Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of sinusoidal stretching and c48/80 on the morphology of dermal fibroblasts and the mechanical… Martel, Hélène 1999

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-0213.pdf [ 9.64MB ]
Metadata
JSON: 831-1.0058617.json
JSON-LD: 831-1.0058617-ld.json
RDF/XML (Pretty): 831-1.0058617-rdf.xml
RDF/JSON: 831-1.0058617-rdf.json
Turtle: 831-1.0058617-turtle.txt
N-Triples: 831-1.0058617-rdf-ntriples.txt
Original Record: 831-1.0058617-source.json
Full Text
831-1.0058617-fulltext.txt
Citation
831-1.0058617.ris

Full Text

E F F E C T O F SINUSOIDAL S T R E T C H I N G A N D C48/80 O N T H E MORPHOLOGY OF DERMAL FIBROBLASTS AND THE MECHANICAL R E S P O N S E O F SKIN  by HELENE MARTEL B.A.Sc, Laval University, 1996 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER IN APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Chemical Engineering We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April 1999 ©Helene Martel, 1999  In presenting this thesis in partial fulfilment  of the  requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or by his  or  her  representatives.  It  is  understood  that  copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  rfH'M-Rtl  Abstract Dermal fibroblasts are connected to each other and to elements of the extracellular matrix (ECM). In vitro studies using fibroblasts grown in collagen gels have demonstrated that fibroblasts are able to deform the gel via cell adhesions. The purpose of this project was to investigate qualitatively the effect of sinusoidal stretching of fresh strips of skin on both the mechanical response of the tissue and the morphology of the fibroblasts. These investigations were carried out for skin soaked in a physiological solution (Kreb's buffer) with and without addition of a mast cell degranulating material, compound 48/80 (C48/80). Two by 10 millimeters strips of skin were removed from the back of 3 male rats. Skin samples were either only soaked in Kreb's buffer in the presence or absence of C48/80, or simultaneously soaked and intermittently sinusoidally stretched in Kreb's buffer in the presence or absence of C48/80.  Controls  consisted of skin samples fixed immediately after excision from the rats or samples only soaked in buffer.  All tissues were fixed and processed for  transmission electron microscopy. Morphometric analysis of the relative surface area of fibroblast cytoplasmic extensions and cell bodies showed that sinusoidal stretching of rat skin results in the retraction of fibroblast cytoplasmic extensions and "rounding up" of the cells.  Incubation with C48/80 alone had no apparent effect on  fibroblast cytoplasmic extensions and morphology. The measurement of tension during sinusoidal stretching in Kreb's buffer with or without addition of C48/80 demonstrated that the tension has a tendency  to decrease as the number of sine wave cycles and stretching sequences increased. In addition, during the 14.33 minute interval between two successive stretching sequences the tissue demonstrated recovery of tension. The decrease in tension observed as a result of biomechanical experiments along with the "rounding up" of fibroblasts demonstrated in the morphological study are consistent with the hypothesis that cell adhesions play a role in the mechanical response of the tissue.  The biomechanical and  morphological results, however, did not support the hypothesis that C48/80 alone induces lost of fibroblast adhesions.  iii  Table of Contents Abstract  ii  Table of Contents  iv  List of Tables  List of Figures  Acknowledgement  viii  ix  xii  1. Introduction  1  2. Literature Review  6  2.1 C E L L INTERACTIONS  6  2.2 M E A S U R E M E N T S O F BIOMECHANICAL P R O P E R T I E S O F SKIN  10  2.3 F I B R O B L A S T BIOMECHANICS IN C O L L A G E N SOLUTIONS  16  2.4 M E C H A N I C A L P R O P E R T I E S O F C O L L A G E N G E L S  19  2.5 T H E R O L E O F C E L L S WITH R E S P E C T T O INTERSTITIAL FLUID P R E S S U R E  22  3. Materials and Methods 3.1 S T R E S S - S T R A I N - T I M E A P P A R A T U S  3.1.1 Basic components of the stress-strain-time apparatus  26 26  26  3.1.2 Electronic and control components of the stress-strain-time apparatus  29  3.1.3 Linearity of the transducer  32  3.2 TRANSMISSION E L E C T R O N M I C R O S C O P E  33  iv  3 . 3 EXPERIMENTAL P R O C E D U R E S  33  3.3.1 Equipment and solutions preparation  33  3.3.2 Harvesting tissue samples  34  3.3.3 Treatment groups  36  3.3.4 Qualitative biomechanical experiments  37  3.3.5 Processing tissue for transmission electron microscopy  39  3 . 4 M O R P H O M E T R Y ANALYSES  39  3.4.1 Micrographs  39  3.4.2 Point counting  41  3.4.3 Roundness of fibroblasts  43  3.4.4 Statistical Analyses  44  4. Results 4.1 PRELIMINARY EXPERIMENTS  47 47  4.1.1 Verification of fibroblast viability  47  4.1.2 Verification of mast cell degranulation  50  4.2 MORPHOLOGICAL R E S U L T S  51  4.2.1 Fibroblast micrographs for each group  52  4.2.2 Surface density ratio  54  4.2.3 Width/length ratio  55  4.2.4 Fibroblast morphology of skin stretched 1 and 4 sequences of sinusoidal waves  57  4.2.5 Fibroblast morphology of skin stretched for 5 sine waves with and without C48/80  61  v  4 . 3 QUALITATIVE BIOMECHANICAL R E S U L T S  64  4.3.1 Stretching in Kreb's buffer 4.3.2 Stretching in C48/80 4.3.3 Stretching of previously frozen tissues 5. D i s c u s s i o n 5.1 S T R E S S - S T R A I N - T I M E  64 ,  69 73 80  APPARATUS  80  5.2 VIABILITY O F FIBROBLASTS  84  5.3 E F F E C T O F SOAKING O N FIBROBLAST M O R P H O L O G Y  87  5.4 E F F E C T O F S T R E T C H I N G O N FIBROBLAST M O R P H O L O G Y  88  5.5 E F F E C T O F C 4 8 / 8 0 O N FIBROBLAST M O R P H O L O G Y  89  5.6 E F F E C T O F S T R E T C H I N G O N SKIN TENSION  92  5.7 E F F E C T O F C 4 8 / 8 0 A N D S T R E T C H I N G O N SKIN TENSION  96  6. C o n c l u s i o n s  98  7. Recommendations  101  8. References  104  Appendix A LINEARITY A N D CALIBRATION O F T H E F O R C E T R A N S D U C E R  110  vi  Appendix B R E C O R D S O F POINT COUNTING  111  Appendix C STATISTICAL A N A L Y S E S  117  Appendix D ADDITIONAL G R A P H S F R O M BIOMECHANICAL EXPERIMENTS  118  vii  List of Tables T A B L E 3.1  INGREDIENTS O F T H E A L T E R E D K R E B ' S SOLUTION (G/L)  List of Figures F I G U R E 1.1  DIAGRAM O F SKIN  F I G U R E 2.1  S C H E M A T I C R E P R E S E N T A T I O N O F T H E S T R U C T U R E O F AN INTEGRIN  RECEPTOR  2  8  F I G U R E 2.2  S T R E S S - S T R A I N C U R V E F O R RAT SKIN  12  F I G U R E 2.3  S T R E S S R E C O R D E D DURING R E P E A T E D EXTENSION  14  F I G U R E 2.4  R E C O V E R Y IN TENSION A T DIFFERENT TIME INTERVALS  15  F I G U R E 2.5  S C H E M A T I C PRESENTATION O F A C E L L O N A GRID WITH A S T E L L A T E A N D  BIPOLAR S H A P E F I G U R E 2.6  S C H E M A T I C DRAWING O F T H E TRANSCAPILLARY-INTERSTITIAL E X C H A N G E  S Y S T E M O F A C O N N E C T I V E TISSUE FIGURE 2.7  19  23  S C H E M A T I C DRAWING O F P R O P O S E D E V E N T S A S S O C I A T E D WITH  D E V E L O P M E N T O F INCREASED NEGATIVITY O F INTERSTITIAL FLUID P R E S S U R E (P.F)  25  F I G U R E 3.1  B A S I C C O M P O N E N T S O F T H E STRESS-STRAIN-TIME A P P A R A T U S  27  F I G U R E 3.2  COMPLETE SYSTEM  30  F I G U R E 3.3  S A W - T O O T H MOTION P R O D U C E S BY DIGITAL P U L S E S A N D L E V E L  DIRECTION  31  F I G U R E 3.4  M E S H GRID WITH TISSUE SECTION  40  F I G U R E 3.5  P O I N T COUNTING  41  F I G U R E 3.6  M E A S U R E M E N T O F FIBROBLAST WIDTH/LENGTH RATIO  44  F I G U R E 4.1  A N A P P A R E N T L Y NORMAL DERMAL FIBROBLAST O B S E R V E D IN A SKIN  S A M P L E S O A K E D IN K R E B ' S B U F F E R F O R 9 0 MINUTES B E F O R E FIXATION  48  ix  FIGURE 4.2  F I B R O B L A S T UNDERGOING NECROSIS (A) A N D A P O P T O S I S (B) IN SKIN  S A M P L E S S T R E T C H E D WITH A STRAIN O F 2 0 % IN K R E B ' S B U F F E R  50  FIGURE 4.3  M A S T C E L L A N D M A S T C E L L DEGRANULATION  51  FIGURE 4.4  FIBROBLASTS FROM EACH TREATMENT G R O U P  53  FIGURE 4.5  S U R F A C E DENSITY RATIO  55  FIGURE 4.6  F I B R O B L A S T S WIDTH/LENGTH RATIO  56  F I G U R E 4 . 7 S U R F A C E DENSITY RATIO FOR VARIOUS N U M B E R O F SINUSOIDAL SEQUENCES FIGURE 4.8  59  W I D T H / L E N G T H RATIO F O R VARIOUS N U M B E R O F SINUSOIDAL  SEQUENCES  60  F I G U R E 4 . 9 S U R F A C E DENSITY RATIO F O R SKIN S T R E T C H E D F O R 5 SINE W A V E C Y C L E S IN K R E B ' S B U F F E R WITH OR WITHOUT ADDITION O F C 4 8 / 8 0 (0.1 MG/ML) ( 2 0 % STRAIN)  62  F I G U R E 4 . 1 0 W I D T H / L E N G T H RATIO F O R SKIN S T R E T C H E D F O R 5 SINE W A V E C Y C L E S IN K R E B ' S B U F F E R WITH O R WITHOUT ADDITION O F C 4 8 / 8 0 (0.1 MG/ML) ( 2 0 % STRAIN) F I G U R E 4.11  63 G R A P H O F TENSION AS A FUNCTION O F SINUSOIDAL W A V E S A N D  S E Q U E N C E S F O R A SKIN S A M P L E S U B J E C T E D T O AN INTERMITTENT SINUSOIDAL S T R E T C H I N G IN K R E B ' S B U F F E R  66  F I G U R E 4 . 1 2 T E N S I O N A S A P E R C E N T A G E O F T H E FIRST SINUSOIDAL S E Q U E N C E F O R A SKIN S A M P L E S T R E T C H E D A T A STRAIN O F 2 0 % IN K R E B ' S B U F F E R E A C H 1 5 MINUTES F O R A PERIOD O F 9 0 MINUTES  68  X  F I G U R E 4 . 1 3 T E N S I O N A S A FUNCTION O F SINUSOIDAL W A V E S A N D S E Q U E N C E S F O R A SKIN S A M P L E S U B J E C T E D T O AN INTERMITTENT SINUSOIDAL S T R E T C H I N G IN K R E B ' S B U F F E R CONTAINING F I G U R E 4.14  C48/80 (0.1  71  MG/ML)  T E N S I O N A S A P E R C E N T A G E O F T H E FIRST SINUSOIDAL S E Q U E N C E F O R A  TISSUE S A M P L E S T R E T C H E D A T A STRAIN O F  20%  IN  C48/80 (0.1  MG/ML) E A C H  15  MINUTES F O R A PERIOD O F 90 MINUTES F I G U R E 4.15  T E N S I O N A S A FUNCTION O F SINUSOIDAL W A V E S A N D S E Q U E N C E S F O R A  SKIN S A M P L E S T R E T C H E D IN K R E B ' S B U F F E R E A C H H O U R F O R A 3 H O U R PERIOD F I G U R E 4.16  72  ..75  T E N S I O N A S A FUNCTION O F SINUSOIDAL W A V E S A N D S E Q U E N C E S F O R A  SKIN S A M P L E PREVIOUSLY F R O Z E N A N D S T R E T C H E D IN K R E B ' S B U F F E R E V E R Y HOUR F O R A 3 H O U R PERIOD  76  F I G U R E 4 . 1 7 T E N S I O N A S A P E R C E N T A G E O F T H E FIRST SINUSOIDAL S E Q U E N C E F O R SKIN S A M P L E S S T R E T C H E D IN K R E B ' S B U F F E R E V E R Y HOUR F O R A 3 H O U R PERIOD  77  F I G U R E 4 . 1 8 T E N S I O N A S A P E R C E N T A G E O F T H E FIRST SINUSOIDAL S E Q U E N C E F O R PREVIOUSLY F R O Z E N SKIN S A M P L E S S T R E T C H E D IN K R E B ' S B U F F E R E V E R Y H O U R FOR A 3 H O U R PERIOD  78  F I G U R E 4 . 1 9 T E N S I O N A S A FUNCTION O F SINUSOIDAL W A V E S A N D S E Q U E N C E S F O R A TISSUE S A M P L E PREVIOUSLY F R O Z E N A N D S T R E T C H E D IN K R E B ' S B U F F E R E V E R Y HOUR F O R A 3 H O U R PERIOD  79  xi  Acknowledgement I would like to thank my supervisor Dr. Joel Bert for his continuous support in completing this project. I would like also to thank Dr. David Walker with whom it has been a great pleasure to discover the world of microscopy. I must also mentioned Dr. Ken Pinder who has always been there to answer my questions, particularly concerning the apparatus. I would like also to thank Fanny Chu for all the tissue sectioning she has done for me and for her patience to teach me the technical procedures in microscopy as well as for her friendship. I am grateful for the financial support provided by the Natural Sciences and Engineering Research Council of Canada (NSERC). Last but not least, I would like to thank Patrick from the bottom of my heart for his continuous encouragement and help throughout this project.  xii  1 . Introduction Skin is one of the largest organs of the human body. Its structure consists of two layers, the epidermis and the dermis. A diagram of the skin is presented in Figure 1.1.  The outer layer of the human skin, the epidermis, is thin (0.07-  0.12 mm) (Silver, 1987) and composed principally of cells (e.g. viable and dead epithelial cells). The second layer, the dermis, is about 1 to 4 millimeters thick (Silver, 1987) and consists primarily of fibrous proteins (collagen and elastin) embedded in a gel-like matrix. The collagen fibres in the upper portion of the dermis, where it contacts with the epidermis, are thin and form loose connective tissue whereas the deeper portion of the dermis is characterized as a dense irregular connective tissue.  Hair follicles, glands and capillaries are other  structural components within the dermis.  Cells are also present in the dermis  and the most numerous ones found in this layer are the fibroblasts.  The  principal function of these cells is the production of collagenous and elastic fibres and some components of the extracellular matrix (e.g. hyaluronan) (Montagna ef al., 1974). Furthermore, it has been demonstrated that fibroblasts are connected to each other and elements of the extracellular matrix (ECM) extensions (Grinnel, 1994; Horwitz, 1997; Rubin etal.,  by cytoplasmic  1995).  Cell-cell and cell-matrix interactions have been a subject of increasing interest for the past 15 years. These adhesions, which are mediated by specific receptors or binding proteins, are of importance because they play a role in the physical organization of tissues and cells, and also in the regulation of complex cellular functions such as embryonic development (Clark et al., 1995; Horwitz, 1  1997), tissue fluid balance (Reed ef al., 1997; Rubin ef al., 1995) and cell migration (Clark etal., 1995; Delvoye etal., 1991; Horwitz, 1997).  Dermis  Macrophage  Elastic fibre  Capillaries  Figure 1.1 Diagram of skin  The structure of skin is composed of two main layers, the epidermis and the dermis. Under the dermis, adipose tissue (fat) is found. The epidermis is divided in two major layers called the stratum corneum (layers of dead cells) and stratum germinativium (formation of new cells). The surface of the dermis where it contacts the epidermis is called the papillary layer. In this layer the collagen fibres (hatched bars) are thinner than in the reticular layer which is the deeper portion of the dermis. Fibroblasts, mast cells and macrophages are cellular components of the dermis. Proteoglycans and hyaluronan (thick coiled lines) are structural components associated with collagen and elastic fibres (thin straight lines).  2  Most of the studies on cell-cell and cell-matrix attachments have been performed in vitro using  fibroblast populated collagen  lattices. These  experiments have demonstrated that cell-cell and cell-matrix attachments have mechanical consequences for both the gel and the cells, such as maintaining a tensional homeostasis (stable tension) across the lattice (Brown et al., 1998; Delvoye et al., 1991; Eastwood et al., 1994).  