary antibody: lgG/Cy3. Scale bar = 1000pm. Figure 29 Fluorescence microscope view of collecting venule within distal PDL of mandibular second premolar of marmoset using JC-704 antibody Endothelial cell outlines are visible B = Bone. Seeondary antibody: lgG/Cy3 Scale bar = 500pm. 68 Figure 30. CLSM image of microvasculature labelled with JC-704 antibody within gingiva (G), bone (B), pulp (P) and PDL of marmoset mandible Secondary antibody: lgG/Cy3 Scale bar = 500pm. Figure 31. CLSM image of microvasculature labelled with JC-704 antibody in bone (B), pulp (P) and PDL of marmoset mandible. Arrows indicate blood vessels traversing cortical plate of tooth sockets Rectangle denotes area presented in Figure 32. Secondary antibody: lgG/Cy3. Scale bar = 500pm. 69 Figure 32. CLSM image of microvasculature labelled with JC-70A antibody adjacent to mesial root apex (M1) of mandibular first permanent molar of marmoset. Arrows indicate communication of blood vessels between alveolar bone and PDL Secondary antibody: lgG/Cy3. Scale bar = 200pm. 'Çç'i,'.. Figure 33 CLSM image of microvasculature labelled with JC-70A antibody in subapical region of marmoset mandible. IAV = inferior alveolar vessels. M2 = apicies of mandibular second permanent molar. Secondary antibody: lgG/Cy3 Scale bar = 500pm. 70 Figure 34 Merged CLSM image of endothelial cells of a collecting venule within PDL of distal root apex of mandibular first permanent molar of marmoset. Primary antibody: JC-704. Secondary antibody: lgG/Cy3. Scale bar = 50pm. ti, : -,$ 4i"- n Figure 35. Merged CLSM image of endothelial cells of a collecting venule within PDL of distal root of mandibular first permanent molar of marmoset. B = lnterradicular bone. Primary antibody: JC-704. Secondary antibody. lgG/Cy3 Scale bar = 50pm. 71 i ..r**sst:, Figure 36. CLSM images showing marmoset mandibular first permanent molar apical vessels selected for quantification of PECAM-1 expression (arrows). CTME: control tooth mesial apex CTDI: control tooth distal apex. Primary antibody: JC-704. Secondary antibody: lgG/Cy3 Scale bars = 100pm 72 T ' a:.':: Figure 37. CLSM images showing marmoset mandibular first permanent molar apical vessels selected for quantification of PECAM-1 expression (arrows) EXME: experimental tooth mesial apex. EXDI: experimental tooth distal apex. Note "basket-like" arrangement of blood vessels enveloping apex in EXDI Primary antibody: JC-704. Secondary antibody. lgG/Cy3 Scale bars = 100pm 73 Figure 38. CLSM images showing marmoset mandibular first permanent molar interradicular PDL vessels selected for quantification of PECAM-'1 expression (arrows). CTIR: control tooth interradicular area. Scale bar = 200pm. EXIR: experimental tooth interradicular area. Scale bar = 100Um B = lnterradicular bone. P = Pulp Primary antibody: JC-704. Secondary antibody: lgG/Cy3 74 Figure 39. Fluorescence microscope view of immunolabelled c¿-SMA within periendothelial cells of marmoset PDL, alveolar bone (B), gingiva (G) and pulp (P). Secondary antibody: lgG/Cy3. Scale bar = 2000pm. Figure 40. Fluorescence microscope view of immunolabelled c¿-SMA within pericytes of collecting vein. Vessel is located in alveolar bone adjacent to mesial aspect of the PDL of mandibular second permanent incisor of marmoset. Solid arrow points to cell body and nucleus of a pericyte. Open arrows locate several cytoplasmic processes extending from cell body. Secondary antibody: lgG/Cy3. Scale bar = 500pm. 75 Figure 41. Fluorescence microscope view of immunolabelled cr-SMA within pericytes of collecting vein located in alveolar bone apical to mandibular first permanent incisor of marmoset. Arrows indicate cr-SMA distributed within abundant cytoplasmic processes of pericyfes. Secondary antibody: lgG/Cy3. Scale bar = 130pm. 76 Figure 42. CLSM image of immunolabelled a-SMA within vascular smooth muscle cells of terminal arterioles located within labial PDL of mandibular permanent lateral incisor of marmoset Boxed area denotes area presented in Figure 43. B = Labial bone. Secondary antibody: lgG/Cy3. Scale bar = 100pm Figure 43. Merged CLSM image of immunolabelled cr-SMA within vascular smooth muscle cells of terminal arterioles located within labial PDL of mandibular permanent lateral incisor of marmoset. Solid arrows indicate cell bodies. Open arrows indicate cellular processes. Secondary antibody: lgG/Cy3 Scale bar = 201.¡m 77 Chapter 6 DISCUSSION 6.1 Material The advantages of breeding marmosets for biological research, together with the fact that their periodontium is considered histologically analogous to man, has resulted in a long history of use of this animal model in dental biological research. However, several disadvantages should be noted. The use of non-human primate material for biological research is costly, although marmosets are considered inexpensive in comparison to other primates. This imposes financial limits on the amount of dental tissues available for such studies. Where methodological difficulties were encountered, as occurred in relation to vascular ET-1 immunolabelling, of primary Goncern was the use of valuable dental material without the production of viable results. lt was only possible to obtain a maximum of three sagittal sections from each side of a mandible, in addition to the most buccal and lingual sections. The pilot studies performed on soft tissues served to largely address this problem. However, experiments on the jaw sections were ultimately required. The most buccal and lingual sections proved useful for initial trials of IHC on dental material, as did the anterior sections. The maxillae were stored in anticipation of performing future research using the technique established in the present investigation. However, they also served as "backup" material for the current study if required. Consideration also had to be given to the consumption of expensive antibodies in repeatedly performing pilot studies with minor alterations when endeavouring to establish a laboratory methodology. 6.2 Load Application The occlusally-directed load applied across the three posterior teeth on the experimental side of the mouth was of "low" magnitude, the rationale being that only light, continuous forces are required clinically to stimulate the necessary cellular activity within the periodontium to elicit bone remodelling and tooth movement. ln his research concerning distal movement of canines in man, Storey (1973) considered a 2509m load on one mandibular canine to be "light", ln the present study, the 120-2}0gm force applied externally to the head would have been partially dissipated by the soft tissues, delivering a light load across the three relatively small posterior teeth in the marmoset. The device 78 used in this study proved adequate for the purpose of application of a light occlusally directed load. lt was also simple and inexpensive to construct. During the experiment, a degree of empiricism was required to determine if a positive load was applied to the teeth by the device. This was achieved by gently pulling the intra-oral pad and checking for resistance of the pad to movement. 6.3 lmmunoh¡stochemical Methodology 6.3.1 Fixation ln past experiments in the Department of Anatomy and Histology at Flinders University, Zamboni's fixative (Stefanini et al., 1967) had been found to provide adequate preservation of tissue structure without affecting antigen availability for lHC. Therefore, it was the fixative of choice for the current experiments. Examination of the tissues under the fluorescence and confocal microscopes showed that tissue integrity was preserved in the majority of experiments. There was some change in tissue morphology noted during the streptavidin-biotin amplification pilot studies using kidney tissue, but this was related to the amplification procedure rather than the method of tissue fixation. A component of Zamboni's fixative, picric acid, may have potentially constituted a barrier to immunolabelling for ET-1. Care was required to ensure the tissues were thoroughly rinsed in graded alcohols to remove as much picric acid as possible prior to starting lHC. The method of fixation was by tissue immersion rather than perfusion, as previous experience in the laboratory had demonstrated that Zamboni's fixative adequately penetrated the tissues using this technique. However, as a precaution, the soft tissue capsular structures were removed and tissue samples were no greater than Smm in width. Also, the oral mucosa was elevated from the jaws prior to complete immersion in fixative. Nevertheless, an advantage of perfusion fixation over immersion fixation is that blood cells are largely flushed out of the vasculature. Red blood cells autofluoresce, which may prove distracting when examining vascular structures' 6.3.2 lmm u nohistochemical Gontrols ln order to test for specificity of the reaction, numerous immunohistochemical controls have been recommended (Van Noorden and Polak, 1983). 1. Pre-absorption of the primary antibody with the purified antigen will result in no labelling if there are no other antibodies present in the solution against other antigens 79 in the tissue. Although only small amounts of each antigen are required for pre- absorption, the cost may be prohibitive. 2. A positive control is a tissue known to have the antigen in question present. This tissue must undergo the same fixation and processing schedule as the experimental material. This will assist confirmation of a negative result in the experimental tissue. 3. A negative control is where either the primary antibody, or the primary and secondary antibody, is replaced by normal serum in antibody diluent. These controls were used routinely in the current study. 4. A substitution control is incubated in pre-immune serum (obtained from the animaljust before it was immunised for production of the antibody), or nonimmune serum (obtained from another non-immunised animal of the same species). Any immunolabelling of the substitution control is due to nonspecific protein binding. Pre- immune serum is not always readily available. The use of controls in the main experiments presented a problem due to the limited availability of jaw sections. Wherever possible, the outer buccal and lingual sections were used for this purpose. 6.3.3 Refinement of the Laboratory Methodology Descriptions of the pilot studies and the main experiments indicate that ongoing changes were made to the laboratory methodology during the course of the investigation. These changes were introduced in an effort to achieve optimal quality immunolabelling, with a high signal-to-noise ratio. A summary of these changes and their rationale may provide a useful reference for f uture i m m u nohistochem ical i nvestig ations. 1. Soft tissue sections were rinsed by pipetting 10% DNS diluted in antibody diluent on to the slides forthree, 10 minute intervals, in preference to dipping the slides into PBS, as it was reasoned that this change in methodology would be of benefit for the following reasons: i) The chance of background staining would be reduced, as the tissues were repeatedly exposed to the blocking agent, DNS. ii) Rinsing by pipetting rather than immersing reduced the amount of DNS required. iii) Rinsing by pipetting rather than immersing minimised the chance of sections floating off the slides as sometimes occurred with repeated washes in PBS baths. 80 2. During incubation of tissues in secondary antibody, the humid box was placed in a drawer, or covered, in order to minimise exposure of the sections to ambient light, so as to reduce the potential for fading of the fluorophor. After mounting, care was taken to shield the sections from light when they were not being viewed. 3. Penetration through the tissue sections with antibodies was at times incomplete. Subsequently, the clearing agent, DMSO, which "opens" the cell membranes, was applied undiluted for longer periods of time to improve antibody access to antigen sites throughout the sections. Penetration of primary and secondary antibodies to a depth of approximately 85¡rm from each side of the jaw sections was achieved despite the fact that the sections had not been demineralised. Higher antibody concentrations and longer incubation times also improved penetration. lt was possible to determine antibody penetration by focusing through the sections to assess the level of fluorescence at differing focal planes. 4. The preferred mounting medium was 20% buffered glycerol, pH 8.6, as this reduced the rate of fading of the fluorophor. 6.3.4 Limitations of the Gurrent Laboratory Methodology Whilst development of the current laboratory methodology has the potential to assist future biological orthodontic research, its inherent limitations should be considered. 1. lmmunohistochemistry is technique-sensitive, and it was the experience of the author that slight variations in methodology could deleteriously affect results. 2. Fading of the fluorophor, sometimes within several days, can be minimised but not avoided. Therefore, tissue sections should be viewed as soon as possible after IHC is completed. 3. Occasionally, superimposition of structures under the fluorescence microscope occurred due to the thickness of the jaw sections such that the location of a blood vessel could be difficult to establish. This was particularly evident at the PDL/bone junction, where a blood vessel within the alveolar bone might appear to be located within the PDL. This problem was eliminated when using the CLSM, as it can produce an image from a focal plane thickness of only 1¡lm. 4. This technique did not allow data collection regarding overall microvascular morphology, such as blood vessel branching patterns, to the extent of other techniques. The SEM examination of corrosion casts as used by Lee (1987) and 81 Stanley (1993) was more suited to this purpose. However, these authors also found corrosion casting to be technique-sensitive. 5. The area of dental tissue available for examination on a section was highly dependent upon the orientation at which the section was cut. Clarification of the exact location around a tooth of the area of PDL under examination at times proved ditficult. 6.4 lmmunoh¡stochemical Labelling w¡th RECA-1 The monoclonal antibody RECA-1 was introduced by Duijvestijn et al. in 1992, and is an lgGl isotype that recognises Rat-Endothelial-Cell-Antigens in vitro and in vivo. lt possesses unique properties when compared with other antibodies used for endothelial cell identification in that it is specific to endothelial cells, and is reactive to all detectable endothelium of the rat vasculature. Duijvestijn ef a/. (1992) found that RECA-1 was not reactive to tissues from other tested species, including those of the mouse, guinea pig, rabbit, goat, pig, sheep, chicken and human. Therefore, it was not surprising in the current pilot study to find the RECA-1 antigen was not detected in the marmoset tissues. However, positive labelling for the antigen occurred in the rat tissues (Figure 13). This pilot study proved worthwhile in that the author's laboratory technique was tested early in the investigation, before proceeding to experiments using marmoset tissue. 6.5 lmmunoh¡stochemical Labelling for ET-l 6.5.1 General Discussion Blood vessels are now considered to be more than simple tubes that are regulated by circulating hormones and the sympathetic nervous system, as "communication" between circulating blood cells and cells of the vessel wall, in addition to local mechanisms, also play important regulatory roles (Lüscher et a1.,1993). The vascular endothelium, which is in a strategic anatomical position between circulating blood and vascular smooth muscle (Lüscher et al., 1993), may be seen as not just a passive barrier to diffusion between blood and vascular smooth muscle, but also as a regulator of vascular function (Haynes et al., 1993). The field of research in biomedical sciences concerning the vascular endothelium has developed rapidly in recent years. A recognition of the fundamental role of vascular endothelium in the regulation of vascular smooth muscle tonus and growth, immunological reactivity (leukocyte emigration), blood coagulation, and 82 lipid metabolism, has emerged with the discovery of endothelium derived prostacyclin, nitric oxide and ET-1 (Webb, 1997). Casasco et al. (1991; 1992) demonstrated endothelin-like immunoreactivity limited to the vascular endothelium of human pulpal tissues and tooth germ. They hypothesised that pulpal blood pressure, blood flow and tooth development may be influenced by endothelin. They did not report investigation of the occurrence of endothelin in the PDL. Gilbert et al. (1992) proposed an hypothesis for a possible pathophysiological role of endothelin in pulpal necrosis following mechanical injury. Endothelin may be synthesized by endothelial cells as part of an intense local response to mechanical injury of the dental pulp. The relatively long duration of the resultant vasoconstriction may produce localised ischaemia and necrosis. Shiohama et al. (1995) reported that ET-1 immunoreaction occurred more intensely in the PDL of those teeth subjected to an orthodontic force. The magnitude of the force exeñed by placement of an elastic band between two rat molar teeth was not repoded, but would have been considerably higher than those applied in clinical orthodontics. lmmunoreaction to ET-1 was also more intense on the tension side of the PDL than the pressure side. Quantification of this immunoreactivity was not provided. ln addition to endothelial cells, ET-1 immunoreactivity was also detected in osteoclasts and fibroblasts. They suggested that ET-1 might play a physiological role in the regulation of bone and PDL metabolism during orthodontic tooth movement. lf it can be demonstrated that ET-1 is synthesised by the vascular endothelium of the PDL and adjacent alveolar bone as an early local response to mechanical stimulation, such as loading from orthodontic appliances, future investigations might be aimed at determining its role in the angiogenic, inflammatory and necrotic processes within the PDL tissues concomitant with orthodontic treatment. ln this context, the PDL microvasculature, and more specifically the endothelial cells, would represent one of the "biological sensors" to the loading demands placed on the tooth, communicating information by releasing cytokines such as ET-1 to elicit a biological response. Therefore, an aim of this investigation was to determine if ET-1 is produced in primate periodontal vascular endothelium and if so, to investigate whether the immunohistochemical expression of ET-1 in the MVB differs between the unloaded and the loaded PDL. 83 6.5.2 Pilot Study for ET-1 lmmunolabelling During the course of this pilot study, sections of kidney were used without the additional sections of skeletal muscle and gut as used initially. Kidney was found to more reliably adhere to the poly-L-ornithine coated slides than the other soft tissues during the washing procedures. Furthermore, Wilkes et al. (1991) had demonstrated endothelin-like immunoreactivity throughout all major regions of the kidney. The establishment of a laboratory protocol for immunolabelling ET-1 in marmoset endothelium that gave predictable results proved to be the most challenging of the pilot studies. lnitially, the only consistent immunolabelling for ET-1 occurred in the multinucleated cells, whereas vascular endothelial cells did not label for ET-1 using the indirect IHC technique. Ehrenreich et al. (1990) reported that cultured human macrophages synthesised ET-1 and ET-3, and demonstrated ET-1 and ET-3 immunoreactivity in tissue macrophages in paraffin sections of human lung tissue. They suggested that macrophage-derived endothelins might play a role in blood vessel physiology. Therefore, the immunolabelling of the multinucleated cells with the anti-ET-1 antibody as occurred during these pilot studies was consistent with the findings in tissue macrophages of other investigators. Early in the current pilot study, the absence of labelling for ET-1 in the vascular endothelial cells was interpreted as a problem possibly attributable to the batch of anti-ET-1 antibody (Peninsula). However, the lack of fluorescence of tissue multinucleated cells in the control sections, together with the consistent immunolabelling of these cells following incubation in anti-ET-1 antibody indicated that the batch of primary antibody was viable. As an extra precaution, another proprietary brand of two anti-ET-1 antibodies (Yamasa), one specific to the N-terminal, the other to the C-terminal of ET-1, was tested. These antibodies, particularly that specific for the N-terminal, demonstrated stronger immunolabelling of occasional epithelial cells of the proximal convoluted tubules within the renal cortex, but no labelling of the vascular elements (Figure 17). Wilkes ef a/. (1991) described endothelin-like immunoreactivity in the most proximal portion of the proximal tubule brush border in the rat kidney, and suggested that ET-1 may have important physiological actions in this nephron segment. lmmunolabelling of these epithelial cells for ET-1 in the current pilot study was, therefore, also supported by the literature. 84 lmmunolabelling for Big ET-1 in vascular endothelium of peritubular capillaries, capillary loops of some glomeruli, and in some, but not all renal blood vessels, was achieved using the same indirect technique as was applied when attempting to label for ET-1 (Figures 14 and 15), Wilkes et at. (1991) described a similar distribution of "endothelin-like" immunoreactivity in rat kidneys using a polyclonal antibody that recognised both ET-1 and Big ET-1. ln the current study, Pre-pro ET-1 was also successfully labelled using the indirect technique. However, unlike Big-ET-1, its vascular distribution was limited to the peritubular capillaries. Pre-pro ET-1 was also identified in some epithelial cells lining the proximal convoluted tubules and in tissue multinucleated cells. The findings indicated that, at any one time under normal physiological conditions in the renal vasculature, ET-1 is present in very small amounts, and that its detection using IHC requires an excellent signalto-noise ratio, or an 'amplification' technique sensitive enough to detect such minute quantities. The streptavidin-biotin amplification technique was tested, and ET-1 was detected within endothelial cells of larger blood vessels of the renal cortex (Figure 18). This technique was useful in that it was sensitive enough to allow such detection of vascular ET-1, however it produced high levels of background labelling. Detection within the renal vasculature of the ET-1 precursors, Pre-pro ET-1 and Big ET-1, without signal amplification indicated that these precursors, particularly Big ET-1, might constitute an intra-cellular pool ready for enzymic conversion to ET-1 as physiological conditions demand. However, Webb (1997) suggested that regulation of endothelin synthesis occurs primarily at the level of gene transcription in response to endothelial cell stimulation. Fujita et al. (1994a) demonstrated in the rat kidney the local enzymic conversion of exogenously administered Big ET-1 to ET-1 in the renal vasculature, and suggested this may be due to the presence of a functional phosphoramidon-sensitive ECE in this tissue. Fujita et al. (1994a) also demonstrated that Big ET-1 is not inactivated during a single passage through the kidney and suggested that the kidney might therefore be a site for inactivation of ET-1 but not of Big ET-1 in the rat. Previously, Abassi ef a/. (1992) had proposed that the kidney is important for the inactivation of circulating ET-1. Furthermore, Fujita et at. (1994b), showed that after 60 minutes of perfusion of the rat kidney with exogenous ET-1, only 10o/o of the applied ET-1 was recovered from this tissue, demonstrating the existence of the process of ET-1 degradation. They postulated that neutral endopeptidase 24.11 (NEP) might be responsible for local proteolytic degradation of ET-1 in the kidney. Similar investigations of Big ET-1 and ET-1 renal 85 physiology in non-human primates are currently lacking. However, the findings from the present investigation are consistent with the aforementioned studies concerning the rat. 6.5.3 Secondary Antibody for Anti-ET-l The donkey anti-mouse lgG secondary antibody conjugated with Cy3rM had been used in the other pilot studies and produced an excellent fluorescent signal at a dilution of 1:100. The secondary antibody, donkey anti-rabbit lgG/Cy3rM (Jackson), also became available whilst the methodological difficulties of immunolabelling for vascular ET-1 were being addressed, and was thereafter used in favour of the DTAF and FITC fluorochromes. ln combination with the primary antibodies for Pre-pro ET-1 and Big ET-1, the lgG/Cy3rM secondary antibody produced a good fluorescent signal. However, this was not the case when labelling for ET-1. Late in the investigation, an lgG/Cy5rM (Jackson) secondary antibody was purchased. lt was found to enhance the signal-to-noise ratio beyond that of lgG/Cy3rM to the extent that it was possible to detect the minute amounts of vascular ET-1 present in the tissues without requiring a signal amplification technique. Therefore, this antibody was used when immunolabelling for vascular ET-1 in the marmoset jaw sections, 6.5.4 ET-l Antibody Cross-Reactivity Specifications of the anti-ET-1 antibody provided by Peninsula Laboratories, lnc. indicated that it possessed l\Oo/o cross-reactivity with porcine Big ET-1 and 760/o cross-reactivity with human Big ET-1. No information was available regarding cross-reactivity with primate Big ET-1. However, it is interesting to note that during this pilot study in marmoset tissues, the patterns of immunolabelling for ET-1 and Big ET-1 were quite distinct. This indicated that significant cross-reactivity between the anti-ET-1 antibody and the marmoset Big ET-1 antigen was unlikely. 