Morphological changes of  fibroblasts have also been observed in the collagen lattices experiments. Cultured fibroblasts are round cells, but in collagen solution they spread through the gel and acquire a stellate form as the number of attachments with other cells and the ECM increases. Changes from stellate to round cells with retraction of cytoplasmic extensions have also been observed in vitro under specific conditions, such as in relaxation experiments in which an applied load across the gel is suddenly released (Grinnell, 1994; Mochitate et al., 1991; Tomasek et al., 1992). Biomechanical properties of tissues such as skin have been studied both in vitro and in vivo. Protecting the body against external injuries and mechanical trauma, being an active barrier between the internal and external environments and allowing movements in response to tension or compression are examples of functions that skin can perform in part because of its mechanical properties. In particular, viscoelastic properties including elasticity and tensile strength, allow the skin to perform many of these functions. In addition to their importance in the basic physiology of skin, the mechanical properties play a significant role in human health because they are  3  related to pathophysiological conditions such as inflammation and also to the wound healing process. Due to the importance of the mechanical properties of the skin, measurements of these properties have been performed under a wide variety of experimental conditions. Maturation (Belkoff et al., 1991), aging (Cook, 1989; Cua et al., 1990; Vogel et al., 1979; Vogel, 1975 and 1989), sex (Cua ef al., 1990), desmotropic drugs (Vogel et al., 1979; Vogel, 1975), anatomical region (Cua et al., 1990) and freezing (Foutz et al., 1992) are examples of aspects that have been studied in relation with the mechanical properties of skin. Fibroblast activity (e.g. cell migration and attachment) has been hypothesized as a potential factor that could affect the mechanical properties of skin (Vogel, 1993), but the role of these cells with respect to biomechanics has yet to be investigated in tissues. The objective of this thesis was to investigate qualitatively the effect of sinusoidal stretching of strips of fresh skin on both the morphology of the fibroblasts and the mechanical response of the tissue.  When skin is  mechanically stimulated, adhesions between fibroblasts and with elements of the ECM may break down and thus allow the fibroblasts to round up as it has been observed in collagen gel studies. These investigations were carried out for skin soaked in a physiological solution (Kreb's buffer) in the presence and absence of a mast cell secretagogue (degranulating material), compound 48/80 (Koller, 1993; Selye, 1965). Mast cells are resident cells of the dermis whose cytoplasm are characterized by large granules that contain heparin, histamine and  4  serotonin. When mast cell degranulation occurs, either as a result of physical (e.g. cold) or chemical (e.g. compound 48/80) stimuli, these compounds are released to initiate an inflammation reaction within the tissue. This inflammation, caused subsequent to mast cell degranulation, is accompanied by a rapid accumulation of fluid within the tissue (edema) and it is hypothesized that this edema is a result in part of breaking of fibroblast adhesions. The project consisted of two parts: a morphological and a biomechanical study. For the morphological study, using transmission electron microscopy, we investigated the effects of stretch and/or exposure to C48/80 on aspects of fibroblast shape in the dermis of the skin. Morphometric analysis of the relative surface area of fibroblast cytoplasmic extensions and cell bodies (section 4.2.2) and morphometric assessment of the roundness of fibroblast cell bodies (sections 4.2.3) were performed to characterize the changes in fibroblast shape. In the biomechanical study, the force resulting from intermittent sinusoidal stretching (i.e. tension) of skin samples was recorded and the change in tissue tension with the number of sine wave cycles was graphically represented for skin stretched in Kreb's buffer with (section 4.3.2) or without the secretagogue C48/80 (section 4.3.1).  5  2. Literature Review Cell-cell and cell-matrix interactions in connective tissues have been the subject of increased research activity during the last decade. A description of the mechanism by which cells connect to each other and to elements of the extracellular matrix is presented in this chapter. Because there is already an extensive literature on the mechanical properties of skin, it is reviewed in this part of the thesis.  Most research on fibroblasts and fibroblasts mechanical  activity relevant to this study has been in vitro studies in which cells were grown in collagen gels and thus, the results obtained in these studies are also considered in this chapter. Finally, this section ends with a consideration of research that addresses the possible relationship between cell to extracellular matrix connections and interstitial fluid pressure.  2.1 Cell interactions  The connective tissues of skin, tendons and cartilage, are composed of a cellular component and an extracellular matrix (ECM).  The ECM of most  connective tissues is composed of fibrous proteins (some collagen and elastin) embedded in a gel-like matrix constituted principally of proteoglycans (covalently linked polysaccharide and protein complexes), glycoproteins and hyaluronan (a polysaccharide) (Montagna et al., 1974; Silver, 1987). These macromolecules found in the ECM interact to form large supramolecular structures. Cells of connective tissues, mostly fibroblasts, may interact also with elements of the  6  ECM and with each other. In the past decade, intensive studies of these cell-cell and cell-matrix interactions have shown that they are important, not only for physical organization of tissues and cells (Delvoye et al., 1991; Grinnell, 1994), but also for regulating complex cellular functions, such as embryonic development (Horwitz, 1997) and tissue fluid balance (Reed et al., 1997; Rubin etal.,  1995). Adhesion between cells and matrix are apparently mediated by specific  receptors or binding proteins located on the cell surface:  Most of these  receptors belong to a family of receptors called integrins (Adams, 1997; Horwitz, 1997; Rubin et al., 1995). Integrins also participate in cell-cell adhesion, but the molecules responsible for most cell-cell interactions belong to groups such as the cadherin, selectin and immunoglobulin families (Adams, 1997; Horwitz, 1997). A schematic representation of the structure of an integrin on cells is presented in Figure 2.1. Integrins are transmembrane heterodimeric cell surface glycoproteins composed of an a and p subunit.  Currently, 16 a- and 8 p-  subunits have been described and these subunits combine into at least 20 different integrins (Clark etal., 1995; Horwitz, 1997). Adhesion between cells and the ECM occurs at specific sites termed focal adhesions (Figure 2.1) (Horwitz, 1997; Rubin et al., 1995).  At these sites  integrins are simultaneously linked with other molecules (ligands) outside the cell and aggregated components of the cytoskeleton, to form the organized focal adhesion complexes. These focal adhesions not only physically link cells to each other or to elements of the ECM, but are also used for signal  7  Collagen fibre Fibronectin  a subunit of integrin  1  B subunit of intearin  Cell membrane  Cytoplasm Focal adhesion  Molecules (e.g. actin, talin, vinculin) Figure 2.1 receptor  Schematic  representation of the structure of an  integrin  Integrins consist of two glycoprotein chains or subunits, a and p, with each subunit contributing to one tail and part of the head of the overall structure. The "head" of the integrin may be attached either to molecules of the extracellular matrix (e.g. fibronectin) or to molecules on other cells. The "tail" of each subunit is attached to the cell's own scaffolding, or cytoskeleton. Integrins connect to the cytoskeleton through an aggregate of molecules (e.g. actin, talin, vinculin) called focal adhesions. Figure based on the paper of Horwitz (1997).  8  transduction between molecules outside the cell and the cellular cytoplasm ("outside-in signaling") (Horwitz, 1997; Rubin etal., 1995). Both gene expression and cell division may result from integrin signal transduction (Horwitz, 1997). Activation of molecules found in focal adhesion complexes seem to be the principal reason for this internal signaling. Integrins also respond to signals coming from inside the cell. This inside-outside signaling can, for example, change the affinity of an integrin for a specific ligand or change the strength with which the integrins bind. Such changes were shown to play a role in inflammatory reactions; after inside-outside signaling integrins on the white blood cells (leukocytes) gained affinity for molecules of the immunoglobulin family on endothelial cells and these attachments help the leukocytes to cross the blood vessel wall into the damaged or infected tissue (Horwitz, 1997). The formation of a network of platelets (small blood cells that lack a nucleus) and proteins at a site of injury in a blood vessel results also from the activation of integrins on platelets by internal signaling; this aggregate of platelets and proteins prevent the blood loss until the wound is repaired (blood clotting) (Horwitz, 1997). Intercellular communication involving the passage of inorganic ions and other small water-soluble molecules from the cytoplasm of one cell to the cytoplasm of the other cell occurs in gap junctions (Alberts et al., 1989). This type of cell junction is found in many tissues (e.g. skin, smooth muscles) and is characterized by a narrow gap of about 3 nm wide present between the membranes of two adjacent cells. The exchange of ions and molecules between  9  cells has important functional implications such as the synchronization of the contractions of heart muscle cells (Alberts etal., 1989). Occluding junctions (tight junctions) and anchoring junctions are other types of cell attachments that are found in tissues (Alberts et al., 1989). Tight junctions seal epithelial cells together whereas anchoring junctions mechanically attach cells from many tissues to their neighbors or the ECM with two proteins, an intracellular attachment protein and a transmembrane linker glycoprotein. Fibroblasts connect within each other and with elements of ECM through gap and anchoring type junctions.  2.2 Measurements of biomechanical properties of skin  Research on the biomechanical properties of skin began in the early years of the 20 century. Over the years, a wide variety of techniques has been th  developed to investigate a number of physical properties of skin (Edwards et al., 1995; Vogel, 1994).  For example, stress-strain relationships during uniaxial  elongation of skin (Colin, 1982; Oxlund etal., 1988; Oxlund etal., 1980; Tong et al., 1976), creeping behaviour, i.e., measurement of the tissue elongation during constant load (Vogel, 1977), stress relaxation, i.e., measurement of the force bore by a tissue under constant deformation (Vogel, 1985; Wan Abas, 1995) and measurement of hysteresis during cyclic loading and unloading (Vogel, 1983 and 1981) are studies that relate to the mechanical characteristics of skin. The force per unit cross-sectional area that a tissue bears is termed the stress and the deformation of a tissue with respect to its initial length is called the 10  strain. A typical stress-strain curve for the skin is shown in Figure 2.2. This curve can be broken into three regions. In the first section [A-B], the collagen fibres in skin are unaligned and therefore little resistance is offered to the deformation of skin. In this initial part of the curve, the straining of the skin is influenced mostly by the network of elastic fibres present in the dermis (Oxlund et al., 1988). At higher strain, elastic fibres have no significant effect on the "stiffness" of skin compared to the collagen network. As the elongation (strain) increases, the collagen fibres become aligned in the direction of the applied force and offer more resistance to the extension of the tissue which results in large increments in stress for small increments in strain. When the majority of collagen fibres are aligned the stress-strain relationship becomes linear [B-C]. The slope of this straight part of the curve is a measure of the "stiffness" of the tissue and is called the Young's modulus or a modulus of elasticity. In the third region [C-D], the fibres begin to break apart until rupture of the specimen. This breaking point is the ultimate strain and ultimate load for the tissue being stressed. Measurement of these parameters in stress-strain experiments, i.e. Young's modulus and ultimate values, can be used for example to demonstrate the effects of aging on the biomechanical properties of skin (Vogel etal., 1977). The first part of the stress-strain curve, the low strain region, is the physiological range in which the skin usually functions.  Therefore, some  researchers have concentrated their efforts on this part of the curve of small forces and strains (Foutz et al., 1994; Mansour et al., 1993; Vogel et al., 1977; Wijn, 1980).  However, due to the difficulty in obtaining repeatable  11  measurements at low strain (Mansour et al., 1993), most of the studies on the mechanical properties of skin have been performed in the high strain region of the curve.  Stress (N/mm 2) A  12.0  —|  0.0  10.0  20.0  30.0  40.0 50.0 Strain (%)  60.0  70.0  80.0  90.0  Figure 2.2 Stress-Strain curve for rat skin  Stress-strain curve for rat skin excised longitudinally to the body axis. The low strain region (A-B) involves the alignment of collagen fibres along the stress direction. When the fibres are aligned, the stress-strain relationship becomes linear (region B-C). In the third region (C-D) fibres begin to yield and there is rupture of the specimen. Reproduced from Vogel (1989).  12  If a tissue is tested by repeated loading and unloading at a constant rate of elongation, it has been demonstrated that the stress-strain curve shifts to the right with an increase of the low strain region (i.e. region A-B on Figure 2.2) (Fung, 1994; Silver et al., 1987). The stress-strain curve shifts during the first three cycles and then becomes repeatable (i.e. the curve no longer shifts). In the first few cycles the internal structure of the tissue changes until it reaches a steady state where no further change occurs unless the cycling routine is changed. The tissue is then said to have been preconditioned. Preconditioning is usually performed in order to obtain reproducible experimental results. Like most biological tissues, skin is a viscoelastic material, i.e. its mechanical properties are time and hence frequency dependent. With steadystate stress-strain measurements, elastic properties of skin but not the viscous properties are determined.  Hysteresis (Vogel, 1983 and 1981), relaxation  (Vogel, 1985; Wan Abas, 1995) and creep (Vogel, 1977) experiments are techniques used to measure the viscous properties of skin. Repeated strain experiments have been used as another approach to scribe the viscoelastic properties of a material (Vogel, 1993; Vogel et al., 1985). With this technique, the skin is repeatedly extended over a specified distance. Vogel et al. (1985) have shown that after repeated loading and unloading, the stress values decrease depending approximately on the logarithm of the number of the cycles (Figure 2.3). This method has also been used to study the in vivo recovery of mechanical properties of skin (Vogel, 1993; Vogel et al., 1985). Vogel (Vogel, 1993) and his collaborators (Vogel etal., 1985) demonstrated that  13  the initial stress value measured in the first cycle was fully recovered after a 16 hour period of rest when skin was stretched perpendicularly to the body axis. For the same resting period a recovery of about 96% was observed in skin stretched longitudinally to the body axis (Figure 2.4). This mechanical recovery is suggested to be related to cellular activity, especially that of fibroblasts (Vogel, 1993; Vogel et al., 1985).  Stress (N/mm 2) A  0.5  —|  0.4  —  0.3  —  —  0.2  0-1  0.0  Cycle number  Figure 2.3 Stress recorded during repeated extension  Stress measured for a rat skin sample extended 30 times up to 50% of its resting length with an extension rate of 100 mm/min. Reproduced from Vogel ef al., 1985. 