6.5.5 lmmunolabelling for ET-1 in Dental Tissues The production of ET-1 within the endothelium of the marmoset paradental microvasculature, including that of the PDL, was confirmed in this study using rabbit anti-ET-1 antibody (Peninsula) together with donkey anti-rabbit lgG/CySrM (Jackson). This cytokine was identified within terminal arterioles, postcapillary-sized venules and collecting venules (Figures 21-25) of the paradental microvasculature. The existence of ET-1 within other cellular elements of the PDL, as reported by Shiohama et al. (1995) was 86 not ascertained in the current study. A well-defined fluorescent signal was only observed within vascular cells, which concurs with the findings of Casasco et al. (1991; 1992). The existence of this potent vasoconstrictor within the MVB of the PDL implies that this vascular system might possess contractile properties. From physiological experiments involving forearm vasculature, Haynes et al. (1993) suggested that ET-1 production plays a role in the maintenance of basal vascular tone. The current findings point to the possibility of such a role for this cytokine within the marmoset PDL. Edwall et al. (1995) stated that there is conflicting evidence as to the existence of a local autoregulatory mechanism in the PDL circulation, and that systemic arterial pressure and sympathetic nerve activity would probably override such a mechanism. The predominance of thin-walled collecting venules and apericytic postcapillary-sized venules within the marmoset PDL (Sims, 1995) is more suggestive of a vascular system with minimal vasconstrictive potential. Further research is required to clarify the roles of ET-1 within the MVB of the PDL. Consideration should be given to the other functions attributed to ET-1 in different body systems. Haynes (1995) has stated that it is also associated with mitogenic activities, salt and water homeostasis and the generation of other vasoconstrictive agents. Pedram ef al. (1997) have also suggested a role for ET-1 in endothelial cell proliferation via the stimulation of vascular endothelial cell growth factor. Dzau et al. (1992) stated that vascular endothelium and smooth muscle cells may play a role in vascular remodelling by the expression of autocrine-paracrine growth factors such as platelet-derived growth factors, angiotensin ll, and endothelin. As ET-1 possesses mitogenic properties, the expression of ET-1 during the process of angiogenesis is a potential area for future research in orthodontic biology. Sims (1995) stated that a combination of techniques is required to elucidate the relationships existing between structure and function in the MVB of the PDL. The methodology used in this investigation represents a technique with potential to enable contribution to such an end. Future research might involve double labelling with the JC-704 and lgG/Cy3rM antibodies together with the anti-ET-1 and lgG/Cy5rM antibodies. This would enable PDL blood vessel identification and determination of ET-1 production according to vessel type. Such research might also involve location of periendothelial cells using anti-c¿-SMA (Page 91) so as to study their relationship with those endothelial cells that produce ET-1. 87 6.5.6 Production of ET-1 in the MVB of the Unloaded and Loaded PDL Quantification of ET-1 expression to enable objective comparison of the unloaded to the loaded PDL was not undertaken by the conclusion of this project. Production of data required for determination of signal-to-noise ratios had been performed for quantification of PECAM-1 expression in one set of jaw sections (Appendix 8.32). This had demonstrated that limitations imposed by time and expense would prohibit the generation of data for ET-1 quantification in the present investigation. However, in those mandibular sections immunolabelled for ET-1, variation in ET-1 production within the MVB of the PDL due to load application was not observed under the fluorescence microscope or the CLSM. Further investigation of ET-1 production within the unloaded and loaded PDL using the laboratory technique established and the method of antigen quantification described is now possible. 6.6 lmmunoh¡stochemical Labelling for Endothelium 6.6.1 Pilot Study for Pan-endothelial Immunolabel ldentification of a suitable pan-endothelial immunolabel during the pilot studies would enable identification of the vasculature throughout the marmoset dental and paradental tissues. The endothelial cell marker, QB-END 10 (Novocastra), an antibody to the CD34 antigen, was tested without success. Another endothelial cell marker, JC-704 (Dako), an antibody to PECAM-1 (CD31) was tested, as it was the preferred antibody over von Willebrand factor and lJlex europaeus agglutinin-1 (UEA-1) by Christenson and Stouffer (1996) when used in the corpus luteum of rhesus monkeys (Macaca mulatta). These investigators found that JC-704 produced homogenous endothelial labelling with little background staining. Similarly in the current pilot study, intense immunolabelling of endothelial cells with little background staining was achieved in tissue sections of marmoset stomach (Figure 19) using the JC-704 primary antibody and lgG/Cy3rM secondary antibody (Jackson). Parums et al. (1990) were the first investigators to report the production of 'JC70', a monoclonal antibody that was found to react with a formalin-resistant membrane bound glycoprotein on endothelial cells. Newman et al. (1990) designated this cell-surface protein as PECAM-1 (CD31) and reported its molecular cloning. ln blood smears, JC70 88 also labelled a variety of non-endothelial vascular cells, including neutrophil polymorphs, lymphocytes, monocytes and platelets. ln tissue sections it labelled megakaryocytes and occasional plasma cells in bone marrow, in addition to endothelial cells. However, Parums et al. (1990) found that in fixed tissue, JC70 essentially only labelled endothelium. Later, Delisser et al. (1994) suggested that PECAM-1 is a unique molecule in that it displays a wide distribution among cells of the vascular compartment. On the basis of these reports and the findings of the current pilot study, this antibody was deemed suitable to test as a pan-endothelial immunolabel for fixed oral tissues in the marmoset. 6.6.2 Endothelial Gells in Dental Tissues lmmunolabelling of endothelial cells using the JC-70A antibody against PECAM-1 within the marmoset jaw sections was successful. 6.6.2,1 Microvascu lar Morphology The JC-704 antibody proved a useful immunohistochemical label of endothelium for the study of microvascular morphology. Overall blood vessel distribution was most readily viewed using the fluorescence microscope. A more detailed examination of vascular structure was possible using the CLSM. Features of the marmoset oral microvasculature evident in the current study were consistent with previous descriptions. Blood vessels within the PDL were oriented in a largely occluso-apical direction, as seen by Lee (1987). Branches communicating between the bone and PDL via perforations in the cribriform plate as described in the rat (Weekes, 1983), mouse (Wong and Sims, 1987) and marmoset (Lee, 1987) were observed in the current investigation (Figure 27). Capillary hairpin loops in the interdental col regions as described by Weekes and Sims (1986) and Lee (1987) were also seen (Figure 28), as were the crevicular capillary loops reported by Stanley (1993). The "basket-like" arrangement of vessels surrounding the apex of a tooth, with numerous anastomoses, as described by Stanley (1993), was also evident in the current study (Figure 37). The observed pattern of PECAM-1 distribution over endothelial cells was also consistent with that of previous studies. Simmons et al. (1990) observed localisation of PECAM-1 to regions of cell-cell contacts, and suggested that this peptide functioned as an intercellular adhesion molecule. Delisser et al. (1994) stated that concentration of this molecule at 89 the borders of endothelial cells was a characteristic feature, and this was illustrated in the current study (Figure 34). 6.6.2.2 Cli nical I m pl ications The ability to locate the PECAM-1 antigen in the periodontium of the marmoset using IHC will potentially enable further investigation into vascular changes incident to orthodontic tooth movement. Recently, studies have implicated roles for PECAM-1 in the processes of inflammation and angiogenesis. Leukocyte recruitment into sites of inflammation from the blood (leukocyte emigration) involves these cells binding weakly to the endothelial cells and "rolling" along the vascular wall. They soon cease rolling and adhere tightly to the endothelial surface before leaving the circulation by squeezing between tightly opposed endothelial cells (Marchesi and Florey, 1960). Muller (1995) reported that after halting PECAM-1 function in an in vitro experiment, the transendothelial migration phase of neutrophil and monocyte emigration was blocked. This may have been caused by interference in the homophilic interaction between PECAM-1 on the leukocytes and PECAM-1 on the endothelial cells. Muller (1995) concluded that PECAM-1 is critical for this phase of leukocyte emigration, whereas other adhesion molecules (selectin and integrins) mediate the preceding phases. Agents that activate or antagonise the adhesive properties of PECAM-1 may therefore provide therapies for acute or chronic inflammatory conditions (Newman, 1997). As inflammation is a central process in tooth movement, these agents might also be applied in orthodontic treatment. Localised administration of a PECAM-1 blocker to the periodontium of an anchor tooth may increase its anchorage potential, therefore reducing or eliminating unwanted tooth movement. DeLisser et at. (1997) used an in vitro model (a three-dimensional endothelial cell culture system) and two in vivo models (neovascularisation induced by pellets in rat corneas and by subcutaneous pellets in mice) to evaluate the consequences of blocking PECAM-1 function on the process of angiogenesis. These authors showed in vitro and rn vivo that angiogenesis was inhibited following the blockade of PECAM-1, concluding that this adhesion molecule is important to this process. At present, the precise role of PECAM-1 in angiogenesis is uncertain. Delisser et al. (1997) stated that there are at least two possibilities. As PECAM-1 is localised to the intercellular areas, it may be required for the initial stabilisation of endothelial cell-cell contacts to enable tube formation. lt may also be 90 involved in the upregulation of integrins, which in turn play a role in endothelial cell migration. ln future, the role of PECAM-1 in angiogenesis should become more clearly defined, where it may emerge in therapies of pathological conditions dependent on the formation of new blood vessels. ln addition to its important function in angiogenesis, PECAM-1 is a potential marker for the study of this process. Page ef al. (1992) have already successfully labelled PEGAM-1 in vasculature undergoing angiogenesis. Both inflammation and angiogenesis are recognised processes in orthodontic tooth movement. Evidence is emerging that PECAM-1 is intimately involved in these processes, therefore, the potential exists for further investigation with direct clinical relevance. The current study has demonstrated that the marmoset may be a suitable animal model, with the aid of lHC, for such investigations. 6.6.2.3 Antigen Expression Variables other than antigen expression might have influenced the intensity of the fluorescent signal from within the blood vessel structure. Therefore, the method of signal- to-noise ratio used in this experiment was favoured in attempting to provide quantification. Data in Appendix 8.32 represent the application of this method to PECAM-1 expression in one section from an unloaded and a loaded molar from one animal. lnterpretation of these data using statistical analyses is not appropriate, as this 'sample' can not be seen to adequately represent the marmoset species. Nevertheless, it is interesting to note that comparison of the average signalto-noise ratios of the six blood vessels examined shows that data derived from the interradicular vessel of the control molar is notably different from those of the other vessels. The interradicular control vessel had the lowest average signal value and the second-highest noise value. The interradicular vessel of the experimental molar had the highest average noise value. These findings necessitate fudher investigation. This method has potential for future research where it could be extended to: i) examine variation of antigen expression in blood vessels of the same category within the same animal using all the available unloaded and loaded jaw sections from that animal, ii) examine such variation between animals. Data from multiple jaw sections could be placed into class intervals and assessed as to whether or not they conform to a normal 91 distribution. Parametric or non-parametric tests could then be applied to derive information regarding the influence of tooth loading on the antigen of interest. The reliability of this method would need to be viewed in the context of limitations, which would need to be accounted for where possible in the experimental design. For example, animal variability due to age and sex would need to be controlled. ldeally, the sections should be processed concurrently with the same batch of antibodies to ensure as far as is possible that the variations in the methodology do not influence results. 6.7 lmmunohistochemical Labelling for Periendothelial Cells 6.7.1 Pilot Study for Periendothelial Gell lmmunolabel It was reasoned that in light of the intention to investigate ET-1 expression in the marmoset periodontium, the ability to locate contractile periendothelial cells within the periodontium might prove useful in the future, as this vasoactive cytokine is thought to act locally at the endothelial cell/smooth muscle cell/pericyte interface. Therefore, a suitable immunolabelfor periendothelial cells was required. Skalli ef a/., (1989) located c¿-SMA, an isoform of the contractile protein actin, in vascular smooth muscle cells and pericytes. The antibody, mouse monoclonal anti-human ø-SMA, reacts with actin from smooth muscle, but does not react with actin from fibroblasts (Skalli et al., 1986). lncubation of sections of marmoset small intestine with this antibody (Maxim Biotech), followed by the secondary antibody, lgG/Cy3rM (Jackson), demonstrated labelling of perivascular cells with little background staining (Figure 20). This immunolabel was deemed suitable for possible identification of pericytes and vascular smooth muscle cells in the fixed oral tissues of the marmoset. 6.7.2 Periendothelial Cells in Dental Tissues The a-SMA antibody enabled immunolabelling of periendothelial cells within the marmoset periodontium and pulpal tissues with a similar degree of success as the JC-704 antibody had labelled endothelial cells. An interesting morphological finding in this study was that of two terminal arterioles within the labial PDL of a mandibular lateral incisor (Figures 42 and 43). Stanley (1993), in an 92 SEM corrosion cast study of the anterior oral microvasculature of the marmoset, found arterial PDL vessels ran predominantly from the apex to the cervical third of the PDL. There were considerably fewer arterial vessels than venous vessels. ln the current study, these arterioles provided the best example of how the CLSM and CoMOS software may be used to provide detailed images of o-SMA antibody-immunolabelled periendothelial cells. ln this context, this antibody has the potential to enable future research into periendothelial cell behaviour during angiogenesis within the periodontium incident to orthodontic tooth movement. With the combined use of the JC-704 and c¿-SMA antibodies, perhaps the current uncertainty about which cell type precedes the other during vascular proliferation could be addressed. 6.8 Suggest¡ons for Future Research The current study has demonstrated that reliable immunohistochemical identification of ET-1, endothelial cells and periendothelial cells is possible in the dental and paradental tissues of the marmoset. lt has also demonstrated that the CLSM and CoMOS software have the potential to enable quantification of antigen expression. lnformation regarding ET-1, PECAM-1 and cr-SMA has been collated with particular reference to biological aspects of orthodontic tooth movement. An account of the resultant implications for future research has been provided in the discussion, and a summary follows. 6,8.1 Endothelin-l 1. Research concerning ET-1 within the PDL microvasculature could initially involve identification of those vessel types that are involved in its production. The endothelial and periendothelial cell immunolabels used in the current study could be used in double-labelling IHC to this end. Triple-labelling IHC would provide a means of identification of all these elements within the same jaw section. When undertaking experiments involving multiple labels, attention to the species in which the antibodies were raised would be required so as to avoid cross-reactivity problems. Also, fluorophores with different excitation wavelengths would be required with each secondary antibody to enable observation of different immunolabelled structures. 2. lt has been reported that ET-1 possesses mitogenic properties (Haynes, 1995), therefore, the expression of ET-1 in the process of angiogenesis within the 93 periodontium during orthodontic tooth movement is a potential erea of future investigation. 6.8.2 Morpholog ical Stud ies Double-labelling IHC could be used for vascular morphological studies. For example, the JC-704 antibody could be used to identify blood vessel size and morphology, whilst the anti-o-SMA antibody could be used within the same jaw section to identify periendothelial cells associated with those blood vessels. 6.8.3 Angiogenes¡s Multiple aspects of the process of angiogenesis within the periodontium concomitant with orthodontic tooth movement could be investigated. 1. lmmunohistochemical studies using the primary antibody to PECAM-1, JC-704, in combination with an lgG/Cy3rM secondary antibody could be performed to identify areas of angiogenesis following various tooth-movement regimens. The effect of variables such as time, force magnitude and force direction could be assessed. 2. The role of PECAM-1 in angiogenesis is still unclear. However, investigations could be undertaken using the marmoset as an animal model. The local administration of agents that influence PECAM-1 function may alter the rate of tooth movement. ln particular, use of a PECAM-1 blocker has the potential to reduce tooth movement. 3. The role that pericytes and vascular smooth muscle cells play in angiogenesis could also be investigated. Double-labelling studies using antibodies to PECAM-1 and o-SMA raised in different species may enable investigation into endothelial and periendothelial cell behaviour during orthodontic tooth movement. The sequence of invasion of these cell types into hyalinised areas of the PDL might then be clarified. 4. The medical concept of "therapeutic angiogenesis" as introduced by Höckel ef a/. (1993) also has implications for clinical orthodontics. Angiogenic cytokines such as TGFp and VEGF are currently undergoing laboratory testing for their potential use in medical hypovascular disorders that lead to ulceration, delayed healing and aseptic necrosis (Höckel et al., 1993). As tooth movement is essentially halted by the inadvertent formation of the avascular (hyalinised) areas in the PDL, such therapies might be applicable to clinical orthodontics, particularly in adults, where tooth movement can be significantly slower than in younger patients. These therapies also have potential applications in oral medicine (delayed healing associated with diabetes 94 mellitus, previous radiotherapy, oral infections) and oral surgery (bone fracture healing). Development of these ideas in dentistry requires laboratory testing, possibly with the aid of IHC as used in the current study. 95 Chapter 7 CONGLUSIONS 1. Tissue fixation of marmosets (Callithrix jacchus) using Zamboni's fixative by immersion of tissue samples without perfusion is satisfactory for immunohistochemical experiments. 2. The rabbit anti-ET-1 antibody (Peninsula) is a suitable immunohistochemical marker for ET-1 when used in combination with the secondary antibody, donkey anti-rabbit lgG conjugated with CyStt (Jackson) in Zamboni's-fixed tissues from marmosets. 3. Endothelin-1 is present within the endothelium of the PDL microvasculature of marmosets in both unloaded and loaded conditions. The hypothesis that vascular ET-1 expression varies between the unloaded and the loaded PDL of the marmoset can be neither accepted nor rejected, as objective quantification was not achieved in this investigation. The laboratory technique for immunolabelling ET-1 established in the present study provides the opportunity for further investigation of its production within the MVB of the PDL under various conditions. All other aims of the investigation were satisfied. 4. The mouse monoclonal anti-human antibody to the PECAM-1 antigen, JC-704 (Dako), is a suitable immunohistochemical cell-surface marker for endothelial cells in Zamboni's-fixed tissues from marmosets. This antibody may be labelled with an anti-mouse lgG secondary antibody conjugated with Cylrtu (Jackson). 5. The mouse monoclonal anti-human o¿-SMA antibody (Maxim Biotech) is a suitable immunohistochemical marker for periendothelial cells in Zamboni's-fixed tissues from marmosets. This antibody may also be labelled with an anti-mouse lgG secondary antibody conjugated with Cy3rM(Jackson). 6. The model of antigen quantification presented in this investigation using signal-to- noise ratios from ClSM-generated images shows promise for future research. 96 7. The present laboratory investigation provides a basis for further immunohistochemical research into angiogenesis in the paradental tissues incident to orthodontic tooth movement. Such research might investigate the possible roles of ET-1, PECAM-1, endothelium and periendothelial cells in angiogenesis. 8. Detailed microvascular morphology might also be further investigated. Assessment of the overall distribution of immunolabelled vascular structures is most easily and inexpensively achieved using the fluorescence microscope. L The CLSM and CoMOS computer software is required for construction of detailed, three-dimensional images of PDL vascular structures located within thick, intact jaw sections by stacking optical sections. 