14  %  110 —|  0  5  10  15  20  25  30  Cycle number  Figure 2.4 Recovery in tension at different time intervals  The stress values of the second stretching test is shown as a percentage of the first stretching test for different periods of rest (0.5, 1, 6 and 16 hours). These results are for a skin sample stretched longitudinally to the body axis. To avoid confusion in the reading of the data, only the negative or positive portion of the standard deviation has been depicted (one sided error bars). Reproduced from Vogel etal., 1985.  All the methods presented above have been used widely and proved to be useful for the characterization of the biomechanical properties of skin. However, because of the viscoelasticity of skin, dynamic oscillation testing seems to be the most appropriate way to measure the relevant properties (e.g elasticity, strength) 15  of this tissue (Fung, 1994). By applying small periodic deformations to a tissue sample, dynamic testing more closely mimics the physiological movement of most living tissues and facilitates model fitting using the theories of linear viscoelasticity (Wang, 1997). One way to perform dynamic testing consists of modulating sinusoidally an input variable or signal at a given amplitude and frequency and then observing the variations of the output quantities in response to that particular sinusoidal forcing (Hougen, 1964; Hougen et al., 1961; Navajas et al., 1995). The procedure is then repeated for the same amplitude using different frequencies of the input disturbance. A more widely used method is the pulse wave propagation (Hougen, 1964; Hougen etal., 1961; Mridha etal., 1992; Pereira etal., 1991; Potts etal., 1983). This technique is usually preferred to the sinusoidal forcing technique because from a single pulse test, the tissue response can be obtained for a wide range of frequencies.  The desired  frequency response can then be extracted mathematically.  2.3 Fibroblast biomechanics  in collagen  solutions  Fibroblasts are the principle cell type of connective tissues, including the connective tissue of skin. Most biomechanical studies of fibroblasts have been done in vitro where fibroblasts are cultured both on and in collagen gels (lattices).  A number of recent studies have focused on contractile forces  generated by fibroblasts grown within collagen gels (Barocas et al., 1997; Barocas etal., 1995; Delvoye etal., 1991; Eastwood etal., 1994). Usually, a decrease in the lattice volume can be observed within hours after the initiation of 16  fibroblast-collagen gel contraction. The contractile forces observed on the collagen lattices are thought to be mediated by the fibroblasts (Grinnell, 1994). However, the exact mechanism by which fibroblasts contract the gel has been disputed for a while.  At first, the contraction of collagen gels by cultured  fibroblasts was suggested to result from their behaving as myofibroblasts (Eastwood etal., 1996; Grinnell, 1994). Subsequently, it was suggested that the contractile forces were generated by fibroblast migration which by exertion of tractional forces on the gel, re-arranged and compacted the collagen fibres (Brown etal., 1996; Eastwood et al., 1996). To better understand the process of gel contraction, effective quantitative measurement of the forces produced in the collagen lattices was necessary. Therefore, devices, such as the Culture Force Monitor (CFM), were recently developed (Eastwood et al., 1996). Using the CFM, Eastwood et al. (1996) measured the force of contraction generated in a collagen gel over a period of 24 hours and monitored fibroblasts morphology throughout the period of contraction. These observations showed that most of the force generated in the collagen gel was a result of fibroblast attachment and movement within the collagen lattice using tractional forces (Eastwood et al., 1996). When cultured in collagen lattices, fibroblasts exhibit morphological and physiological modifications.  Their shape changes from round to stellate or  bipolar (Figure 2.5) (Eastwood et al., 1996; Grinnell, 1994) and the variety of ECM molecules synthesized also changed (Rubin et al., 1995).  Fibroblast  17  sensitivity to growth factor stimulation was also altered when they were cultured on collagen lattices (Nakagawa etal., 1989). It has been demonstrated that gel contraction in vitro is modulated by some inflammatory factors such as PDGF (Platelet-derived growth factor) (which stimulates gel contraction) and prostaglandin E (which inhibits collagen gel 2  contraction). Therefore, events and mechanisms associated with collagen gel contraction may be of importance with respect to in vivo studies concerning aspects of inflammation (Rubin etal., 1995). Connective tissue ECM contraction by myofibroblasts is regulated in vivo by a number of growth factors and cytokines (substances related to the regulation of the immune system) that are produced locally during wound healing (Grinnell, 1994; Rubin et al., 1995). This would support the assumption that collagen gel contraction observed in vitro is analogous to the tissue contraction that occurs during wound healing in vivo and therefore constitutes a suitable model for in vitro study (Grinnell, 1994; Rubin ef al., 1995).  18  Stellate shape  Bipolar shape  Figure 2.5 Schematic presentation of a cell on a grid with a stellate and bipolar shape  When a cell has a stellate shape its extensions are on each side of the body whereas in a bipolar shape the cell is elongated and the extensions are concentrated at two extremities of the cell body.  2.4 Mechanical properties of collagen gels  Mechanical activity of cells plays an important role in the tissue organization (Delvoye etal., 1991; Grinnell, 1994). Contraction during the wound healing process is an example of mechanical function performed in part by the cells. In fibroblast populated collagen lattices, cells moving through the lattices were seen to mechanically rearrange the collagen network and in so doing contract the gel (Delvoye etal., 1991; Harris et al., 1981; Stopak etal., 1982). To move through the collagen lattices and contract the gel, cells attached to the matrix and spread across the gel. These attachments generate and subsequently maintain a significant endogenous tension in the matrix, i.e., a  19  tension created within the gel by the biological components of this gel (Brown et al., 1998; Delvoye et al., 1991; Eastwood et al., 1994). To clarify the process by which an endogenous tension is maintained in the matrix and to study the effect of mechanical stimulation on cells, Brown et a/(1998) used a tensioning CFM to apply specific tensile loads across fibroblast populated collagen lattices (Brown et al., 1998). With this equipment, the external loads were applied such that the total tension in the matrix was above or below the endogenous level. The cellular response to this external force was then observed by measuring changes in the matrix tension following the loading. Both single static loading and cyclical loading were used during mechanical testing. Independently of the nature of the test performed, results showed that cells reacted to modify the endogenous matrix tension in the opposite direction to externally applied loads.  That is,  fibroblast-mediated contraction decreased after an increase of the external loading and the contraction by fibroblasts increased after the external force was unloaded.  Therefore, according to this study, the maintenance of a stable  tension (tensional homeostasis) within the ECM could be the result of rapid and synchronous adjustments in endogenous contraction by fibroblasts acting against external forces. An endogenous tension has been observed in vivo in many connective tissues, such as smooth muscle and skin (Brown et al., 1998). The mechanical activity of fibroblasts and their contribution to matrix tension observed in collagen gels in vitro has been suggested to be analogous to that observed in the dermis of skin (Delvoye etal., 1991; Stopak etal., 1981; Vernon etal., 1992).  20  Application of uni-axial mechanical loads across collagen lattices has also been used to study the morphological changes of fibroblasts when the gel is stimulated mechanically (Eastwood et al., 1998).  In this study it was  demonstrated that fibroblasts respond to a mechanical load by realigning themselves in the direction of the maximum strain, i.e., the cells become oriented parallel to the axis of the strain with a bipolar morphology. These observations, showing that the alignment and shape of cells may be controlled by mechanical stimulation, could be useful for the study of fibroblast function in the development and repair of connective tissues architecture (Eastwood et al., 1998). Changes of fibroblast morphology have also been studied during stress relaxation in collagen gels, i.e., when the external load across the lattices is suddenly released (Eastwood etal., 1998; Grinnell, 1994; Mochitate etal., 1991; Tomasek et al., 1992).  In these experiments the collagen lattice remains  attached to the culture dish for a few days such that a force is created across the gel. During this period, fibroblasts attached to the matrix and become stellate in shape. When the edges of the gel are released from the culture dish, there is contraction of the gel and the morphology of fibroblasts changes from a large stellate shape to rounded cells with the retraction of pseudopodia and the depolymerization of actin bundles Since collagen lattices have been suggested to mimic the behaviour of the dermal connective tissue (Bell et al., 1979), the mechanical properties of collagen lattices, such as the force of isometric retraction (Delvoye et al., 1988)  21  and the stress-strain characteristics of the gel (Chapuis et al., 1992), have been studied for comparison to that connective tissue.  Using a uniaxial traction  apparatus, stress-strain curves have been obtained for fibroblast populated collagen lattices. The shape of the stress-strain curves obtained for collagen lattices was similar to the ones obtained for skin (Figure 2.2) (Chapuis et al., 1992). Effect of diverse pharmacological agents on the mechanical properties of skin is of interest for the dermatologists, physiologists and cosmetologists and therefore the possibility to measure the mechanical properties of collagen lattices is of significant interest.  2.5 The role of cells with respect to interstitial fluid pressure  In connective tissues, the space outside the vascular and lymphatic systems and the cells is called the interstitium. In this space, fluid, solutes and other dissolved constituents can be found. This tissue space plays an important role in homeostasis and in a number of pathological states. The interstitium plays a significant role in the transport of fluid and solutes between the capillaries, the tissue cells and the lymphatic system. Figure 2.6 summarizes the transcapillary-interstitial exchange system of a typical connective tissue (e.g. skin). Under normal conditions, the microvascular exchange system is assumed to be at or near steady-state (Bert et al., 1984). Stabilization of the interstitium has been assumed to occur in part through control of interstitial fluid volume. One factor that plays an important role in the control of interstitial fluid volume is the interstitial fluid pressure (P ). In skin, under normal conditions, P is slightly if  if  22  subatmospheric at approximately-1 mmHg (Reed etal., 1997). The relationship between the interstitial volume and interstitial fluid pressure is characterized as a compliance relationship. The term compliance is defined as the ratio between the change in interstitial fluid volume and the corresponding change in Pn.  Capillary Proteoglycan  I  Hyaluronan  Figure 2.6 Schematic drawing of the transcapillary-interstitial exchange system of a connective tissue  A capillary, a lymphatic and the major structural components of the interstitium (e.g. collagen bundles) are represented in this diagram. Two pressures act across the capillary wall: a colloid (e.g. plasma protein) osmotic pressure (COP) and a hydrostatic pressure (P). Subscripts c and i denote capillary and interstitium respectively. K and o are the capillary filtration coefficient and capillary reflection coefficient for proteins respectively. Reproduced from Reed etal., 1997. f  23  The volume-pressure relationship is such that when a tissue swells in response to some perturbation, its volume and hydrostatic pressure increase (in skin, a change in interstitial fluid volume of 14% will change interstitial fluid pressure by 1mmHg). This increased Pn will then act across the capillary wall to counteract further fluid filtration (Reed  etal.,  1997).  Recent studies have shown that Pit doesn't act only as a passive controller to maintain a constant interstitial fluid volume, but can also play an active role in respect to transcapillary fluid dynamics (Reed et al., 1997). In fact, in experiments it has been demonstrated that the rapid accumulation of fluid within the tissue (edema) observed in burn injuries and in several acute inflammatory reactions was created in part via an increased negativity of P . if  This change in fluid pressure was also investigated after mast cell degranulation (Koller et al., 1993) and when neurogenic inflammation occurs (Woie et al., 1993), and in both studies, an increased negativity of Pit was observed. Fibroblasts and structural components of connective tissues (e.g. collagen and elastic fibres) are thought to be involve in this increased negativity of Pn. The process by which the fibroblasts and structural components of connective tissues could alter P is suggested to be similar to the one observed in fibroblast if  collagen gel contraction. As seen in collagen lattices, it seems reasonable that fibroblasts in dermis might also perform a contractile function. Therefore, dermal fibroblasts could regulate P by exerting a tensile force on the collagen network if  which would restrain the swelling tendency of the hyaluronan-proteoglycan gel present in the connective tissues (Figure 2.7). The tensile force applied on the  24  collagen network is generated via integrin binding; particularly collagen-binding p 1 integrins.  The activity of these integrins can be modulated by cytokines,  growth factors and prostaglandins, to either increase or decrease the number of fibroblast-matrix attachments. Such changes in cell interactions would then lead to the compaction or swelling of the hyaluronan-proteoglycan gel, in turn affecting P and the water content of the tissue (Rubin et al., 1995). if  Fluid  N o r m a l situation  F l u i d influx  Figure 2.7 Schematic drawing of proposed events associated development of increased negativity of interstitial fluid pressure (P )  with  if  Acute inflammation and/or perturbation of integrin function (particularly p integrin) disrupt cellular attachments (e.g. adhesions between cells and fibres network as shown in this diagram) which allows expansion of the tissue and a decreased of Pif. Reproduced from Reed etal., 1997. r  25  3. Materials and Methods To  carry  out  the  biomechanical  experiments  a  stress-strain-time  apparatus was used. The principal components of this apparatus as well as the electronic system are described in this chapter.  For this project, experiments  were carried out on skin samples from male rats.  The procedure utilized to  harvest the skin from the animal is presented followed by a description of the biomechanical experiments performed on the tissue samples.  As previously  mentioned, transmission electron microscopy was used to study the changes in fibroblast morphology.  Before a tissue sample can be analyzed under the  electron microscope, it is submitted to diverse chemical treatments. After a brief description of the principal steps involved in the processing of samples to be used  in the electron microscope is presented, this chapter ends  with  a  presentation of the way morphometric analyses were done using the electron micrographs.  