97 Chapter I APPENDICES 8.1 Anaesthetic Pitman-Moore Australia Ltd. North Ryde, NSW, Australia Saffan is a steroid anaesthetic, which can be used in cats and monkeys. The active constituents are two pregnanedione derivatives, alphaloxone (3a -hydroxy-5a-pregnane- 11, 2O-dione) and alphadolone acetate (21-acetoxy-3a-hydroxy-Sa-pregnane-11, 20- dione), which are solubalised in saline by 2Oo/o w/v polyoxyethylated castor oil. Saffan enables sedative control of monkeys for handling or surgical anaesthesia. lt may be administered by intra-venous or intra-muscular routes. ln the monkey, intra-muscular administration is favoured due to handling difficulties often encountered with the species. The recommended sites of injection are the triceps or the quadriceps muscle groups. Deep intra-muscular injection is recommended to minimise the variability of response that may occur when Saffan is administered by this route. The onset, duration and depth of sedation are directly related to the dose given. Sedation to enable handling and examination is achieved using a dose rate of 12mg (1ml) per kg. lnduction and maintenance of surgical anaesthesia requires a dose rate of 12to 18mg perkg. Saffan is reported to have a high therapeutic index. lt produces good relaxation of the abdominal muscles and respiration is usually well maintained. lt is also well tolerated by the tissues (Pitman-Moore). Description: Clear,slightlyviscoussolution. Presentation: l0mlvials Storage: At room temperature (below 25"C). Refrigeration is contra-indicated as this may cause precipitation of the steroid constituents. Protection from light essential. Open ampoules should be discarded, as it does not contain an anti-bacterial agent. S.2Zamboni's Fixative (Stefanini et al., 1967) Consists of 2o/o formaldehyde and 15% picric acid in 0.1M phosphate buffer at pH 7.0 Preparation 2 solutions are made up and kept in fridge as stock solutions: 98 Part A: 31.29m NaHzPO¿.2HzO in l ltr distilled water (.02M). Part B: 28.399m NazHPO¿, or 35.599 NazHPO¿.2H2O, in 1 ltr distilled water (.02M). ln a lltr container add: i) 200mls part A ii) 300mls part B iii) 200mls 10% paraformaldehyde solution iv) 150mls saturated picric acid filtered through No 1 Whatman filter paper Make up to l ltr with distilled water. Storage At 4"C in refrigerator. 8.3 Phosphate Buffered Saline (PBS) Preparation: To make l ltr working solution; add to l ltr of double distilled water: i) 8.59m NaCl ii) 1.079m NazHPO¿, or 1.349m Na2HPO4.2H2O iii) 0.3459m NaHzPO¿.HeO, or 0.399m NaHzPO¿.2HzO Adjust pHlo7.2-7.4 using 1M NaOH or HCl. 8.4 PBS I 30% Sucrose I 0.1% Sodium Azide Preparation: To make l ltr working solution; add to l ltr of double distilled water: i) 8.59m NaCl ii) 1.079m NazHPO¿, or 1.349m NazHPO¿.2HzO iii) 0.3459m NaHzPO+.H2O, or 0.399m NaHzPO¿.2HzO iv) 3009m Sucrose v) 10mls 10% sodium azide 8.5 Dimethyl Sulphoxide (DMSO) Merck Pty. Ltd., Kilsyth, Victoria, Australia. Description: Practically colourless and odourless liquid. Used to clear the cellular cytoplasmic membranes and expose intracellular antigens for immunolabelling. Storage: ln sealed bottle at room temperature. Toxicity: Skin contact causes primary irritation with redness, itching, and sometimes scaling. 99 8.6 Coating Slides To atford better adhesion of the tissues to the slides, the slides are coated with 0.01% poly-L-ornithine (Sigma BioSciencesrM, St Louis, Missouri, USA.) Procedure: i) Wash slides in warm Decon 90 for t hour. ii) Rinse in running tap water for 2-3 hours. iii) Soak in double distilled water 2 X 3Omins. iv) Shake off excess water and dip slides in poly-L-ornithine at room temperature for t hour. v) Drain slides and dry at room temperature overnight, or at 37'C for rapid drying. vi) Coated slides are stored at room temperature and used within 2 weeks if possible. 8.7 Drying Slides w¡th Tissue Sections After the tissue sections have been placed on the slides, they are dried before being stained. They are placed in a vacuum dessicator containing phosphorus pentoxide (PrOs) for 30 mins. The slides are then ready for processing for immunohistochemistry, or may be stored in lightproof plastic boxes at 4'C for up to 2 weeks. Prior to processing, the slides are removed from the fridge and allowed to reach room temperature before being removed from the box in order to reduce the effects of condensation on the tissues. Slides that have not been used may be again stored in the fridge. 8.8 Antibody Diluent (0.02M) Description: Hypertonic PBS with 0.01o/o sodium azide. The NaCl concentration helps to reduce the non-specific binding of antibodies to the tissue. Preparation: To make l ltr solution; add to 1 ltr double distilled water: i) 17gm NaCl ii) 1.079m NazHPO¿, or 1.349m NazHPO¿.2HzO iii) 0,3459m NaHzPO¿.H2O, or 0.399m NaHzPO¿.2HzO iv) 1ml10o/o sodium azide v) Adjust PH = 7.1 100 8.9 Donkey Normal Serum Jackson lmmunoResearch Laboratories, lnc., West Grove, Pennsylvania, USA. (Product information) Description. A clear, straw coloured liquid obtained from a normal donor herd. Protein concentration of 60.7m9/ml. Used as a normal blocking agent or negative control in immunoassays. Storage: For extended storage, is frozen in working aliquots. Othenruise stored at 0-5'c. 8.10 RECA-1 Antibody Serotec: purchased via Jomar Diagnostics, Magill, South Australia. (Product information) Specificity: Pan endothelium (RECA-1 antigen) Description: Clear liquid containing 0.2M Tris/HCl, pH 7.4, and 8% foetal calf serum. lmmunogen is stromal cells from rat lymph node. Mouse host. Storage: Store at 4"C for 1 month, or at -20"C for longer. Avoid repeated freezing and thawing, as this can denature the antibody. 8.1 I Rabbit Anti-Endothelin-1 Antibody Peninsula Laboratories lnc., San Carlos, Cal., USA. (Product information) Specificity: lQOo/o cross-reactivity with ET-1 (porcine, human) 91% cross-reactivity with ET-2 (human) 100% cross-reactivity with Big ET-1 (porcine) 76% cross-reactivity with Big ET-1 (human) Description: Clear liquid diluted with equal portions of PBS and glycerol Rabbit antiserum to Endothelin-1 (porcine, human). Storage: Aliquoted and stored as stock solution al -20"C. 8.12 Rabbit Anti-Big Endothelin-1 Antibody Peninsula Laboratories lnc., San Carlos, Cal., USA. (Product information) Specificity: lQQo/o cross-reactivity with Big ET-1 (rat) 1O0o/o cross-reactivity with ET-1 (human, porcine, canine, rat, mouse, bovine) 35% cross-reactivity with Big ET-1 (porcine) 17% cross-reactivity with Big ET-1 (human) Description Clear liquid diluted with equal portions of PBS and glycerol antiserum to Big Endothelin-1 (rat). Storage Aliquoted and stored as stock solution at -20'C. 8.13 Rabbit Anti-Pre-proendothelin Antibody Peninsula Laboratories lnc., San Carlos, Cal., USA. (Product information) Specificity: No cross-reactivity with ET-1 , ET-2 or Big ET-1. Description. Clear liquid diluted with equal portions of PBS and glycerol. Rabbit antiserum to Pre-pro Endothelin-1 (Porcine). Storage: Aliquoted and stored as stock solution at -20"C. 8.14 Monoclonal Anti-Endothelin Antibodies Yamasa: purchased via Sapphire Bioscience Ltd. Alexandria, NSW. (Product information) Specificity: 1. Mouse antiserum to Nterminal of endothelin. 2. Mouse antiserum to C-terminal of endothelin. Description: Clear liquids diluted with equal portions of PBS and glycerol. Storage: Aliquoted and stored as stock solution at -20'C. 8.15 Endothelial Gell Marker: QB-END 10 Antibody Novocastra Laboratories Ltd., Newcastle upon Tyne, UK. (Product information) Synonym: CD34 Specificity: Human endothelial cells, some cross reactivity with basement membrane collagen. Description: Clear liquid. Raised against CD34 antigen in human endothelial cell membranes and haemopoietic progenitor cells. Mouse host. Storage: Store at 4"C in desiccated and reconstituted form. For longterm storage aliquot antibody into smaller volumes and store at -20"C. 8.16 Endothelial Cell Marker: Anti- PECAM-î Antibody Dako (Australia) Pty. Ltd. Botany, NSW, Australia. (Product information) Synonym: JC-704 Specificity: Strong reactivity with formalin-resistant epitope on CD31 in endothelial cells in normal tissues and in benign and malignant proliferations. Also labels megakaryocytes, platelets and occasionally plasma cells. 102 Description Clear liquid. Raised against CD-31 antigen. Mouse host Storage: ln liquid form at 2-8"C until expiry date. 8.17 Mouse Monoclonal Anti-Human Alpha-Smooth Muscle Actin Antibody Maxim Biotech, Inc. San Francisco, California. (Product information) Specificity: Reacts with normal and neoplastic smooth muscle. Does not react with actin from fibroblasts (beta- and gamma-cytoplasmic), striated muscle (alpha-sarcomeric) and myocardium (alpha-myocardial). ls not reactive with other mesenchymal cells or all epithelial cells, except myoepithelial cells. Description: Clear liquid. Mouse host. Storage: Store at 4-8'C; do not freeze. Stable for 1 year. 8.18 Gytrna Gonjugated Donkey Anti-Mouse lgG Antibody Jackson lmmunoResearch Laboratories, lnc,, West Grove, Pennsylvania, USA. (Product information) Description: Freeze-dried powder re-constituted with distilled water. Antibody concentration of 1.4m9/ml. Suggested dilution range: 1:100-1:800 for most applications. Reacts with heavy chains on mouse lgG and with light chains common to most mouse immunoglobulins. Filter: Green Storage: Freeze-dried product stored at 4' until opened. Restored to 0.4m1 with distilled water, stable at 4"C for several weeks. For extended storage (up to one year) after reconstitution; aliquot and freeze at -20'C or below. lt is preferable to add an equal volume of glycerol for long term storage at -20'C, but protein concentration will be halved. Expiration date: one year from date of reconstitution. Ll9 Gy$rru Gonjugated Donkey Anti-Rabbit lgG Antibody Jackson lmmunoResearch Laboratories, lnc., West Grove, Pennsylvania, USA. (Product information). Description: Freeze-dried powder re-constituted with distilled water. Antibody concentration of 1.5m9/ml. Suggested dilution range: 1:100-1:800 for 103 most applications. Reacts with heavy chains on rabbit lgG and with light chains common to most rabbit immunoglobulins. Filter: Green Storage Freeze-dried product stored at 2-8'C until opened. Restored to 0.4m1 with distilled water, stable at 2-8"C for several weeks. For extended storage (up to one year) after reconstitution; aliquot and treeze at -20"C or below. lt is preferable to add an equal volume of glycerol for long term storage at -20"C, but protein concentration will be halved. Expiration date: one year from date of reconstitution. 8.20 Cy5ru Gonjugated Donkey Anti-Rabbit lgG Antibody Jackson lmmunoResearch Laboratories, lnc., West Grove, Pennsylvania, USA. (Product information). Description: Freeze-dried powder re-constituted with distilled water. Suggested dilution range: 1:100-1:800 for most applications. Reacts with heavy chains on rabbit lgG and with light chains common to most rabbit immunoglobulins. Filter: Red Storage: Freeze-dried product stored at 2-8"C until opened. Restored to 0.4m1 with distilled water, stable at 2-8'C for several weeks. For extended storage (up to one year) after reconstitution; aliquot and lreeze at -20"C or below. lt is preferable to add an equal volume of glycerol for long term storage at -20'C, but protein concentration will be halved. Expiration date: one year from date of reconstitution. 8.21 Dichlorotriazinyl Amino Fluorescein (DTAF) Antibody Jackson lmmunoResearch Laboratories, lnc., West Grove, Pennsylvania, USA. (Product information). Description: DTAF (fluorophore)-conjugated affinipure donkey anti-rabbit lgG. Freeze-dried powder re-constituted with distilled water. Antibody concentration of 1.5m9/ml. Suggested dilution range 1:50 -1:200 for most applications. Reacts with the heavy chains on rabbit lgG and with light chains common to most rabbit immunoglobulins. Filter: Blue Storage: Freeze-dried product stored at 4"C until opened. 104 Re-constituted; is stable for several weeks al4"C. For extended storage (up to one year); is added to equal volume of glycerol at -20'C, with or without aliquoting. Once diluted, should be used on the day. 8.22 Biotin-SP-Gonjugated Affinipure Donkey Anti-Rabbit lgG Jackson lmmunoResearch Laboratories, lnc., West Grove, Pennsylvania, USA. (Product information) Description: Freeze-dried powderre-constitutedwithdistilledwater. Antibody concentration of 1 .1mg/ml. Suggested dilution range: 1:50, 000 - 1'.1,000, 000 for ELSA using enzyme-conjugated streptavidin. Reacts with heavy chains on rabbit lgG and with light chains common to most rabbit immunoglobulins. Storage: Freeze-dried product stored at 2-8'C until opened. Restored to 0.5m1 with distilled water, stable at 2-8'C for several weeks. For extended storage (up to one year) after reconstitution; aliquot and freeze at -20'C or below. lt is preferable to add an equal volume of glycerol for long term storage at -20'C, but protein concentration will be halved. Expiration date: one year from date of reconstitution. 