3.1 Stress-Strain-Time apparatus A stress-strain-time apparatus (Wang, 1997) was used to stretch skin samples.  Both static and dynamic tests can be performed with this apparatus.  Only the basic features of the apparatus will be described in this chapter.  3.1.1  Basic components of the stress-strain-time apparatus The figure below (Figure 3.1) presents the basic components of the  equipment. 26  Figure 3.1 Basic components of the stress-strain-time apparatus.  Clips to hold the tissue sample (#1 and #2), transducer (#3), moveable table (#4), stepper motor (#5) and bath (#6).  The tissue sample is held vertically by two stainless steel clips (#1 and #2 in Figure 3.1). The curvature of those clips allows a good grip on the tissue and therefore prevents it from slipping. The upper clip (#1 in Figure 3.1) is fixed and connected to a force transducer (#3 in Figure 3.1) which can support a maximal force of 25 grams (model: MBL/5514-02, Sensotec, Ohio). To ensure that a higher force is not applied to the transducer during the mounting of the tissue sample, a locking device is incorporated into the transducer support. With this mechanism in place, all the imposed force is carried by the locking device  27  instead of the transducer. The lower clip (#2 in Figure 3.1) is attached to the opposite end of the rectangular piece of tissue.  It is also connected to and  moved by a motorized linear position table (#4 in Figure 3.1) whose position is controlled by a stepper motor (#5 in Figure 3.1). To ensure the viability of the cells during biomechanical testing, the tissue, which is mounted between the clips, is surrounded by a moveable tissue bath (#6 in Figure 3.1) in which a physiological solution can be circulated. The bath, which can slide up and down along a fixed stainless steel rod facilitating the mounting of the tissue, has glass windows on its sides to allow observation of the tissue during experiments. To maintain the physiological solution at 37°C, a heater cartridge is embedded in one of the walls of the bath and controlled electrically. To avoid physical disturbances of the tissue in the bath, physiological solutions are bubbled with 95% oxygen (0 ) and 5% carbon dioxide (C0 ) in a reservoir 2  2  separated from the bath instead of being oxygenated directly in the bath. The reservoir is connected to an opening at the bottom of the bath by plastic tubing and is elevated on a shelf such that the solution can be fed by gravity. The circulation of the solution in the bath is controlled by a vacuum at the top of the outlet of the bath:  in this manner the pH of the bathing solution can be  maintained at 7.4-7.5 in the bath. An isolation table, made from steel tubes and plates, is used to support the entire mechanical apparatus. The presence of tennis balls between the top  28  plates and base legs reduce the vibration leading to a reduction of noise in the tension signal sensed by the transducer.  3.1.2 Electronic and control components of the stress-strain-time apparatus A diagram of the complete system for biomechanical experiments is presented in Figure 3.2. One of the most important pieces in this system is the stepper motor, which controls the movement of the linear table. Digital pulses are used to operate the motor. These pulses are generated by the conversion of the amplitude and/or frequency of the forcing function (e.g. sinusoidal function) to pulse number or time interval between the pulses. Once the pulse calculation is done, it is downloaded through an lO-Tech General Purpose Instrument Bus (GPIB) (in the microcontroller) to a position indexer (compumotor) and then read by the motor that is programmed to move according to the pulse calculation. The motor is programmed such that it can either move the linear table from one position to another along a straight line, or oscillate the linear table at each position according to sinusoidal functions, or move the linear table to generate a single triangular pulse.  29  Motor  Clips  Moveable table Vr<*-- Kreb's reservoir Temperature Controller  IBM Compatible  Manual control panel  Figure 3.2 Complete system The principal components of the system used for the project are presented in this figure. The apparatus is controlled by a PC (IBM). A microcontroller and compumotor are used to control the movement of the stepper motor which moves the linear table and at the same time the lower clip that hold the tissue sample. In case of emergency the motor can be stopped manually using the manual control panel connected to the microcontroller. The force transducer is connected to the PC such that the tension signal can be recorded and the temperature of the bathing solution is kept at 37°C by a temperature controller.  For each pulse sent to the motor, the table will move for a specific incremental distance determined by the user of the apparatus while respecting the limitations of the equipment. For a sinusoidal test, the smaller the increment is (minimum of 0.25um), the smoother will be the sine waves, but, for a given frequency, the maximal amplitude will be less. The time required by the motor to change direction causes this limitation in the combinations of frequency and amplitude of the position forcing function. In this project, an increment of 2.5um 30  was used.  This step size provides a relatively smooth sine wave with a  maximum amplitude of 800um that can attain a frequency of 1 Hz. To control the direction of the table motion, a level sensitive direction digital input is placed on the motor driver. Figure 3.3 shows how a saw-tooth motion could be obtained using the digital pulses and the level direction:  Table position  Index pulse  Direction  Figure 3.3  Saw-tooth motion produces by digital pulses and level direction  The table position is controlled by the index pulse and the level direction to provide the desired forcing function. In this figure, the table is controlled to produce a saw-tooth motion. Reproduced from Wang (1997).  The  parameters specific for each experiment, such as the rate and  amplitude of the movement, are controlled and sent to the motor via a PC. The software used with the PC is a program developed in Visual Basic Professional Version 3.0 (Wang, 1997), with a user friendly interface that facilitates the operation of the apparatus. Starting up and zeroing the equipment, initiating the calibration routine and adjusting calibration factors, downloading waveforms to the controller, on line graphing of the tension and position signals, and high 31  sampling frequency data are some of the functions that can be performed using this software program (Wang, 1997). During an experiment, the PC allows the display and recording of the motor position and the tension of the tissue sample sensed by the transducer. Both signals are read at the same set intervals (sampling frequency), specified at the beginning of the experiment by the user. To obtain digital position signals that can be read by the data acquisition system, a 12-bit up-down counter is used.  For the reading of the tension coming from the transducer, an  analog/digital (A/D) converter is inserted in the system. By using an lO-Tech GPIB interface between the position indexer and the PC, the data acquisition and the control of the motor can be done conveniently by the same computer. A manual control panel is connected to the microcontroller to manually control the motor movement at three different speeds. An emergency button is incorporated into this panel to stop the motor in case of emergency.  3.1.3 Linearity of the transducer To ensure an accurate reading of the tension the transducer must be calibrated.  For the calibration, standard weights are hung to the upper clip of  the apparatus and the tension sensed by the force transducer is compared to the actual weight. The results of calibration tests showed an excellent linearity of the transducer; the coefficient of determination or R value (R =1-SSE/(SSR+SSE), 2  2  with SSE: error sum of squares and SSR: regression sum of squares) 32  corresponding to the linear regression of the calibration data was about 0.99948±0.00060. An example of calibration result is shown in Appendix A.  3.2 Transmission electron microscope  Transmission electron microscopy was necessary to monitor the changes in fibroblast morphology.  The transmission electron microscope used was a  Philips 400, operated at 80kV.  3.3 Experimental Procedures  3.3.1 Equipment and solutions preparation Before each experiment, the calibration of the transducer was performed to ensure an accurate reading of the tension during testing. Standard weights were used to calibrate the force transducer (see Appendix A). The physiological solution used for this project was an altered version of the classic Kreb's solution. The Kreb's solution was modified to eliminate the Ca  ++  ion, which may facilitate the denaturation of collagen in the tissue and thus  may affect the mechanical properties of the sample (Wang, 1997).  The  ingredients of the altered Kreb's solution are presented in Table 3.1. The pH of the solution was adjusted to 7.4 - 7.5 using hydrochloric acid (HCI) and sodium hydroxide (NaOH). Once the solution was made, it was poured into the reservoir connected to the bath of the stress-strain-time apparatus and purged through the tubing connections. The solution in the reservoir was then bubbled with 95% 0  2  33  -5% CO2 and the temperature controller of the bath was turned on to maintain the Kreb's solution at 37°C.  Table 3.1 Ingredients of the altered Kreb's solution (g/l)  NaCI  KCI  6.9  0.35  MgS0 .7H 0 0.29 4  2  KH.P0 0.16  4  NaHC0 2.1  3  Glucose 2.0  An additional 500 ml of Kreb's solution was set aside in which 50mg of compound 48/80 (C48/80, Sigma Chemical Co., St-Louis, MO, USA) was dissolved. A concentration of 0.1 mg/ml (0.1 mg of C48/80 dissolved in 1 ml of Kreb's solution) was chosen for this project according to previous studies (Koller etal., 1993; Woie etal., 1993). Additional Kreb's solution was set aside for soaking of tissue samples.  3.3.2 Harvesting tissue samples Wistar-Moller male rats weighing 200-250g (6-8 weeks old) were used for this project. The rats were anesthetized by halothane inhalation. After verifying that the animal was not subject to pain anymore the back of the rat was shaved. Since skin is naturally under tension, the length of the sample to be used in the biomechanical experiment was measured, before the skin was harvested. Indian ink was used to mark the places on the skin where the clips of the stress-strain apparatus were to be attached. This was done to ensure that the tissue sample  34  was at its natural tension at the baseline condition for experiments.  A line  parallel to the body axis of the animal was also drawn with ink to serve as a guide for consistently stretching the tissue in the same physiological orientation (because skin is anisotropic, Edward etal., 1995, Vogel, 1994). To facilitate access of the fixative to the dermis of skin to be used the fat layer under the dermis was gently removed with a scalpel while the skin was being harvested. On the other hand, the epidermis was left intact since it is a thin layer difficult to remove. After a 1x2 cm piece of skin was excised, the opening on the back of the animal was stitched closed to limit the loss of blood. The rat was kept under anesthesia so that a second piece of tissue could be taken from the same animal after the first biomechanical experiment was performed.  With this  procedure, fewer animals were used for the project, the cells viability was assured in the second piece of skin and each rat provided tissue samples for each of the treatment groups, which removes a factor of error in the comparison of results. After the second piece of skin was harvested, the rat, still under anesthesia, was killed by carbon dioxide asphyxiation. All of the experiments performed in this study were approved by the Animal Care Committee of the University of British Columbia (Vancouver, BC, Canada).  35  3.3.3 Treatment groups From the first piece of skin harvested from each animal, a sample (about 10mm x 3mm) was cut and fixed immediately in 4% glutaraldehyde. To improve the fixation, the tissue sample was stretched and pinned with needles in wax dishes while immersed in glutaraldehyde. This specimen was later considered as the virginal control for all of the morphometric analyses. The remainder of the first piece of skin from each animal, as well as the second piece, were cut in two.  One part was used for the biomechanical  experiment in which the skin was stretched using the stress-strain-time apparatus described previously. The other half of each sample was immersed in the same solution used for the biomechanical experiments, i.e. Kreb's buffer or Kreb's buffer containing C48/80 (0.1mg/ml), and for the same amount of total time (90 minutes) as for the tissue subjected to biomechanical experiments. The soaking was done in plastic dishes containing wax at the bottom. To facilitate the penetration of the solution into the tissue, the skin was stretched slightly while pinned with needles in the wax. To preserve viability of cells, the solution in the dish was bubbled with 95% 0  2  and 5% C 0  temperature was maintained at around 37°C.  2  (Wang, 1997) and the  The same environmental  conditions were applied to the skin sample subjected to biomechanical experiments, i.e., the bathing solution was oxygenated and kept at 37°C. Once the biomechanical experiment was done, both the piece of skin used in mechanical test and the skin only immersed in the dish, were fixed in 4% glutaraldehyde. All of the tissue samples were stretched and pinned in wax  36  dishes while fixing in glutaraldehyde. All tissue samples were fixed within 120 minutes of harvesting: 90 minutes to carry out an experiment (either biomechanical experiment or static soaking in buffer) and about 30 minutes of manipulations. The experimental procedure described above, produced five treatment groups for which tissues were obtained from each animal: 1) virginal control, 2) skin soaked in Kreb's buffer, 3) skin soaked and stretched in Kreb's buffer, 4) skin soaked in C48/80 in buffer (0.1 mg/ml) and 5) skin soaked and stretched in C48/80 in buffer (0.1 mg/ml).  3.3.4 Qualitative biomechanical experiments A tissue sample was cut from the piece of skin previously harvested with dot and arrow markings indicating normal in vivo dimensions and orientation of the body. The sample was mounted between the clips of the stress-strain-time apparatus such that it was stretched longitudinally to the body axis of the animal. After the mounting of the sample, the dimensions of the tissue were taken using an optical micrometer and a caliper.  The specimen of skin used for the  biomechanical experiments was about 2 mm wide, 1 mm thick and 4 mm long (length between the two clips which was equivalent to the length marked on the back of the animal). In most biomechanical studies, the tissue is preconditioned prior to any measurements. Repeated loading and unloading account for the preconditioning of a tissue sample. Theoretically, the load-extension curves should shift to the 37  right during roughly the first three cycles and then become repeatable (Fung, 1994; Silver etal., 1987). Preconditioning is usually performed in order to obtain reproducible experimental results. In life, skin is not preconditioned before it is submitted to any tensional forces. Therefore, to obtain more representative results of how the cells react when real skin is stretched, the preconditioning of the skin samples was omitted in this project. However, during the first 5 to 10 minutes following the mounting of the sample the tissue undergoes relaxation, i.e., the tissue tension decreases with time and therefore no experiment was performed until the tension reading stayed constant for about 10 seconds. Sine wave testing was chosen as the strain driving force for the skin samples.  All the experiments were performed at a frequency of 1 Hz (1  sinusoidal cycle/second), but the amplitude of the sine wave (which corresponds to the stretching elongation) was varied depending on the amount of extension desired. As previously mentioned (section 2.2), under normal in vivo conditions skin is usually subjected to small extensions. Therefore, we decided that the strain during mechanical experiments should not exceed 20%. Each mechanical experiment consisted of a sequence of 40 sine waves performed each 15 minutes for a period of 90 minutes. Consequently, a period of rest of 14.33 minutes was used between each sequence of sine waves. The tension and strain perturbation applied on the tissue sample during the stretching was recorded on the PC and analyzed subsequently. For each sinusoidal cycle 50 data points were recorded to ensure an accurate sinusoidal tracing.  38  3.3.5 Processing tissue for transmission electron microscopy A standard procedure was followed to process the tissue for transmission electron microscopy (Glauert, 1975). glutaraldehyde  in 0.2M  Sodium  After an overnight fixation in 4%  Cacodylate  buffer (pH=7.34) at room  temperature, the tissue samples were cut into pieces averaging 1 cubic millimeter using double-edged razor blades. These pieces were then washed three times in 0.1 M Sodium Cacodylate buffer and then post-fixed in 2% osmium tetroxide in 0.2M Sodium Cacodylate buffer at pH=7.34 at room temperature for 1 hour. After washes in distilled water, the tissues were dehydrated through a grade series of alcohol. Infiltration and embedding were done in Epon (PELCO Eponate Kit, Ted Pella Inc., CA, USA). All pieces were embedded and sectioned such that a cross section of skin perpendicular to the epidermis was observed in every section in the electron microscope. Thin sections, 0.06um to 0.07pm thick, were obtained from blocks and picked up on 200 mesh copper grids. The sectioning was done using an RMC MT6000-XL ultramicrotome with assistance from Ms Fanny Chu. Sections were stained with uranyl acetate and Reynold's lead staining.  3.4 Morphometric analyses  3.4.1 Micrographs From each grid 5 negatives of dermal connective tissue containing fibroblasts were made in a systematically random fashion at a magnification of 39  X1650. All negatives were printed at a final magnification of X4125 on 8x10" photographic paper. To ensure systematic random sampling of the dermis connective tissue, sections were picked up on the grids with no specific orientation and a micrograph was taken at each third grid square that contained connective tissue. For example, if each number on the mesh grid of Figure 3.4 represents a grid square with connective tissue a micrograph would be taken at #1, #4, #7, etc. Some fields, represented by unnumbered gray grid squares in Figure 3.4, were not utilized because of the presence of other tissues such as blood vessels, hair follicles, subaceous glands or large areas of section damage. With this method only dermal connective tissue was sampled in a systematically random fashion through the depth of the dermis.  /  /  1  A  A  2 3 6 8 10  7  /  4 5  \  K  \  \ \  9  11 12  \  13  \  14 15 16  \  /  I 1  N  Figure 3.4  Mesh grid with tissue section  The gray area represents the tissue section on the grid. Numbered small squares designate fields of dermal connective tissue that could be used for micrographs because they do not contain other tissues (e.g. blood vessels, glands) or large areas of section damage. 40  3.4.2 Point counting To analyze the relative surface area of fibroblast cytoplasmic extensions and cell bodies, point counting (Reith etal., 1988) was done on the micrographs. For that purpose, a transparency containing dots and lines was placed over the micrographs as shown in Figure 3.5. The lines on the transparency were 8 mm long and had one dot of 1 mm diameter at each end. All the lines (including the dots) on the transparency were 7 mm apart in all directions from each other.  2.7 um  Figure 3.5 Point counting This figure provides an example of how point counting was performed using micrographs and a transparency on which lines and dots are printed. Examples of line intercepts are denoted by white lines for fibroblast body and lines indicated by emptied arrowheads (<—) for fibroblast extension. Examples of hits are denoted by white dots for fibroblast body, dots indicated by filled arrowheads (*~) for fibroblast extension, dots surrounded by squares for collagen fascicles and dots surrounded by circles for interstitial fluid space.  41  The text below Figure 3.5 gives an example of how the point counting was performed. For the purpose of this study, both the line intercepts and the dots (or hits) were counted with respect to the transparency and the micrographs. An intercept was counted when a line on the transparency crossed a fibroblast cell body or a fibroblast cytoplasmic extension. To differentiate between a fibroblast extension and a fibroblast body we used the rule that if a profile of fibroblast cytoplasm was smaller than a mitochondrion it was counted as a cytoplasmic extension and if it was bigger it was counted as a cell body. If no mitochondrion was present, the distinction between fibroblast extension and fibroblast body was left to the judgement of the investigator. The line intercepts data were used to calculate a surface density ratio, defined as the number of intercepts with fibroblast cytoplasmic extensions to the number of line intercepts with the plasma membranes of fibroblast cell bodies. If retraction of fibroblast cytoplasmic extensions occurs in skin subjected to stretching and/or chemical disruption (C48/80) of fibroblast-fibroblast and fibroblast-ECM contacts, then the surface area occupied by cell extensions should decrease relative to the number of intercepts with fibroblast cell body plasma membranes, and thus a lower surface density ratio should be measured in groups where skin was stretched. The number of dots was counted for several components of the connective tissue including fibroblast bodies, fibroblast extensions, collagen fascicles, spaces filled with extracellular fluid and cells other than fibroblasts. The relative proportion of each component in one micrograph was calculated by  42  dividing the number of hits calculated for the tissue component by the total number of hits counted for the whole micrograph. The number of hits and the relative proportion of each element were used only as additional information; no results were obtained from these data.  3.4.3  Roundness of fibroblasts To test the hypothesis that a disruption of fibroblast adhesions might  result in the rounding up of fibroblasts as cytoplasmic extensions are withdrawn, the width/length ratio was calculated for the fibroblasts observed in all micrographs as in Figure 3.6. The closer this ratio comes to one, the more round is the fibroblast; a value of one indicates a circle. The width/length ratio was calculated as follow; the longest dimension of the fibroblast body was measured and set as the denominator of the ratio (i.e. length of the fibroblast) (denoted by B in Figure 3.6) with the largest dimension perpendicular to that length taken as the ratio's numerator (i.e. width of the fibroblast) (denoted by A in Figure 3.6). In this ratio calculation the fibroblast extensions were excluded using the criteria outlined previously. That is the mitochondrion rule, as explained for the point counting, was used in the selection of widths and lengths to calculate the ratio and thereby estimate the roundness of the cell bodies.  43  1.5 nm  Figure 3.6 Measurement of fibroblast width/length ratio  To calculate the width/length ratio the longest dimension of the fibroblast body (denoted by the letter B) was measured as well as the largest dimension perpendicular to that first measurement (denoted by the letter A). The width/length ratio corresponded to the division of the largest dimension by the longest dimension, i.e., A/B for the fibroblast (F) of this figure.  3.4.4 Statistical Analyses The surface density ratios and width/length ratios presented in this thesis are given as means ± standard deviation. A paired t-test was used to compare the ratios of the virginal control group to the tissues soaked in plain Kreb's buffer 44  (without addition of C48/80 in buffer).  The sample size for this t-test  corresponded to the total number of ratios calculated for each of the two treatment group, i.e., virginal control and skin soaked in Kreb's buffer. For the surface density ratio 45 ratios were obtained per treatment group, i.e., n=45. The number of fibroblasts on each micrograph was different and thus the sample size for the t-test on the width/length ratio was not the same for both treatment group; 74 ratios were obtained for the virginal control compared to 101 ratios for tissues soaked in Kreb's buffer. The null hypothesis set for the t-tests was that there was no difference between the average surface density ratio or width/length ratio of the virginal control and the ratio of the tissue soaked in Kreb's buffer. If there was a significant difference (p<0.05) between the virginal control group and the skin soaked in Kreb's buffer for one of the two ratios (i.e. surface density or width/length ratio), the data from the virginal control group were not used for further comparisons; the data for Kreb's soaked tissue were then used as the control group for the other statistics. A significant difference between the virginal control and skin soaked-only in Kreb's indicates that soaking affects the cells morphology and therefore, since soaking is present in all the treatment groups, excepted in the virginal control, it is more appropriate to use tissue soaked-only in Kreb's buffer as the control. To study statistically the effect of stretching and the effect of C48/80 on the relative surface area of fibroblast extensions and cell bodies and on the roundness of fibroblast bodies, a two-way randomized Block ANOVA test was used. The ANOVA analysis indicated that the variability between the tissue  45  samples was greater than the variability between the rats. Therefore, the tissue sample was chosen as the blocking factor, i.e., the "experimental unit" for statistical analyses. Since there were 3 rats and 3 tissues per rat, the sample size was 9 (n=9) for all the statistics obtained from an ANOVA test. For the effect of stretching, the null hypothesis chosen for this test was that there was no difference between the ratio of skin that was stretched while in a bathing solution and skin that was only soaked in the same solution for the same period of time.  For the C48/80 effect, Kreb's soaked tissues were  compared to C48/80 soaked tissues with the hypothesis that there was no difference between the two groups. The level of significance for the statistics was p<0.05. Details on the statistical analysis are presented in Appendix C.  46  4. Results The initial aspect of this project was involved with the assessment of tissue fibroblast viability following soaking in buffer. The occurrence of mast cell degranulation in the skin in response to C48/80 dissolved in Kreb's buffer was also investigated.  Once these matters were verified, morphological and  qualitative biomechanical experiments were conducted. The results obtained from the morphometric analysis of the relative surface area of fibroblast cytoplasmic extensions and cell bodies as well as the results of the roundness assessment of fibroblast cell bodies are presented in this chapter.  The tension of tissue samples recorded during mechanical  perturbation experiments is also shown as a function of sinusoidal cycle number for skin soaked both in Kreb's buffer and skin soaked in Kreb's buffer containing C48/80. In this chapter is also presented the results for sinusoidal mechanical stimulation of frozen tissues and morphological results obtained for fresh skin subjected to a varying number of sinusoidal perturbations.  4.1 Preliminary experiments  4.1.1 Verification of fibroblast viability Before any mechanical experimentation was begun, it was important to determine whether cells of the dermal connective tissue were still alive after soaking for up to 90 minutes, the time needed to carry out a typical mechanical experiment.  This was verified by the transmission electron microscope 47  examination of fibroblasts in the dermis of soaked skin. Figure 4.1 shows an example of a typical fibroblast observed in the dermis of a skin sample soaked in Kreb's buffer.  A continuous plasma membrane, the presence of normal  mitochondria, a non swollen rough endoplasmic reticulum (RER) and an intact nucleus are some of the ultrastructural characteristics indicative of the viability of the fibroblasts at the end of the 90 minutes buffer incubation when the tissues were fixed.  Figure 4.1 An apparently normal dermal fibroblast observed in a skin sample soaked in Kreb's buffer for 90 minutes before fixation  Intact plasma (PM) and nuclear (NM) membranes and a non swollen rough endoplasmic reticulum (RER) (difficult to see at this low magnification) are some of the characteristics of the healthy fibroblast shown in this figure. The cell's nucleus is denoted by the letter N.  48  The examination of micrographs taken during the realization of this project showed fibroblasts in good condition in the virginal control group and the soakedonly skin samples (with or without the presence of C48/80), but a significant number of dying or dead cells were seen in skin samples that had been stretched. Two distinctive processes of cell death appeared to occur in stretched skin: necrosis and apoptosis.  Environmental perturbations (e.g. changes in  temperature, hypoxia, exposure to toxins) are often the cause of necrosis whereas apoptosis involves intracellular stimuli (Wyllie et al., 1980). A swelling endoplasmic reticulum, the degradation of the cytoplasm and the rupture of plasma and nuclear membranes are some of the morphological changes observed in a cell undergoing necrosis (Figure 4.2A).  Apoptosis is also  characterized by a morphological transformation of the cell, including plasma and nuclear membrane convolution, cell shrinkage, and chromatin condensation and fragmentation (Figure 4.2B).  49  A  1.4 (am  B  2.3 nm  Figure 4.2 Fibroblast undergoing necrosis (A) and apoptosis (B) in skin samples stretched with a strain of 2 0 % in Kreb's buffer  The two fibroblasts presented in this figure are dying following two distinctive cell death processes: necrosis (A) and apoptosis (B). In a necrotic cell (A) the nuclear membrane (NM) is kept intact until cytoplasmic degradation advanced, whereas in an apoptotic cell there is progressive convolution of the nuclear membrane and chromatin condensation (CC).  4.1.2 Verification of mast cell degranulation C48/80 is a compound that causes mast cell degranulation. To make sure that the compound was efficacious, that the chosen concentration was high enough and that the soaking was long enough, to provoke mast cell degranulation, micrographs of mast cells have been taken from a skin sample soaked in C48/80 in buffer for 90 minutes (Figure 4.3B).  For comparison  purposes, micrograph of a mast cell from a skin sample fixed immediately following removal from the animal (virginal control) is also shown (Figure 4.3A). The granules outside the cytoplasm seen in Figure 4.3B is a characteristic of  50  mast cell degranulation. It was concluded that the experiments involving C48/80 caused mast cell degranulation in the tissue samples.  