8.23 Cy3rtu Conjugated Streptavidin Jackson lmmunoResearch Laboratories, lnc., West Grove, Pennsylvania, USA. (Product information) Description: Freeze-dried powder re-constituted with distilled water. Streptavidin concentration 1.8m9/ml. Suggested final working dilution 0.5-2pg/ml for most applications. Filter: Green Storage: Freeze-dried product stored at 2-8"C until opened. Restored to 0.65m1 with distilled water, stable at 2-8'C for several weeks. For extended storage (up to one year) after reconstitution; aliquot and freeze at -20'C or below. lt is preferable to add an equal volume of glycerol for long term storage at -20"C, but protein concentration will be halved. Expiration date: one year from date of reconstitution. 105 8.24 Immunohistochemical Labelling Procedure - Soft Tissues. Indirect Technique 1. Obtain tissue samples and place in Zamboni's fixative at 4oC Íor 24 hours. 2. Remove picric acid with graded ethanol solutions (70%,80o/o,99o/o and 100%), 5 mins each. 3. Clear in DMSO 3 X 1Omin. 4. Wash in PBS 3 X 10min. 5. Store in PBS/30% sucrose + 0.01% sodium azide at 4oC overnight on a bench shaker. 6. Prepare 14pm thick sections on cryostat and place on poly-L-ornithine coated slides. 7. Dry slides over phosphorus pentoxide in vacuum dessicator for 3Omins. 8. Block with 10% DNS diluted in antibody diluent for 3Omins. 9. lncubate in primary antibody and antibody diluent containing 10% DNS in humid box overnight at room temperature. 10. Drain antibody off slides and rinse by pipetting 10o/o DNS diluted in antibody diluent ontosections3XlOmin. ll.lncubate with secondary antibody and 10% DNS diluted in antibody diluent in humid box for t hour at room temperature. Cover to prevent entry of ambient light. 12. Drain antibody off slides and rinse by pipetting 0.02M PBS onto sections 3 X 10min. 13. Dry slides with high-grade blotting paper and mount in 20o/o buffered glycerol, pH 8.6. 14. View sections. Store slides in lightproof container. 8.25 lmmunohistochemical Labelling Procedure - Soft Tissues. Streptavid i n -Bioti n Am pl ification Tech n iq ue L Proceed as for the indirect technique until reaching the secondary antibody step. 2. Apply the biotin antibody appropriate for the primary antibody at 1:50 for 2 hours at room temperature in humid box. For example, for a mouse primary antibody, apply donkey anti-mouse biotin. 3. Drain biotin off slides and rinse by pipetting 10% DNS diluted in antibody diluent onto sections 3 X 10min. 4. lncubate in streptavidin conjugated with a fluorophore in humid box for 1.5 hours. 5. Wash and mount sections in same manner as above. 106 8.26 lmmunohistochemical Labelling Procedure Mandibular Sections. lndirect Technique 1. Obtain jaws ensuring mucoperiosteum has been lifted from bone and place in Zamboni's fixative at 4oC for 24 hours. 2. Remove soft tissues and wash whole jaws in 0.01M PBS 3 X 1Omin. 3, Store in PBS/30% sucrose + 0.01% sodium azide at 4oC. 4. Hemisection mandible and mount bodies on individual perspex blocks using LoctiterM. 5. Cut thin sagittal sections with a low speed bone sectioning saw using 0.01M PBS as a lubricant, 6. Remove picric acid with graded ethanol solutions (70o/o, 80o/o, 90o/o and 100%), 5 mins each. 7 . Clear in 100% DMSO 3 X 2Omins. 8. Wash in 0.01M PBS 3X10min and store in PBS/30% sucrose and 0.01% sodium azide at 4oC until starting immunohistochemistry. 9. Block in 10o/o DNS diluted in antibody diluent for 2 hours on bench shaker, 10. lncubate in primary antibody and antibody diluent containing 10% DNS on bench shaker for 4 days. 11. Wash in 10o/o DNS diluted in antibody diluent 3 X 30mins. 12. lncubate in secondary antibody and antibody diluent containing 10% DNS on bench shaker for 2 days. Cover vials with aluminium foil. 13. Wash in antibody diluent 3 X 30mins. 14. Mount with large coverslips on each side in 20o/o buÍfered glycerol, pH 8.6. Seal with nail varnish. 15. View sections and store in lightproof container. 8.27 Chequerboard for Pilot Study No I Slide numbers were used to identify dilutions of primary and secondary antibody applied to tissue sections. ilsG / cy3l BLANK 1 in 100 fRECA-11 BI ANK 12 34 1in4 56 1in8 78 1in16 9lo 1in32 11 12 107 8.28 Ghequerboards for Pilot Study No 2 Slide numbers were used to identify dilutions of primary and secondary antibody applied to tissue sections. a. / [gG BLANK 1in25 1in50 1 in 100 1 in 200 DTAFI fAnti-ET-lt BLANK 12 34 56 7A 910 1 in IOO 11 12 13 14 15 16 17 l8 1 in 2OO l9 20 21 ?? 23 24 25 26 I in 400 27 28 29 30 31 32 33 ?A I in 800 35 36 37 38 39 40 4'l 42 b. c / flgG / [sG BLANK 1 in 100 cv3l BLANK I in 100 cv3l [Anti-ET-1 or lAnti-ET-lt Anti-Big ET-ll BI ANK 12 34 BI ANK 12 34 1in25 56 1in25 56 l in5O 78 1in50 78 1 in 100 9lo I in 100 910 [AntiBig ET-l] 1in50 11 1 in 100 12 d. [sG / cv3l BLANK 1 in 100 fAnti-Prepro ET-ll BLANK 12 34 1in25 56 1in50 78 1 in l0O 910 ET-ll [Anti-Big 11 12 1in50 e. ilgG / cv3l BLANK 1 in 100 lAntiET-lt BLANK 12 34 1in25 - 56 N Terminal 78 C Terminal 1in50- I 10 N Terminal 1 1 12 C Terminal 1 in 100 - 1 3 4 N Terminal I 5 6 C Terminal 108 f. flsG / cv3l BLANK 1 in 100 fAntiET-11 BI ANK 12 34 Streptavidin 1in25 - 1:200: 5 6 1:400: 7 I Streptavidin 1in50- 1:200: I 10 1:400: 11 12 Streptavidin 1in 100- 'l:200: 13 14 1:400: 15 16 8.29 Chequerboard for Pilot Study No 3 Slide numbers were used to identify dilutions of primary and secondary antibody applied to tissue sections. UgG / cv3l BLANK 1 in 100 ]QB.END IOI BI ANK 12 34 1in4 56 I in8 78 lin16 910 1in32 11 12 8.30 Ghequerboard for Pilot Study No 4 Slide numbers were used to identify dilutions of primary and secondary antibody applied to tissue sections. ilgG / cv3l BLANK 1 in 100 fJc-704t BLANK 1? 34 lin10 56 lin20 78 1in40 910 1in80 11 '12 8.31 Ghequerboard for Pilot Study No 5 Slide numbers were used to identify dilutions of primary and secondary antibody applied to tissue sections. flgG / cv3l BLANK 1 in 100 tAnti- o-SMAI BLANK 1? 34 I in 1O0 56 1 in 200 78 109 8.32. Quantification of PECAM-I Expression in PDL Vessels in Unloaded and Loaded First Molars. IINI OADED MOLAR MESIAL ROOT APEX INTFFIRÀÍIIÍ:III ÄR ÀRFA SIGNAI NôISF SIGNÂI NOISF SIGNÂI NOISF 101 4 107 6 108 18 60 I 94 5 127 14 132 I 105 6 100 17 136 9 78 3 74 I 83 7 106 6 110 10 90 10 98 6 78 11 153 10 56 5 143 14 39 I 101 6 78 4 88 7 80 7 82 7 71 7 116 5 63 4 163 12 88 4 62 6 135 7 106 5 53 I 46 5 100 4 103 13 61 5 78 4 56 11 148 6 72 3 74 17 130 5 65 't2 74 22 128 6 110 7 88 29 70 7 114 7 70 12 33 5 98 7 74 20 47 6 84 7 8l 10 TôTÂI 1914 142 lR56 .t.t^ I t^ ÂVFRAGF ^qR 957 71 q?R R¿q 1)A SIGNAI - TO - NOISE^74 RATIO 135 1^1 A6 I OÂTìFD MOLAR MFSIÀI ROOT APEX DISTAL ROOT APEX INTFPFIAÍìIíìIII ÂP ÀRF srêNÂl NôISF SIGNAI NOISF SIGNÂI NOISE 166 12 93 10 220 16 159 10 136 6 219 12 154 11 122 6 189 12 169 7 104 11 197 9 167 7 123 12 196 10 185 6 128 I 190 16 184 5 157 18 223 10 177 I 156 11 178 8 180 6 154 10 211 13 156 6 113 12 206 15 188 4 136 I 164 11 150 4 111 I 203 19 188 23 109 7 203 7 191 13 114 b 172 10 162 5 170 10 200 13 179 4 166 7 177 21 130 13 182 14 212 14 139 7 166 11 202 15 147 15 161 9 200 16 194 1? 1)5 7 202 14 TOTAL 11^5 17F- 27)â 192 3964 261 AVERAGE 1âA 25 Âq 136 3 g6 198 2 l? fìà SIGNAL - TO - NOISE RATIO lßq 142 152 110 Chapter 9 REFERENCES Abassi Z. A., Tate J. E., Golomb E. and Keiser H. (1992) Role of neutral endopeptidase in the metabolism of endotheli n. Hype fte nsio n. 20: 89-95. Bartholomeuz B., Hardy K. J., Nelson A. S. and Phillips P. A. (1996) Bosentan ameliorates cyclosporin A-induced hypertension in rats and primates. Hypeftension.2T:1341-1345. Bassett C. A. L.. and Becker R. O. (1962) Generation of electrical potentials by bone in response to mechanicalstress. Science 137: 1063-1064. Berkovitz B. K. B. (1990) The structure of the periodontal ligament: an update. Eur J Oñhod. 12 51-76. Berkovitz B. K. 8., Moxham B. J. and Newman H. N. (1995) The Periodontal Ligament in Health and Disease. 2nd edn. Mosby-Wolfe, London. Bernick S. and Levy B. M. (1968a) Studies on the biology of the periodontium of marmosets: l. Development of bifurcation in multirooted teeth in marmosets (Callithrix iacchus). J Dent Res. 47: 21-33. Bernick S. and Levy B. M. (1968b) Studies on the biology of the periodontium of marmosets: lV. lnnervation of the periodontal ligamenl. J Dent Res.47: 1158-1165. Bernick S., Levy B. M., Dreizen S. and Grant D. A. (1977) The intraosseous orientation of the alveolar component of marmoset alveolodental fibers. J Dent Res. 56: 1409-1416. Bien S. M. (1966) Hydrodynamic damping of tooth movement. J Dent Res.45: 907-914. Blood C. H. and Zelter B. R. (1990) Tumor interactions with the vasculature: angiogenesis and tumour metastasis. Biochim Biophys Acta. 1032= 89-'1 18. Brudvik P. and Rygh P. (1993a) The initial phase of orthodontic root resorption incident to local compression of the periodontal ligament. Eur J Orthod. 15:. 249-263. Brudvik P. and Rygh P. (1993b) Non clast cells start orthodontic root resorption in the periphery of hyalinized zones. Eur J Orthod.15:467-480. Brudvik P. and Rygh P. (1994) Root resorption beneath the main hyalinized zone. Eur J O¡Ihod. 16:249-263. Brudvik P. and Rygh P. (1995) The repair of orthodontic root resorption: an ultrastructural study. Eur J Oñhod. l7: 189-198. Cameron J. (1995) Endothelial junctions in the periodontal ligament microvasculature of young and aged mice. MDS Thesis, The University of Adelaide, South Australia. Casasco 4., Calligaro A., Casasco M., Springall D. R., Tenti P., Marchetti C., Poggi P. and Polak J. M. (1991) lmmunohistochemical localization of endothelin-like immunoreactivity in human tooth germ and mature dental pulp. Anaf Embryol.183: 515-520. 111 Casasco 4., Casasco M., Ciuffreda M., Springall D. R., Calligaro 4., Bianchi S. and Polak J. M. (1992) lmmunohistochemical evidence for the occurrence of endothelin in the vascular endothelium of normal and inflamed human dental pulp. J Denf Res. 71= 475-477. Castelli W. A. and Dempster W. T. (1965) The periodontal vasculature and its responses to experimental pressures. J Am Dent Assn. 70: 890-905. Chakravarthy U., GardinerT. A., Anderson P., Archer D. B. and Trimble E. R. (1992) The effectof endothelin 1 on the retinal microvascular pericyte. Microvasc Res. 43: 241-254. Chintavalakorn S. (1993) TEM stereology of blood vessels and nerves in marmoset periodontal ligament following incisor crown fracture, endodontic therapy, orthodontic extrusion and long term retention. MDS Thesis, The University of Adelaide, South Australia. Christenson L. K. and Stouffer R. L. (1996) Proliferation of microvascular endothelial cells in the primate corpus luteum during the menstrual cycle and simulated early pregnancy. Endocrinology 137 : 367 -37 4. Chumbley A. B. and Tuncay O. C. (1986) The effect of indomethacin (an asprin-like drug) on the rate of orthodontic tooth movement. Am J Oñhod.89l.312-314. Clark A. (1986) A quantitative analysis of the effect of intrusion on apical periodontal ligament fenestrae of the rat maxillary molar. MDS Thesis, The University of Adelaide, South Australia. Clozel M. and ClozelJ. P. (1989) Effects of endothelin on regional blood flows in squirrel monkeys J Pharmacol Exp Ther. 250l. 1125-1131. Cogoli A., Valluchi-Morf M., Mueller M. and Briegleb W. (1980) Effect of hypogravity on human lymphocyte activation. Aviat Space Environ Med. 51= 29. Cohn S. A. (1972) A re-examination of Sharpey's fibres in alveolar bone of the marmoset. Archs Oral Biol. 17:261-269. Crowe P. (1988) A TEM stereological analysis of the marmoset periodontal ligament following incisor crown fracture, root canal therapy and orthodontic extrusion. MDS Thesis, The University of Adelaide, South Australia. DavenportA.P.andMaguireJ.J.