2.3 nm  A  B  1.9 ixm  Figure 4.3 Mast cell and mast cell degranulation  A normal mast cell from a fixed virginal control skin sample (A) and a degranulating mast cell from a fixed skin sample soaked in C48/80 in buffer (0.1 mg/ml) (B) are shown in this figure. The granules outside the cell's plasma membrane in figure B (arrowheads) indicate mast cell degranulation.  4.2 Morphological Results  Because of the large amount of work required to do a morphological analysis using transmission electron microscopy (TEM), only 3 Wistar-Moller male rats were studied for this part of the project. According to the experimental procedure described in chapter 3, each rat provided 5 differently treated skin samples which were virginal control, tissue soaked in Kreb's buffer, tissue 51  soaked and stretched in Kreb's buffer, tissue soaked in C48/80 in buffer (0.1 mg/ml) and tissue soaked and stretched in C48/80 in buffer (0.1 mg/ml). From each of these treatment groups, 3 tissue sections for TEM were obtained. Since 5 negatives were made for each section (see section 3.4.1 for details), 15 micrographs were printed for each group. Therefore, for each 3 animals a total of 45 micrographs were analyzed for each treatment group.  4.2.1 Fibroblast micrographs for each group To analyze the changes in fibroblast morphology in the different treatment groups, TEM was used and micrographs of samples from each group were printed. Figure 4.4 provides a typical example of fibroblast morphology observed for each treatment. The principal visual morphological difference noticed on these micrographs resides in the presence or absence of cytoplasmic extensions and the overall shape of the cells. In the virginal control (Figure 4.4A), skin soaked in Kreb's buffer (Figure 4.4B) and skin soaked in C48/80 (Figure 4.4D) fibroblasts have long cytoplasmic extensions, but these extensions seem to disappear when the skin is stretched (Figures 4.4C and 4.4E).  As for the  appearance of fibroblast bodies, a round shape is observed in stretched skin samples (Figures 4.4C and 4.4E) whereas elongate bodies are found in the other treatment groups (Figures 4.4A, 4.4B and 4.4D).  52  B  1.8 nm  Q  1.9 nm  1.9 nm  1.7 nm  Figure 4.4 Fibroblasts from each treatment group  Fibroblasts from the virginal control group (A), soaked in Kreb's buffer (B), soaked and stretched in Kreb's buffer (C), soaked in C48/80 in buffer (0.1 mg/ml) (D) and soaked and stretched in C48/80 in buffer (0.1 mg/ml) (E). Fibroblast bodies (*) and fibroblast cytoplasmic extensions (arrowheads) are indicated.  53  4.2.2 Surface density ratio To quantify the difference between the fibroblast morphology observed in the different treatment groups, the relative surface areas of fibroblast cytoplasmic extensions and cell bodies were analyzed. To perform such an analysis, a surface density ratio was calculated using point counting analysis (details in Chapter 3). The surface density ratio is the division of the number of line intercepts for the fibroblast cytoplasmic extensions by the number of line intercepts for the cell bodies. Thus, the smaller the ratio is the smaller is the surface occupied by the fibroblast cytoplasmic extensions compared to the surface area of cell bodies. Examples of number of line intercepts and number of hits counted for skin samples of different treatment groups are presented in Appendix B. Figure 4.5 shows the mean of the surface density ratio for each treatment group. An average ratio of 0.82±0.38 was measured for the virginal control group and the paired t-test indicated that this ratio was significantly smaller (p<0.001) than the 1.45±0.69 ratio obtained for skin soaked in Kreb's buffer. A ratio of 1.65±0.99 was calculated for skin soaked in Kreb's buffer containing C48/80 (0.1 mg/ml). The ANOVA analysis used to study the effect of C48/80 on the surface density ratio resulted in no significant difference (p=0.17) between the ratio of skin soaked with or without addition of C48/80 in Kreb's buffer. Two ways ANOVA also indicated that the ratios for skin stretched with or without addition of C48/80 in Kreb's buffer were significantly lower (p<0.001) than the ratios for skin soaked-only in the same bathing solution. The surface  54  density ratios calculated for the skin stretched in the absence and presence of C48/80 were respectively 0.04±0.07 and 0.06±0.11. Therefore, the surface area covered by the fibroblast cytoplasmic extensions appears to be less in tissue samples that have been subjected to mechanical stimulation.  .2  2  5  2  >.  2  I£  1-5  t  CO  1  1  0.5  Virginal  Soaked  Soaked +  Soaked  Control  in Kreb's  stretched  inC48/80  in Kreb's  Soaked + stretched inC48/80  Treatment group Figure 4.5 Surface density ratio  The value of the surface density ratio depicted in this figure for each treatment group correspond to the arithmetic mean of 45 ratios. The error bars represent the positive standard deviation.  4.2.3 Width/length ratio Width/length ratios of fibroblast bodies have been calculated to analyze the "roundness" of cells. Since the width is always smaller than the length, ratio values will always be between 0 and 1. The average ratio for each treatment group is presented in Figure 4.6. The paired t-test used to compare the virginal  55  control to tissues soaked in Kreb's buffer indicated that the two average ratios, which are respectively 0.30±0.13 and 0.31 ±0.10, were not significantly different (p=0.27). According to ANOVA analysis, significantly (p<0.001) higher ratios were measured for the groups in which the skin was stretched at 20%, in the absence (0.62±0.17) or presence (0.61 ±0.15) of C48/80, compared to skin soaked-only in the same bathing solution (0.31 ±0.10 without C48/80 in buffer and 0.32±0.11 with C48/80 in buffer). Therefore, fibroblast bodies apparently round up when skin is sinusoidally stretched.  0.8 0.7  o  0.6 0.5  c  0)  I ^  0.4 0 3  0.2 0.1 0 Virginal  Soaked  Soaked +  Soaked  Control  in Kreb's  stretched  in C 4 8 / 8 0  in K r e b ' s  Soaked + stretched in C 4 8 / 8 0  Treatment group  Figure 4.6 Fibroblasts width/length ratio  The value of the width/length ratio depicted for each treatment group correspond to an arithmetic mean. Since there was a different number of fibroblasts on each micrograph the number of ratios used for the mean of each group varied as follow: virginal control: 74, soaked in Kreb's: 101, soaked and stretched in Kreb's: 77, soaked in C48/80: 86, soaked and stretched in C48/80: 72. The error bars represent the positive standard deviation. 56  No statistical difference (p=0.94) was observed between the width/length ratio of skin soaked-only in plain Kreb's buffer (0.31 ±0.10) and skin soaked-only in presence of C48/80 (0.32±0.11). Thus, no apparent effect of C48/80 on the width/length ratio can be determined.  4.2.4 Fibroblast morphology of skin stretched 1 and 4 sequences of sinusoidal waves In the experiments described previously, skin was stimulated mechanically seven times by a series of 40 sine waves before it was fixed in glutaraldehyde. To determine if the retraction of cytoplasmic extensions was progressive, skin samples were fixed in individual pilot studies after the first sequence of 40 sine waves and after the fourth sequence. These experiments were done only in plain Kreb's buffer (i.e. without addition of C48/80 in buffer). Micrographs were printed for these tissue samples and surface density ratios as well as width/length ratios were calculated. Two rats were used for skin stretched during 1 sequence and only one rat was used for skin stretched for 4 sequences. A total of 10 micrographs were made for skin stretched during 1 sequence and 6 micrographs for skin stretched during 4 sequences.  Because of the small  number of animals and micrographs used in these additional experiments, no statistical analysis was performed on the data. The average surface density ratios calculated for skin stretched for one and four sequences of sine waves are shown in Figure 4.7. From one animal, a tissue sample was only soaked in Kreb's buffer to serve as control data (3 57  micrographs were printed). The average surface density ratio corresponding to this group of treatment also appears on Figure 4.7 (Soaked-only in Kreb's). For comparison purpose, the average ratio of skin stretched seven times in Kreb's buffer presented in Figure 4.5 is depicted again in Figure 4.7. As indicated by the graph of Figure 4.7, a ratio of 1.88+1.15 was calculated for skin soaked in Kreb's buffer, which is higher but in the same order of magnitude than the 1.45±0.69 surface density ratio presented previously for skin subjected to the same treatment (Figure 4.5). The surface density ratio of skin stretched for only one sinusoidal sequence (0.40±0.15) ranges between the ratio of skin soaked in Kreb's buffer (1.88±1.15) and the ratio of skin stretched for four sequences (0.16±0.14).  This later is slightly higher than the ratio  measured for seven sinusoidal sequences (0.039±0.070). According to these preliminary results, the surface area covered by fibroblast cytoplasmic extensions decreases rapidly during the first stretching sequence followed by a progressive retraction of cell extensions as the number of sinusoidal sequences increases.  58  3.5 3 o  f 2.5  1 o u  2  1.5  CO  0.5 Soakedonly in Kreb's  Stretched in Kreb's, 1 sequence  Stretched in Kreb's, 4 sequences  Stretched in Kreb's, 7 sequences  Treatment group  Figure 4 . 7 sequences  Surface  density  ratio for various  number  of  sinusoidal  Average surface density ratio of skin samples fixed after 1 (n=10), 4 (n=6) and 7 (n=45) sinusoidal sequences. An average ratio is also shown for a skin sample only soaked in Kreb's buffer (n=3). The error bars represent the positive standard deviation. Note: the n in the caption indicates the number of ratio values used for the arithmetic mean of each treatment group ratio.  The average fibroblast width/length ratio for Kreb's soaked-only tissue and for skin stretched for one, four and seven (data from Figure 4.6) sinusoidal sequences are presented in Figure 4.8. These width/length ratios were obtained using the same micrographs used for the calculation of the surface density ratio (Figure 4.7).  59  0.8 0.7 ,2 0.6 0.5 TO C CD  •o  0.4 0.3 0.2 0.1 0 Soakedonly in Kreb's  Stretched in Kreb's, 1 sequence  Stretched in Kreb's, 4 sequences  Stretched in Kreb's, 7 sequences  Treatment group  Figure 4.8 Width/length ratio for various number of sinusoidal s e q u e n c e s  Average width/length ratio for skin samples stretched for 1 (n=12), 4 (n=7) and 7 (n=77) sinusoidal sequences. An average ratio is also shown for a skin sample soaked-only in Kreb's buffer (n=6). The errors bars represent the positive standard deviation. Note: the n in the caption indicates the number of ratio values used for the arithmetic mean of each treatment group ratio.  A width/length ratio of 0.30±0.12 was calculated for the Kreb's soakedonly skin, which is very close to the 0.31 ±0.10 ratio obtained in the previous results for Kreb's soaked-only tissue (Figure 4.6). According to the graph presented in Figure 4.8, the average width/length ratio of skin soaked in Kreb's is lower than the ratio of stretched skin (independent of the number of sinusoidal sequences) which is in agreement with the previous results (section 4.2, Figure 4.6). However, no statistical analysis was performed on the data of Figure 4.8 and thus the difference between the ratio of soaked-only skin and stretched skin 60  is only based on a graphical interpretation. Similar width/length ratios, 0.49±0.17 and 0.50±0.15, were measured respectively for skin stretched during one sinusoidal sequence and skin stretched for four sequences. The average ratio of skin stretched for seven sequences (0.62±0.17) appears to be higher than the ratio of both, skin stretched for one and four sequences, but with the magnitude of the standard deviation for each treatment group this difference between the width/length ratios may be not statistically different. Therefore, according to the graph depicted in Figure 4.8, the "rounding up" of fibroblast bodies mainly occurs in the first stretching sequence and increases slightly as the number of sinusoidal sequences increases.  4.2.5 Fibroblast morphology of skin stretched for 5 sine waves with and without C48/80 Skin samples were stretched for five sine wave cycles (instead of 40) in Kreb's buffer with and without addition of C48/80 and then fixed for morphological analysis. This test was performed to determine if an effect of C48/80 on fibroblast morphology could be observed after the first few sine wave cycles. This was only a preliminary test and thus only one rat and one skin sample was studied for each of the two treatment groups, i.e., skin stretched in plain Kreb's buffer and skin stretched in Kreb's buffer containing C48/80, and 5 micrographs were made per tissue sample. The results of the calculation of an average surface density ratio and width/length ratio are shown respectively in Figures 4.9 and 4.10. 61  3.5  rat  o  3 2.5  >.  <h 2 c  ed  Oi o (0  r3  CO  1.5 1  T  0.5  I-  0 Soakedonly in Kreb's  Stretched for 5 sine waves in Krebs  Stretched for 5 sine waves in C48/80  Treatment group  Figure 4.9 Surface density ratio for skin stretched for 5 sine wave c y c l e s in Kreb's buffer with or without addition of C48/80 (0.1 mg/ml) (20% strain)  The ratios shown in this graph correspond to an arithmetic mean (soaked-only tissues: n=3; stretched in Kreb's: n=4; stretched in C48/80: n=5). The error bars represent the positive standard deviation.  Ratios calculated for skin samples soaked-only in Kreb's buffer are used as control data. The soaked-only in Kreb's ratio is the same ratio as the one depicted in Figure 4.7 because both experiments (i.e. experiments presented in sections 4.2.4 and 4.2.5) used the same animals.  According to the graph  presented in Figure 4.9, a surface density ratio of 1.88±1.15 was measured for skin soaked-only in Kreb's buffer which is higher than the 0.41 ±0.27 ratio obtained for skin stretched in plain Kreb's buffer and a ratio of 0.15±0.15 for skin stretched in C48/80.  For the width/length ratio (Figure 4.10), a value of  0.30±0.12 (same ratio as the one depicted in Figure 4.8) was calculated for 62  Kreb's soaked-only tissue which is smaller than the 0.36±0.13 ratio obtained for skin stretched in Kreb's buffer. A higher value was measured for the skin stretched in presence of C48/80, that is 0.51+0.11.  0.7 0.6  o  •fO k-  *L  0.5 0.4  0)  0.3  !2  0.2 0.1 Soaked-  Stretched  Stretched  only in  for 5 sine  for 5 sine  Kreb's  waves in  waves in  Krebs  C48/80  Treatment group  Figure 4.10 Width/length ratio for skin stretched for 5 sine wave c y c l e s in Kreb's buffer with or without addition of C48/80 (0.1 mg/ml) (20% strain)  The ratios shown in this graph correspond to an arithmetic mean (soaked-only tissues: n=6; stretched in Kreb's: n=6; stretched in C48/80: n=7). The error bars represent the positive standard deviation.  Since no statistical analysis was performed on these data the differences in ratios between C48/80 exposed and unexposed tissues cannot be proven to be significant.  However, the smaller surface density ratio obtained for skin  stretched in C48/80 suggest that the retraction of cytoplasmic extensions after 5 sine wave cycles is more pronounced in skin stretched with than without C48/80. 63  Also, a closer width/length ratio to 1 for skin stretched in Kreb's containing C48/80 indicates that fibroblasts are "rounder" after 5 sinusoidal cycles in tissue samples stimulated mechanically in C48/80 than in Kreb's buffer alone.  4.3 Qualitative biomechanical  results  In addition to the animals used for the morphological study, about ten other rats were used to study the biomechanical response of the different groups to repeated sinusoidal strain tests. The experimental procedures described in chapter three were followed for these additional mechanical experiments, except that soaked-only tissues were not studied and no samples were fixed for electron microscopy.  4.3.1 Stretching in Kreb's buffer In this part of the study, skin was subjected to intermittent sinusoidal stretching. A sequence of 40 sine waves at 1 Hz was applied to the tissue sample every 15 minutes for a period of 90 minutes. Thus, during a complete experiment, a total of 7 sinusoidal sequences were performed on the tissue sample. During each sequence of sine waves, which lasted 40 seconds, the tension of the skin sample was recorded. The variation of the tension with the number of sine wave cycles during the seven sinusoidal sequences is presented in Figure 4.11 for a tissue soaked in Kreb's buffer and stretched with an extension of 20%.  The tension data  64  points shown in that figure are the maximum tensions corresponding to the amplitude peak of each sine wave, i.e. when the skin sample is fully extended. At a frequency of 1 Hz we expect there to be no delay between tension and strain. Even if tennis balls were used to reduce noise through the apparatus (section 3.1), a noisy tension signal was still recorded. The high sensitivity of the transducer and the equipment restriction on the use of tissue samples longer than 4 mm, probably caused most of the error observed in Figure 4.11. To obtain an estimate of the magnitude of the error in tension measurement, the mean, the standard deviation and the biggest difference between the data and the mean were calculated for the tension data points between the fifteenth sine wave cycle and the fortieth cycle for each of the last three sequences, i.e., the sequences performed at 60, 75 and 90 minutes. The range of the data points for the error estimation was chosen according to the assumption that the tension is constant after the fifteenth sine wave cycle for the last sinusoidal sequences. The following means and standard deviations were obtained respectively for the 5 , 6 th  th  and 7  1190.1±252.0 mg and 1123.9±253.6 mg. f  f  th  sequenced202.7±275.5 mg,,  The biggest difference measured  between the data point and the mean was 483.5 mg , 496.1 mg and 585.7 mg, f  f  for the 5 , 6 and 7 sinusoidal sequence. According to this calculation, most th  th  th  data points are found within a range of about 500mgf, as shown by the standard deviations, and an error up to about 586 mgf can be expected in the tension data presented in Figure 4.11.  65  T e n s i o n (mgf)  4000 Time zero, Y=-493*log(X)+3465, R 2=0.53 A  15 mins, Y=-307*log(X)+2561, R 2=0.41 A  30 mins, Y=-291*log(X)+2279, R 2=0.38  3500  A  45 mins, Y=-175*log(X)+1806, R 2=0.23 A  60 mins, Y=-107*log(X)+1591, R 2=0.17 A  75 mins, Y=-107*log(X)+1551, R 2=0.14 A  3000  90 mins, Y=-147*log(X)+1586, R 2=0.22 A  2500  2000  1500  1000  .  *+  500  I II 0  I  III IIIII 10  I  II!II  I  15  20  S""  1  IIIIII 25  I  Vi*  TT  I  30  Sine wave cycle number  III II 35  I  III I 40  45  Figure 4.11 Graph of tension as a function of sinusoidal waves a n d s e q u e n c e s for a skin sample subjected to an intermittent sinusoidal stretching in Kreb's buffer  Skin sample was stretched at an elongation of 20% 7 times every 15 minutes. The data point traced on the graph represent the tissue tension recorded at the maximal extension. The data were fitted with logarithmic equations which appear in the legend with their corresponding coefficient of determination (R 2). Other graphs resulting from biomechanical experiments are shown in Appendix D1. A  66  However, despite the presence of noise in the data, it can be observed that the tension has a tendency to decrease with the number of sine wave cycle during a sequence of 40 sine waves and also to decrease with the number of sequences. This trend is more visible when the data points are fitted with a logarithmic equation, i.e., tension = b*log(sine wave cycle number) + a, where a and b are two coefficients. Among other types of fitting curves, the logarithmic fit was found to be the more accurate one for these data and furthermore, Vogel (1994) demonstrated that the tension decreases, after repeated loading, approximately with the logarithm of number of cycles (Vogel ,1994). The fitted curves for each sinusoidal sequence are drawn on Figure 4.11 and the logarithmic equations are presented in the graph legend with their corresponding coefficient of determination (or R-squared value). Coefficients closer to one (0.53, 0.41 and 0.38) were obtained for the three first sequences which means that the logarithmic fitting correlates better the maximal tensions recorded during these first sinusoidal sequences. A certain recovery of the tension between each sinusoidal sequence, i.e. that the tension at the beginning of a sequence is higher than the tension at the end of the previous sequence can also be observed on the graph of Figure 4.11. The true interval between the end of one sequence and the beginning of the next one is 14.33 minutes.  During this time the tissue demonstrates recovery of  tension as measured in the first sine wave of the following sequence. Figure 4.12 presents the tension at specific sine wave cycles for each sinusoidal sequence as a percentage of the first sequence. The data points on  67  the graph represent the arithmetic mean of tension percentages calculated using four graphs of tension versus sine wave cycle number (Graph of Figure 4.11 and 3 graphs presented in Appendix D1).  Figure 4.12 Tension as a percentage of the first sinusoidal s e q u e n c e for a skin sample stretched at a strain of 2 0 % in Kreb's buffer each 15 minutes for a period of 90 minutes.  The tension percentages were calculated using fitting curves of four graphs of tension versus sinusoidal cycle number. To avoid confusion in the reading of the data only one portion, positive or negative, of the standard deviation is depicted on the graph (one sided error bars).  68  According to Figure 4.12, the relative decrease in tension as compare to the first sinusoidal sequence was more pronounced in the first sine wave cycles than in the middle and last ones. In the second sequence, almost 80% (78.8%±5.5) of the first tension was recovered at the first sine wave cycle and 87.5%±9.7 at the last cycle. This recovery of the first sequence decreases progressively along with the sequences until it reaches, in the last third sequences, a relatively constant tension equal to about 59% (between 57.4%±11.0 and 59.5%±13.4) of the first sequence at the first sine wave cycle and around 76% at the last cycle.  4.3.2 Stretching in C48/80 Qualitative biomechanical testing were also performed on tissue samples stretched in Kreb's buffer containing C48/80 (0.1 mg/ml). In these experiments, the first sinusoidal sequence was performed in plain Kreb's buffer (without addition of C48/80), such that the first sequence could be used as a reference point for the comparison of biomechanical results obtained for skin stretched in absence of C48/80 and skin stretched in presence of C48/80. The tension recorded during the seven sinusoidal sequences is presented in Figure 4.13 in function of the sine wave cycle number. The data points represent the tension measured at the maximal extension. The scattered data points indicate the presence of noise in the results and to estimate the variability of the tension measurements, means and standard deviation were calculated for the tension recorded between the 15 and 40 sinusoidal cycle for the last three th  th  69  sinusoidal sequences. The following means and standard deviation were calculated respectively for the 5 , 6 th  th  and 7  th  sequence: 1372.3mg ±149.3, f  1414.9mg ±168.2 and 1356mg ±142.9. According to these results, most of the f  f  data points are found within a region of about 300mg. To obtain an estimation f  of the maximal error induced in the tension measurement, the biggest difference between the data point and mean was calculated in the three last sequences from the 15 cycle to the 40 cycle. The following maximal differences were th  th  obtained respectively for the 5 , 6 and 7 sequences: 396.7mg,, 390.9mg and th  th  th  f  334.4mg,. Thus, an error up to approximately 400mg, may be expected in the measurement of the tension. Logarithmic fitting (tension=b*log(cycle number)+a) was performed on the tension data and the resulting equations are shown in the graph legend of Figure 4.13 as well as the corresponding coefficients of determination. As shown in Figure 4.13, the same trends in tension observed with tissue soaked in Kreb's buffer without addition of C48/80 are observed in presence of C48/80. Again the tension decreased with the number of sinusoidal waves and sequences and recovered to some degree between the sequences of sine waves. The tension also appears to be constant after a few sinusoidal waves in each sequence.  70  Tension (mgf) 3500.0 — r O  0  5  10  Time zero (Kreb's), Y=-433*log(X)+2924, R"2=0.76  —A—  15 mins, Y=-269"log(X)+2494, R 2=0.61  —B  30 mins, Y=-165"log(X)+2051, R*2=0.37  —+—  45 mins, Y=-150*log(X)+1907, R 2=0.43  A  15 20 25 30 Sine wave cycle number  A  35  40  45  Figure 4.13 Tension as a function of sinusoidal waves and s e q u e n c e s for a skin sample subjected to an intermittent sinusoidal stretching in Kreb's buffer containing C48/80 (0.1 mg/ml)  Skin sample was stretched 7 times every 15 minutes. The first stretching sequence was performed in plain Kreb's buffer and the following sequences were performed in presence of C48/80 in buffer (0.1 mg/ml). The data point traced on the graph represent the tissue tension recorded at the maximal extension (20%). The data were fitted with logarithmic equations which appear in the legend with their corresponding coefficient of determination (R 2). Similar graphs are presented in Appendix D2. A  71  Figure 4.14 presents the tension for each sinusoidal sequence as a percentage of the first sequence as a function of the sine wave cycle number. The  data points on the graph represent the arithmetic mean of tension  percentages calculated using two graphs of tension versus sine wave cycle number (Graph of Figure 4.13 and 1 graph presented in Appendix D2) and the corresponding standard deviation is represented by error bars.  %  120.0 110.0 100.0 90.0 80.0 70.0  15 mins - A  60.0  H  50.0 40.0  I I I | I I I I I I I  10  i  i i i  i i I  I  3 0 mins  —FJ —  4 5 mins  — + —  6 0 mins  - ~1r -  7 5 mins  —  9 0 mins  —0 I  _  I I i i i i I i i i i I i i I  15 20 25 30 Sine wave cycle number  35  40  45  Figure 4.14 Tension as a percentage of the first sinusoidal s e q u e n c e for a tissue sample stretched at a strain of 2 0 % in C48/80 (0.1 mg/ml) each 15 minutes for a period of 90 minutes.  The tension percentages were calculated using fitting curves of two graphs of tension versus sine wave cycle number. To avoid confusion in the reading of the data only one portion, positive or negative, of the standard deviation is depicted on the graph (one sided error bars). 72  As observed with skin stretched in plain Kreb's buffer (Figure 4.12), the relative decrease in tension as compared to the first sequence was more important in the first cycles than in the last ones. A recovery in tension of 84.5%±1.1 was calculated in the first sine wave cycle of the second sinusoidal sequence compare to a recovery of 110.3%+4.11 in the last cycle. As previously seen in the absence of C48/80 (Figure 4.12), the recovery in tension of the first sequence decreases progressively with the number of sequences until it reaches, in the last fourth sequences, a relatively constant tension. According to Figure 4.14, this constant tension is about 69% (between 68.82%±15.40 and 70.57%±14.97) of the first sequence at the first sine wave cycle and around 97% (between 95.9%±0.3 and 99.2%±4.1) for the last cycle. The tension as a percentage of the first sequence was in general superior for the skin stretched in presence of C48/80 (Figure 4.14) than for skin stretched only in Kreb's buffer (Figure 4.12), but since the magnitude of the error bars is important in most of these data the comparison between the two treatment groups is difficult and may be biased.  4.3.3 Stretching of previously frozen tissues To determine if cells adhesion was playing a role in the variation of tension as a function of sinusoidal cycle number, stretching was performed on previously frozen skin samples. By freezing skin samples in aluminum foil at 70°C, it was expected to kill the cells and therefore, fibroblasts activity should not interfere in the tension measurement.  Intermittent sinusoidal stretching was 73  performed on the previously frozen tissue samples, but instead of stretching the sample every 15 minutes it was extended every hour for a 3 hour period. The stretching conditions were different in these experiments with previously frozen tissues because the tests were conducted prior to the final experimental protocol described in this thesis and a period of rest of 1 hour between the sinusoidal sequences was investigated in these preliminary experiments. Non previously frozen tissues were also stretched in Kreb's buffer every hour for a 3 hour period. The measured tension as a function of sinusoidal cycle number of a fresh tissue sample stretched in Kreb's buffer and of a previously frozen skin sample stretched at 20% is shown respectively in Figure 4.15 and 4.16. A less sensitive force transducer was in placed when these experiments were performed which resulted in less scattered data points. Higher tensions are also observed in these graphs which could be explained by a more important stretching of the tissue sample at the baseline condition of the experiments; no marks were done on the back of the animal to assure that the skin was at its natural tension when mounted in the apparatus. The tension trend observed in both graphs is similar to the trend observed with skin stretched each 15 minutes in Kreb's buffer and with non previously frozen tissues, i.e., the tension decreases with the number of sinusoidal waves and sequences.  74  Tension  (mgf)  9000.0 — r —  8000.0 —\  7000.0 H  6000.0  4000.0 H  3000.0  2000.0 0  5  10  15  20  25  Sine wave cycle  30  35  40  45  number  Figure 4.15 Tension as a function of sinusoidal waves and s e q u e n c e s for a skin sample stretched in Kreb's buffer each hour for a 3 hour period  The data point depicted on the graph represent the tissue tension recorded at the maximal extension (20%). The data were fitted with logarithmic equations which appear in the legend with their corresponding coefficient of determination (R 2). Additional graphs are presented in Appendix D3. A  75  Tension  (mgf)  5500.0  m~  5000.0 —\  4000.0 —\  3500.0 —\  3000.0 —  2500.0 —\  0  5  10  15  20  25  Sine wave cycle  30  35  40  45  number  Figure 4.16 Tension as a function of sinusoidal waves and s e q u e n c e s for a skin sample previously frozen and stretched in Kreb's buffer every hour for a 3 hour period  The data point depicted on the graph represent the tissue tension recorded at the maximal extension (20%). The data were fitted with logarithmic equations which appear in the legend with their corresponding coefficient of determination (R 2). Another graph obtained with previously frozen tissue is presented in Appendix D4. A  76  The tension as a percentage of the first sinusoidal sequence is presented in Figure 4.17 for non previously frozen pieces of skin (stretched each hour for a 3 hour period) and in Figure 4.18 for previously frozen tissues  Figure 4.17 Tension as a percentage of the first sinusoidal s e q u e n c e for skin samples stretched in Kreb's buffer every hour for a 3 hour period  The percentages were calculated using fitting curves of three graphs of tension versus sine wave cycle number. To avoid confusion in the reading of the data only one portion, positive or negative, of the standard deviation is depicted on the graph (one sided error bars).  