(1994) lsendothelin-inducedvasoconstrictionmediatedonlyby ETa receptors in human s? Trends Pharmacol Scr. 15: 9-1 I . Davidovitch Z. (1991a) The state of basic research. ln: Orthodontics af fhe Crossroads. Trends in Contemporary Ofthodontics. (Eds. Ghafari J. G. and Moorrees J. G.) pp121- 157. Superior Sale, Toronto. Davidovitch Z. (1991b) Tooth movement. Crit Rev Oral Biol Med.2:411-450. Davidovitch Z. (1995) Cell biology associated with orthodontic tooth movement. ln: The Periodontat Ligament in Health and Disease. (Eds. Berkovitz B. K. 8., Moxham B. J. and Newman H. N.)2nd edn, pp259-278. Mosby-Wolfe, London. Davidovitch 2., Nicolay O. F., Ngan P. W. and Shanfeld J. L. (1988) Neurotransmitters, cytokines and the control of alveolar bone remodelling in orthodontics. Dent Clin North Am. 32l. 411-435. 112 DeLisser H. M., Christofidou-Solomidou M., Strieter R. M., Burdick M. D., Robinson C. S., Wexler R. S., Kerr J. S., Garlanda C., Merwin J. R., Madri J. A. and Albelda S. M. (1997) lnvolvement of endothelial PECAM-1/CD31 in angiogenesis. Am J Pathol. 151l. 671-677. Delisser H. M., Newman P. J. and Albelda S. M. (1994) Molecular and functional aspects of PECAM-1/CD31. lmmunol Today. 15: 490-495. Dinarello C. A. (1988) Biology of interleukin-1 . FASEB J. 2= 108-1 I 5. Douvartzidis l. (1983) A morphometric examination of the periodontal ligament vasculature of the marmoset mandible. MDS Thesis, The University of Adelaide, South Australia. Dreizen S., Levy B. M., Bernick S., Hampton J. K. Jr. and Kraintz L. (1967) Studies on the biology of the periodontium of marmosets: lll. Periodontal bone changes in marmosets with osteomalacia and hyperparathyroidism. J Med Scl. 3: 731-738. Duijvestijn A. M., van Goor H., Klatter F., Majoor G. D., van Bussel E. and van Breda Vriesman P. J. C. (f992) Antibodies defining rat endothelial cells: RECA-1, a pan-endothelial cell- specific monoclonal antibody. Lab lnvest.66: 459-466. Dupuis J., Stewart D. J., Cernacek P. and Gosselin G. (1996) Human pulmonary circulation is an important site for both clearance and production of endothelin-1. Circulation 94= 1578- 1 584. Dzau V. J., Pratt R. E. and Cooke J. P. (1992) Endothelin as a growth factor in vascular remodeling and vascular disease. ln: Endothelin. (Ed Rubanyi G M.) pp137-143. Oxford University Press, Oxford. Edwall L. G. A. and Aars H. (1995) The physiology of the vasculature of the periodontal ligament. ln: The Periodontal Ligament in Health and Disease. (Eds. Berkovitz B. K. 8., Moxham B. J. and Newman H. N.)2nd edn, pp121-131. Mosby-Wolfe, London. Ehrenreich H., Anderson R. W., Fox C. H., Riekmann P., Hoffman G. S., Travis W. D., Coligan J. E., Kehrl J. H. and Fauci A. S. (1990) Endothelins, peptides with potent vasoactive properties, are produced by human macrophages. J Exp Med. 172:.1741-1748. Epker B. N. and Frost H. (1965) Correlation of bone resorption and formation with the physical behaviour of loaded bone. J Denf Res. 44: 33-41. Ferreira S. H., Romitelli M. and de Mecci G. (1989) Endothelin-1 participation in overt and inflammatory pain. J Cardio Pharmacol.l3 (Suppl 5l= 5220-222. Folkman J. (1971) Tumour angiogenesis: therapeutic implications. N Engl J Med. 285l. 1182-1186. Foong K. W. C. (1993) A TEM stereological investigation of the blood vessels in human mandibular premolar periodontal ligament. MDS Thesis, The University of Adelaide, South Australia. Freeman E. (1994) Periodontium.ln: Oral histology: development, structure, and function. (Ed Ten Cate A. R.)4th edn, pp276-312. Mosby Year Book, St Louis. Freezer S. (1984) A study of periodontal ligament mesial to the mouse mandibular first molar MDS Thesis, The University of Adelaide, South Australia. 113 Freezer S. and Sims M. R. (1987) A transmission electron-microscope stereological study of the blood vessels, oxytalan fibres and nerves of the mouse periodontal ligament. Arch Oral Biol.32= 407-412. Fujita K., Matsumura Y., Kita S., Hisaki K., Takaoka M. and Morimoto S. (199aa) lntrarenal conversion of big endothelin-'1 to endothelin-1 in the rat. Life Scr. 55: 1285-1291. Fujita K., Matsumura Y., Kita S., Hisaki K., Takaoka M. and Morimoto S. (1994b) Phosphoramidon-sensitive conversion of big endothelin-1 and degradation of endothelin- 1 in rat kidney. Hypertension 24= 227-233. Garrels J. l. and Gibson W. (1976) ldentification and characterization of multiple forms of actin. Cell. 9:793-805. Gianelly A. A. (1969) Force-induced changes in the vascularity of the periodontal ligament. Am J Otihod.55: 5-11. Gilbert T. M., Pashley D. H. and Anderson R. W. (1992) Response of pulpal blood flow to intra- arterial infusion of endothelin. J Endod. '18= 228-231. Gould T., Melcher A. and Brunette D. (1977) Location of progenitor cells in periodontal ligament of mouse molar stimulated by wounding. Anat Rec. 188: 1 33-142. Goss A. N. (1984) A comparison of tooth eruption patterns between two colonies of young marmosets (callithrix jacchus). J Dent Res. 63: 44-46. Grant D A., Bernick S., Levy B. M, and Dreizen S. (1972) A comparative study of periodontal ligament development in teeth with and without predecessors in marmosets. J Periodontol. 43: 162-169. Grieve lll W. G., Johnson G. K., Moore R. N., Reinhardt R. A. and DuBois L. M. (1994) Prostaglandin E (PGE) and interleukin-1p (lL-1p) levels in gingival crevicularfluid during human orthodontic tooth movemenl. Am J Orthod Dentofac Oñhop.105: 369-374. Haralabopoulos G., Grant D., Kleinman H., Lelkes P., Papaioannou S. and Maragoudakis M. (1994) lnhibitors of basement membrane collagen synthesis prevent endothelial cell alignment in matrigel in vitro and angiogenesis ln vivo. Lab lnvest. 71: 575-582. Hayashi K. (1983) Synopsis; and fluorescence microscopy. ln lmmunofluorescence in Medical Scrence. (Eds. Kawamura A. Jr. and Aoyama Y.) pp9-16 and pp75-92. University of Tokyo Press. Haynes W. G. (1995) Endothelins as regulators of vascular tone in man. Clin Scl. 88: 509-517 Haynes W. G. and Webb D. J. (1993) The endothelin family of peptides: local hormones with diverse roles in health and disease? Clin Sci.84: 485-500. Haynes W. G., Clarke J., Cockcroft J. and Webb D. J. (1991) Pharmacology of endothelin-1 in vivo in man. J Cardiovasc Pharmacol. 17 (Suppl. 7): 5283-5285. Hino 4., Tokuyama Y., Kobayashi M., Yano M., Weir 8., Takeda J., Wang X., Bell G. l. and MacDonald R. L. (1996) lncreased expression of endothelin B receptor mRNA following subarachnoid hemorrhage in monkeys. J Cereb Blood Flow Metab. l6: 688-697. Höckel M., Schlenger K., Doctrow S., Kissel T. and Vaupel P. (1993) Therapeutic angiogenesis Arch Surg. 128l. 423-429. 114 lnoue 4., Yanagisawa M. and Kimura S. (1989) The human endothelin family: three structurally and pharmacologically distinct isopeptides predicted by three separate genes. Proc Natl Acad Sci USA. 86: 2863-2867. James W. W. (1980) The jaws and teeth of primates. Pitman Med. Pub. Co. Ltd., London. Johnson C. (1994) Exploring confocal microscopy. Lab news. August: '10 and 16. Reed Business Publishing, Chatswood. Johnston G. W., Dreizen S. and Levy B. M. (1970) Dental development in the cotton ear marmoset (callithrix jacchus). Am J Phys Anthrop. 33l. 41-48. Joyce N. C., Haire M. F. and Palade G. E. (1985) Contractile proteins in pericytes. I lmmunoperoxidase localization of tropomyosin. J Ce// Biol. 100:. 1379-1386. Kato J., Tanne K., lchikawa H., Matsuo S., Wakisaka S., Akai M., Kurisu K. and Sakuda M. (1992) Distribution of calcitonin gene-related peptide and substance P-immunoreactive nerve fibers and their correlation in the periodontal ligament of the mouse incisor. Acta Anat. 145:101-105. Kuwabara T. and Cogan D. G. (1963) Retinal vascular patterns. Vl. Mural cells of the retinal capillaries. Arch Opthalmol. 69: 492-502. Kvinnsland S., Heyeraas K. and Øljord E. S. (1989) Effect of experimental tooth movement on periodontal and pulpal blood flow. Eur J Orthod. 11= 200-205. Lansman J. 8., Hallam T. J. and Rink T. J. (1987) Single stretch-activated ion channels in vascular endothelial cells as mechanotransducers. Nature 325: 811-813. LeBeux Y. J. and Willemot J. (1978) Actin- and myosin-like filaments in rat brain pericytes. Anaf Rec. 190: 811-826. Lee D. (1987) A scanning electron microscopic study of the normal marmoset periodontal ligament microvascular bed and the vascular response to pericision using corrosion casts. MDS Thesis, The University of Adelaide, South Australia. Leung D. Y. M., Glagov S. and Mathews M. B. (1976) Cyclic stretching stimulates synthesis of matrix components by arterial smooth muscle cells rn vitro. Science. '191= 475-477. Levy B. M. (1971) The nonhuman primate as an analogue for the study of periodontal disease. J Dent Res. 50= 246-253. Levy B. M. (1976) Animal model: chronic destructive periodontitis in marmosets. Am J Path.83 637-640. Levy B. M. and Bernick S. (1968a) Studies on the biology of the periodontium of marmosets: ll. Development and organization of the periodontal ligament of deciduous teeth in marmosets (callithrix jacchus). J Dent Res. 47: 27-33. Levy B. M. and Bernick S. (1968b) Studies on the biology of the periodontium of marmosets: V Lymphatic vessels of the periodontal ligament. J Dent Res. 47: 1166-1170. Levy B. M., Dreizen S. and Bernick S. (1972) Effectof aging on the marmoset periodontium. J Oral Path.'l: 61-65. 115 Levy B. M., Dreizen S., Bernick S. and Grant D. (1970) Comparative study of the periodontium of aged marmosets and humans. J Dent Res. 49 (special issue): No 604. 199. Lew K., Sims M. R. and Leppard P. l. (1989) Tooth extrusion effects on microvessel volumes, endothelial areas, and fenestrae in molar apical periodontal ligament. Am J Orthod Dentofac Oñhop. 96l. 221 -231. Lewis P. J. (1996) A model for the response of blood vessels to tensile forces. Masters Thesis. University of Sydney, New South Wales. Lilja E., Björnestedt T. and Lindskog S. (1983) Cellular enzyme activity associated with tissue degradation following orthodontic tooth movement in man. Scand J Dent Res. 91:381- 390. Lopez J.4., Armstrong M. L., Piegors D. J. and Heistad D. D. (1990) Vascular responses to endothelin-1 in atherosclerotic primates. Afteriosclerosis. '10l. 1113-1 1 18. Lüscher T. F., Rubanyi G. M., Masaki T., Vane J. R. and Vanhoutte P. M. (1993) Endothelial control of vascular tone and growth. Supp Circulation. 87:. V-1-V-1. Marchesi V. T. and Florey H. W. (1960) Electron micrographic observations on the emigration of leukocytes. J Exp Physiol. 45= 343-347. Matsumoto Y., lshibashiT., Niiya 4., Yamada H,, Kurokawa K. and Kinoshita S. (1997) Distribution of endothelin and endothelin-A receptor in the lacrimal glands of the monkey (macaca fuscata). Exp Eye Res.64: 127-132. Matsuo M., Kishi Y. and Takahashi K. (1987) The periodontal vascular changes and bone resorption incident to experimental teeth movement using corrosive resin cast with SEM. J Jpn Orthod Soc. 46:.217-229. McCulloch C. and MelcherA. (1983) Cell density and cellgeneration in the periodontal ligamentof mice. Amer J Anat.167: 43-58. Mclntosh G. H. and Looker J. W. (1982) Development of a marmoset colony in Australia. Lab An Sci. 32:677-679. MelcherA. H. (1989) An overview of the anatomy and physiology of the periodontal ligament. ln: The Biology of Tooth Movement. (Eds. Norton L. A. and Burstone C. J.) ppl-7. CRC Press, Boca Raton. Millar C., Poyer J. F., Gabelt B. T. and Kaufman P. L. (1995) Endothelin subtypes: effect on isolated rhesus monkey ciliary muscle. J Pharmacol Exp Ther.275:. 1143-1147. Morris C. E. (1990) Mechanosensitive ion channels . J Memb Biol. 113= 93-107. Moxham B. J. and Berkovitz B. K. B. (1995) The effects of external forces on the periodontal ligament. ln: The Periodontal Ligament in Health and Disease. (Eds. Berkovitz B. K. 8., Moxham B. J. and Newman H. N.)2nd edn, pp215-241. Mosby-Wolfe, London. Moxham B. J., Shore R. and Berkovitz B. K. B. (1985) Fenestrated capillaries in the connective tissues of the periodontal ligamenl. Microvasc Res. 3O= 116-124. Muller W. A. (1995) The role of PECAM-1 (CD31) in leukocyte emigration: studies in vitro and in vivo. J Leukoc Biol. 57l. 523-528. 116 Murrell E. F., Yen E. H. K. and Johnson R. B. (1996) Vascularchanges in the periodontal ligament after removal of orthodontic forces. Am J Orthod Dentofac Oñhop.'l1O:.280-286. Nathan C. and Sporn M. (1991) Cytokines in context. J Cell Biol. 113: 981-986. Nehls V. and Drenckhahn D. (1991) Heterogeneity of microvascular pericytes for smooth muscle type alpha-actin. J Cell Biol.113:147-154. Nehls V., Denzer K. and Drenckhahn D. (1992) Pericyte involvement in capillary sprouting during angiogenesis in situ. Ce// Iissue Res. 270: 469-474. Newman P. J. (1997) The biology of PECAM-1 . J Clin lnvest.l00: Supp S25-S29 Newman P. J., BerndtM. C., GorskiJ.,WhiteG. C., LymanS., PaddockC. and MullerW.A. (1990) PECAM-1 (CD-31) cloning and relation to adhesion molecules of the immunoglobulin gene superfamily. Scrence 247l' 1219-1222. Nicola N. A. (1994) lntroduction to cytokines. Guidebook to Cytokines and Their Receptors pp8-16. Oxford University Press, Oxford. Ohlen 4., Raud J., Hedqvist P. and Wiklund N. P. (1989) Microvascular effects of endothelin in the rabbit tenuissemus muscle and hamster cheek pouch. Microvasc Res. 37: 1 15-1 18. Orgul S, Cioffi G. A., Bacon D. R. and Van Buskirk E. M. (1996) An endothelin-1 -induced model of chronic optic nerve ischemia in rhesis monkeys. J Glaucoma. 5: 135-138. Page C., Rose M., Yacoub M. and Pigott R. (1992) Antigenic heterogeneity of vascular endothelium . Am J Pathol. 141: 673-683. Page R. C., Ammons W. F., Schectman L. R. and Dillingham L. A. (1974) Collagen fibre bundles of the normal marginal gingiva in the marmoset. Archs Oral Biol.19: 1039-1043. Parlange L. M. (1991) A TEM stereological analysis of blood vessels and nerves in marmoset periodontal ligament following orthodontic extrusion. MDS Thesis, The University of Adelaide, South Australia. Parlange L. M. and Sims M. R. (1993) A TEM stereological analysis of blood vessels and nerves in marmoset periodontal ligament following endodontics and magnetic incisor extrusion. Eur J Otfhod.15l.33-44. Parums D. V., Cordell J. 1., Micklem K., Heryet A. R., Gatter K. C. and Mason D. Y. (1990) JC70: a new monoclonal antibody that detects vascular endothelium associated antigen on routinely processed tissue sections. J Clin Pathol. 43= 752-757. Pedram A., Razandi M., Hu R. and Levin E. (1997) Vasoactive peptides modulate vascular endothelial cell growth factor production and endothelial cell proliferation and invasion. J Biol Chem.272= 17097-17103. Pohl U. and Busse R. (1989) Differential vascular sensitivity to luminally and adventitially applied endothelin 1 . J Cardiovasc Pharm. 13 (Suppl.): 5188-5190. Polverini P.J. (1995) The pathophysiology of angiogenesis. CnT Rev Oral Biol Med. 6:230-247. Reitan K. (1951) The initialtissue reaction incident to orthodontic tooth movement as related to the influence of function. Acta Odont Scand. Suppl. 6. 117 Reitan K. and Kvam E. (1971) Comparative behaviour of human and animal tissue during experimental tooth movement. Angle Orthod. 11l. 1-14. Rhodin J. A. G. (1967) The ultrastructure of mammalian arterioles and precapillary sphincters J Ultrastrucf Res. 18: 181-223. Rhodin J. A. G, (1968) Ultrastructure of mammalian venous capillaries, venules, and small collecting veins. J Ultrastruct Res. 25: 452-500. Risau W., Sariola H., Zen¡ves H. G., Sasse J., Ekblom P., Kemler R. and Doetschman T. (1988) Vasculogenesis and angiogenesis in embryonic-stem-cell-derived embryoid bodies. Development 102l. 47 1-478. RobertsW. E. (1994) Bone physiology, metabolism, and biomechanics in orthodontic practice. ln: Orthodontics. Current Principles and Techniques. (Eds. Graber T. M. and Vanarsdall R. L.) 4th edn, pp193-234. C.V . Mosby Co., St. Louis. Rosenberger A. L. (1978) Loss of incisor enamel in marmosets. J Mammal. 59: 207-208. Rost F. W. D. (1992) Fluorescence Microscopy Volume 1. Cambridge University Press, Cambridge. Rygh P. (1974) Elimination of hyalinized periodontal tissues associated with orthodontic tooth movement. Scand J Dent Res. 82: 57-73. Rygh P. (1984) Periodontal responses to orthodontic forces. ln: Malocclusion and the periodontium. (Eds. McNamara J. R. and Ribbens K. A.) Monograph No. 15, Craniofacial Growth Series, 17-42. The University of Michigan, Ann Arbor. Rygh P., Bowling K., Hovlandsdal L. and Williams S. (1986) Activation of the vascular system: A main mediator of periodontal fiber remodeling in orthodontic tooth movemenl. Am J Orthod.89: 453-468. Rygh P. and Brudvik P. (1995) The histological responses of the periodontal ligament to horizontal orthodontic loads. ln: The Periodontal Ligament in Health and Disease. (Eds. Berkovi2 B. K. 8., Moxham B. J. and Newman H. N.)2no edn, pp243-258. Mosby-Wolfe, London. Sachs F. (1988) Mechanical transduction in biological systems. Crit Rev Biomed Eng.'16= 141- 169. Saito M., Saito S., Ngan P. W., Shanfeld J. and Davidovitch Z. (1991) lnterleukin-1 beta and prostaglandin E are involved in the response of periodontal cells to mechanical stress in vivo and in vitro. Am J Oñhod Dentofac Oñhop.99:.226-240. Sandstedt C. (1904) Einige beiträge zur theori der zahnregulierung. Nord Tandlaeg Tidskr.5: 236- 256. Sandstedt C. (1905) Einige beiträge zur theori der zahnregulierung. Nord Tandlaeg Tidskr.6:1- ,18. Sandy J. R. (1992) Tooth eruption and orthodontic movement. Brit Dent J. 172:141-149 Sandy J. R., Farndale R. W. and Meikle M. C. (1993) Recent advances in understanding mechanically induced bone remodeling and their relevance to orthodontic theory and practice. Am J Orthod Dentofac Ofthop. 103= 212-222. 118 Sasano T., Kuriwada S., Sanjo D., lzumi H., Tabata T. and Karita K. (1992) Acute response of periodontal ligament blood flow to external force application. J Periodont Res. 27: 301- 304. Sato Y., Tsuboi R., Lyons R., Moses H. and Rifkin D. B. (1990) Characterization of the activation of latent TGF-B by co-cultures of endothelial cells and pericytes or smooth muscle cells: a self-regulating system. J Cell Biol. 111: 757 -763. Schor A. M., Canfield A. E., Sutton A. 8., Arciniegas E. and Allen T. D. (1995) Pericyte differentiation. Clin Orthop. 313: 81-91. Scott J. H. and Dixon A. D. (1978) Anatomy for Students of Dentistry. 4th edn. Churchill Livingston, London. Shiohama Y., Sasaki T. and Shibasaki Y. (1995) Histological and immunohistochemical study of endothelin-1 localization in periodontal ligament during experimental tooth movement. Dent in Jpn. 32 75-78. Simmons D. 1., Walker C., Power C. and Pigott R. (1990) Molecular cloning of CD31, a putative intercellular adhesion molecule closely related to carcinoembryonic antigen. J Exp Med. 171l. 2147-2152. Sims D. E. (1986) The pericyte - a review. Iissue Cell. '18= 153-174. Sims M. R. (1980) Angular changes in collagen cemental attachment during tooth movement. J Perio Res. 15: 638-645. Sims M. R. (1987) A model of the anisotropic distribution of microvascular volume in the periodontal ligament of the mouse mandibular molar. Aust Otthod J. 10:.21-24. Sims M. R. (1995) The morphology of the vasculature of the periodontal ligament. ln: Ihe Periodontal Ligament in Health and Disease. (Eds. Berkovitz B. K. 8., Moxham B. J. and Newman H. N.)2nd edn, pp107-120. Mosby-Wolfe, London. Sims M. R., Leppard P. 1., Sampson W. J. and Dreyer C.W. (1996) Vascular luminal changes in aged mouse periodontal ligament. J Dent Res. 75: 1503-1511. Sims M. R., Sampson W, J., Leppard P. l. and Dreyer C. W. (1992) Ageing changes in the periodontal ligament microvascular bed-luminal volumes. J Dent Res. 677: Abst 96. Skalli O., Pelte M., Peclet M., Gabbiani G., Gugliotta P., Bussolati G., Ravazzola M. and Orci L. (1989) c¿-smooth muscle actin, a differentiation marker of smooth muscle cells, is present in microfilamentous bundles of pericytes. J Histochem Cytochem. 37:. 315-321. Skalli O., Ropraz P., Trzeciak 4., Benzonana G., Gillessen D. and Gabbiani G. (1986) A monoclonal antibody against cr-smooth muscle actin: a new probe for smooth muscle differentiation. J Cell Biol. 103:2787-2796. Skalli O., Vandekerckhove J. and Gabbiani G. (1987) Actin-isoform pattern as a marker of normal or pathological smooth-muscle and fibroblastic tissues. Differentiation 33:. 232-238. Skougaard M. R., Levy B. M. and Simpson J. (1970) Collagen metabolism and periodontal membrane of the marmoset. Scand J Dent Res. 78: 256-262. Speiser P., Gittelsohn A. M. and Patz A. (1968) Studies on diabetic retinopathy. lll. lnfluence of diabetes on intramural pericytes. Arch Ophthalmol. S0= 332-337. 119 Stanley C. M. (1993) S.E.M. stereopair study of marmoset oral microvasculature: using corrosion casts to assess the gingival and PDL vessels. MDS Thesis, The University of Adelaide, South Australia. Stefanini M., DeMartino C. and ZamboniC. (1967) Fixation of ejaculated spermatozoa forelectron microscopy. Nature. 216= 17 3-17 4. Storey E. (1973) The nature of tooth movement. Am J Ofthod. 63= 292-314 Storti R. V. and Rich A. (1976) Chick cytoplasmic actin and muscle actin have different structural genes. Proc Natl Acad Sci. USA73:2346-2350. Sunderkötter C., Steinbrink K., Goebeler M., Bhardwaj R. and Sorg C. (1994) Macrophages and angiogenesis. J Leukoc Biol. 55l. 410-422. Sussman R. W. and Kinzey W. G. (1984) The ecological role of the callitrichidae'. a review. Am J Phys Anthrop. 64: 419-449. Tang M. P. F. and Sims M. R. (1992) A TEM analysis of tissue channels in normal and orthodontically tensioned rat molar periodontal ligament. Eur J Orthod. 14:. 433-444. Trueta J. (1963) The role of the vessels in osteogenesis. J Bone Joint Surg. 458: 402-418. Vandekerckhove J. and Weber K. (1978) At least six different actins are expressed in a higher mammal: an analysis on the amino acid sequence of the amino-terminal tryptic peptide. J Mol Biol.126:783-802. Van Noorden S. and Polak J. M. (1983) Techniques and practice. ln'. lmmunocytochemistry Practical Applications in Pathology and Biology. (Eds. Polak J. M. and Van Noorden S.) pp11-42. Wright PSG lnc., Bristol. Waldo C. M. and Rothblatt J. M. (1954) Histologic response to tooth movement in the laboratory rat: procedure and preliminary observation. J Dent Res. 33: 481-486. WarnerT. D., ElliottJ. D. and Ohlstein E. H. (1996) California dreamin"boutendothelin: emerging new therapeutics. Irends Pharmacol Sci.'17=177 -179. Webb D. J. (1997) Endothelin: from molecule to man. Br J Clin Pharmacol. 44t 9-20. Weekes W. T. (1983) Vascular morphology of rat molar periodontium. MDS Thesis, The University of Adelaide, South Australia. Weekes W. T. and Sims M. R. (1986) The vasculature of the rat molar gingival crevice. J Perio Res. 2l= 177-185. Weideman M. P. (1984) Architecture. ln:Handbookof Physiology. (Eds. Renkin E. M. and Michel C C ) Section 2: The Cardiovascular System. Volume lV, Microcirculation. pp1'1-40. Am Physiol. Soc., Bethesda, Maryland. Weir A. P. (1990) The marmoset periodontal ligament: a T.E.M. stereological analysis following incisor crown fracture, endodontic therapy, orthodontic extrusion and long term retention. MDS Thesis, The University of Adelaide, South Australia. Wilkes B. M., Susin M., Mento P., Macica C. M., Girardi E., Boss E. and Nord E. (1991) Localization of endothelin-like immunoreactivity in rat kidneys. Am J Physiol.260: (Renal Fluid Electrolyte Physiol. 29) F913-F920. 120 Wills D., Picton D. and Davies W. (1976) A study of the fluid systems of the periodontium in macaque monkeys. Archs Oral Biol.21:175-185. Wilson N. H. F. and Gardner D. L. (1982) The postnatal development of the temperomandibular joint of the common marmoset (callithrix jacchus). J Med Primatol. 1l : 303-31 1 . Wong M. E. K., Hollinger J. O. and Pinero G. J. (1996) lntegrated processes responsible for soft tissue healing. Oral Surg Oral Med Oral Pathol Oral Radiol Endod.82:475-492. Wong R. S. T. and Sims M. R. (1987) A scanning electron-microscopic study of methacrylate corrosion casts of the mouse palatal and molar periodontal microvasculalure. Archs Orat Biol. 32:557-566. Yamasaki K., Shibata Y. and Rukuhara T. (1982) The effect of prostaglandins on experimental tooth movement in monkeys. (macaca fuscata). J Dent Res 61l.1444-1446. Yamasaki K., Shibata Y., lmaiS., TaniY., ShibasakiY. and Fukuhara T. (1984) Clinicalapplication of prostaglandin E1 (PGEr) upon orthodontic tooth movement. Am J Ofthod.85: 508-518. Yanagisawa M., Kurihawa H., Kimura S., Tomobe Y., Kobayashi M., Mitsui Y., Yazaki Y., Goto K. and Masaki T. (1988) A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nafure (London). 332: 41 1-415. Zimmermann K. W. (1923) Die feinere bau der blutcapillaren. Z Anat Entwickl. 68: 29-109.