The data points of Figure 4.17 result from the averaging of percentages from 3 graphs (Figure 4.15 and two more graphs shown in Appendix D3). Two graphs (Figure 4.16 and 1 graph in Appendix D4) were used to obtain the data of 77  Figure 4.18. Slightly higher recovery of tension was obtained for tissues that were previously frozen. For example, at the first sine wave cycle of the second sinusoidal sequence, the tension was 78.9%±10.7 of the first sequence for the non frozen tissue and 93.2%±5.1 for the previously frozen skin samples. Furthermore, the relative lost in tension as compare to the first sequence appears to reach a constant trend faster when skin is previously frozen.  % 150.0 140.0 130.0 120.0 110.0 100.0 90.0  H  80.0 70.0 60.0 50.0 40.0  -0—  1 hour  -±—  2 hours  -j—I— 3 hours I I II  |I I i I |I I I I |I I I I |I I I I |I I I  5  10  15 20 25 30 Sine wave cycle number  35  40  45  Figure 4.18 Tension as a percentage of the first sinusoidal s e q u e n c e for previously frozen skin samples stretched in Kreb's buffer every hour for a 3 hour period  The tension percentages were calculated using fitting curves of two graphs of tension versus sine wave cycle number. To avoid confusion in the reading of the data only one portion, positive or negative, of the standard deviation is depicted on the graph (one sided error bars).  78  A different pattern in the variation of the tissue tension was obtained in one experiment with previously frozen skin (Figure 4.19). In this experiment, the tension decreased with the number of sine wave cycles during the first sinusoidal sequence, but a nearly constant tension was obtained after the first sequence instead of the gradual decreased in tension with the number of sequences observed in the other biomechanical experiments.  Tension (mgf) 6000.0 —0— 5500.0  Time zero, Y=-853*log(X)+5318, R*2=0.94  - A" • 1 hour, Y=-222*log(X)+2374, R*2=0.62 • 2 hours, Y=-281*log(X)+2417, R*2=0.73  —FJ  5000.0  - -+ -  3 hours, Y=-147*log(X)+1991, R 2=0.42 A  4500.0  4000.0  3500.0  3000.0  2500.0  2000.0 ' * .  13 _  D  A  A  A  ti  1500.0  *  +  1000.0  A  +"-*2iA"~- ?~'a»---b--A^+  *  i  ii  j  5  ii  i  I |I I  10  i  i |  i i i  I |I  •  M  *  I  •  A  Sol.  | ii ii | I I  15 20 25 30 Sine wave cycle number  A  +  T H S R B -  i i | i  35  I ii  | .1  40  II  45  Figure 4.19 Tension a s a function of sinusoidal waves and s e q u e n c e s for a tissue sample previously frozen and stretched in Kreb's buffer every hour for a 3 hour period The data point depicted on the graph represent the tissue tension recorded at the maximal extension (20%). The data were fitted with logarithmic equations which appear in the legend with their corresponding coefficient of determination (R 2). A  79  5. Discussion The stress-strain-time apparatus used in this project was built originally for another research project. Thus some of the apparatus capabilities were not directly suitable for quantitative analysis for the study presented in this thesis. The equipment constraints related to the stress-strain-time apparatus are discussed in this chapter.  The variability observed in cells viability for the  different treatment groups is also discussed in this section of the thesis followed by a discussion of the effect of soaking on morphological results. Finally, this chapter discussed the effect of stretching as well as the effect of C48/80 on cells morphology and tissue tension.  5.1 Stress-Strain-Time apparatus  The stress-strain-time apparatus used in this project has many attributes. This apparatus is capable of a diversity of dynamic and static testing and the simultaneous recording of tissue sample tension and position. In addition this equipment is controlled by a user-friendly software. However, the apparatus was built to the specifications of a previous research project (Wang, 1997) which had different experimental conditions than those studied here. Therefore, some of the capabilities of the apparatus have resulted in constraints to the present project. The limitation in the combination of frequency and amplitude for sinusoidal testing is one such constraint. As a result, we were restricted to 80  experimentation with short tissue samples. By choosing a frequency of 1 Hz, the maximal amplitude of the sine wave in this apparatus was 800pm for a motor step size of 2.5 um/step. Since we wanted to stretch the skin to a maximal strain of 20%, we were restricted to a tissue length of 4 mm at zero strain. An increase in the motor step size would have allow a higher sine wave amplitude and given us the possibility to use a longer piece of tissue, but the next step interval available on the compumotor was 5um/step; a large increment that would have provided an inaccurate sinusoidal tracing. One disadvantage of working with short pieces of tissue relates to handling of the tissues. The mounting of the tissue sample in the stress-straintime apparatus is a delicate task. While mounting, special care must be taken to avoid stretching of the tissue sample and damaging tissue organization which could influence the experimental results. The shorter the sample is, the greater is the risk of damaging the tissue. Furthermore, a short piece of tissue imposed a high width/length ratio for the specimen. To avoid a minimum of disturbance in the tissue structure, a width smaller than 2 mm could not be obtained when the tissue samples were cut with a scalpel.  Therefore, the tissue sample held by the clips of the  apparatus had an averaged width/length ratio of 1:2, which is high compared to the standard 1:5 to 1:8 ratio used in many studies for quantitative analysis and determination of biomechanical properties (Colin, 1982; Purslow et al., 1997; Vogel, 1975; Vogel etal., 1977 and 1979; Wan Abas, 1995). The width/length ratio of a specimen plays a role in the uniformity of stretching throughout the 81  gauge length of an experimental sample. When a tissue sample is stretched, the elongation of the region close to the holding clips may be different from that in the region closer to the center of the sample ("end effects"). To avoid the effect of a non uniform stretching on biomechanical measurements, researchers usually used dumbbell-shaped specimens and the strain is calculated according to the gauge length, which is the length between the two bells (Belkoff et al., 1991; Foutz etal., 1992; Vogel, 1975 and 1989; Vogel etal., 1977 and 1970). Determination of biomechanical properties with a tissue sample 4 mm long and with the equipment available to us is unreasonable and therefore, qualitative experiments were more suitable and acceptable for the goals of this project. Because of the end effects, fibroblasts were stretched a bit more or less than the average 20% elongation expected, but since the project was based on qualitative analysis the measurement of the precise amount of stretching was not required. A non uniform stretching throughout the skin sample may have also caused differences in fibroblast morphological changes along the different regions of the tissue sample. Consequently, to ensure consistency between the morphological results, the pieces of skin that were analyzed by transmission electron microscopy were always taken from the middle region of the stretched specimen since that is the region most likely to be uniformly stretched. Another problematic feature of the apparatus was the force transducer. The transducer used for the biomechanical experiments was too sensitive for this kind of experiments and as a result significant noise was noted in the tension signal. An estimation of the error in tension measurements was performed with  82  the graph of Figure 4.11.  The mean, standard deviation and the biggest  difference between a data point and a mean were calculated for the 5 , 6 and th  7  th  th  sinusoidal stretching sequence using the data between the fifteenth and  fortieth sine wave cycle; in this region of the curve a constant tension is assumed.  A standard deviation of about ±250mgf was calculated for each  sequence and the biggest difference measured between a data point and the mean was 586mg,. That is for each sequence, most data points were found within a range of about 500mg and an error in the tension measurement up to f  586mgt is possible in the results. The changes in tension between successive stretching sequences is not very different. For example, in Figure 4.11 the maximal difference found between two successive sequences using the fitting curves was about 1000mgt and this difference is only at the first sine wave cycle; the difference drops quickly in the successive sequences. With a variability of about 500mg in the data and with a possible error in the tension measurement f  around 586mg,, a difference of 1000mg between successive sequences may not f  be significant. Thus, the presence of noise in the tension measurements causes uncertainty in the interpretation of the results. Financial and time restraints prevented us from correcting this shortcoming of the equipment. A less sensitive transducer was previously used with the apparatus, but it was not robust enough to stretch skin and it was easily damaged. Since these transducers are very expensive and since we were not measuring any biomechanical properties, we opted for a more robust but, unfortunately, more sensitive transducer.  83  5.2  Viability of fibroblasts  Fibroblasts examined in micrographs from the virginal control samples and skin soaked both in Kreb's alone and in Kreb's containing C48/80 were healthy, i.e., the cells had an intact plasma membrane, a normal rough endoplasmic reticulum and an intact nucleus. The good condition of the cells in the virginal control group, in which skin was fixed immediately after it was removed from the animal's body, indicates that the handling of animals and the harvesting of the skin samples was not too disruptive and that the dermal fibroblasts could be adequately fixed throughout the thickness of the skin. The healthy cells observed in tissue soaked in Kreb's buffer suggest that the Kreb's solution provides all the nutrients necessary to keep the cells alive. It also demonstrates that the environmental conditions in the soaking dishes, i.e., temperature, pH and oxygenation, were sufficient to maintain cell viability. The Kreb's solution containing C48/80 appears also to have had no dramatic effects on fibroblast morphology based on the finding of intact cells in the tissues treated with this substance. This indicates that the compound 48/80 dissolved in Kreb's buffer at a concentration of 0.1 mg/ml doesn't appear to have compromised fibroblast viability. In sharp contrast, a significant number of disrupted fibroblasts were observed in skin samples that had been sinusoidally stretched with and without the addition of C48/80 in Kreb's buffer. Since fibroblasts in soaked-only tissues were in good shape, and since soaked-only skin and stretched skin were exposed to the same environmental conditions (i.e. temperature at 37°C, Kreb's  84  buffer bubbled with 95%02-5%C02), we suspect strongly that stretching may damage these cells. The examination of micrographs demonstrated that about half of the dead fibroblasts followed a programmed cell death called apoptosis. This process of cell death is controlled via cellular receptors, such as integrins. As mentioned in the introduction, integrins are glycoproteins that bind cells to elements of the extracellular matrix (ECM) and in some cases to neighboring cells. In addition to physically binding cells, integrins participate in signal transduction, that is extracellular signals can be sent inside the cell and provoke a specific cellular response. In the past few years it has been demonstrated that in the absence of appropriate ECM contacts, some cells (e.g. epithelial and endothelial cells) undergo programmed cell death or apoptosis (McConkey ef al., 1994; Meredith etal., 1997; Ruoslahti etal., 1994). It is hypothesized that apoptosis is a default pathway that cells enter in the absence of extracellular signals instructing them otherwise. The observation in stretched skin samples of both, retraction of fibroblast cytoplasmic extensions and apoptotic fibroblasts, corroborate these findings. Fibroblasts may have died because they lost ECM attachments when skin was stimulated mechanically and in the absence of extracellular signals via cellular receptors (e.g. integrins) they underwent programmed cell death (apoptosis). The magnitude of the stretching elongation may also be responsible in part of the death of fibroblasts. The force applied on the skin sample elongated by 20% may be more than the force that these cells can support. However, an  85  elongation of 20% is relatively small compared to the approximately 50% elongation used in the studies by Vogel on in vivo recovery of mechanical properties (Vogel et al., 1985; Vogel, 1993). In these in vivo experiments, skin samples were subjected to 30 repeated loading up to 50% of their initial length. After different periods of rest, the skin was stretched again by the same cyclic loading and the tension measured during this second test was compared to the tension measured during the first stretching test. After repeated loading, the tissue tension decreased with the number of loading cycles and Vogel demonstrated that this loss of tension is fully recovered after a period of rest of 16 hours.  In Vogel's works, no electron microscopy examinations were  performed on skin samples after they were stretched to investigate if cells were intact after the repeated strain of 50%. Vogel hypothesizes that the recovery is attributable to cellular functions. The fact that stretching was performed on a piece of skin removed from the animal body [in vitro) instead of being stretched directly on the animal, i.e. in vivo, is another possible explanation for the death of fibroblasts. Cells in skin that are subjected in vitro to mechanical stimulation may need more than adequate environmental conditions (i.e. physiological solution, temperature, oxygenation) to stay intact. Many factors may play a role in the destruction of fibroblasts during stretching.  To determine the causes of cells death in these studies, an  investigation of different experimental conditions on cells viability is warranted.  86  5.3 Effect of soaking on fibroblast morphology  A few skin samples were fixed immediately after they were removed from the animal; these are the virginal control samples. By comparing these samples with the skin samples soaked in Kreb's buffer we can determine if soaking alters the shape characteristics of cells. The measurement of the fibroblast surface density ratio (i.e. line intercepts for cell extensions/line intercepts for cell bodies) demonstrated that soaking in Kreb's buffer increases the relative surface area covered by fibroblast cytoplasmic extensions.  The surface density ratio  calculated for the skin soaked in Kreb's solution was significantly (p<0.001) greater than the ratio of the virginal control group (Section 4.2.2 and Figure 4.5). When skin is soaked as in this work, osmotic forces cause the penetration of the bathing solution into the tissue with subsequent tissue swelling. This increased fluid volume in the tissue may cause the increased extension of the fibroblasts. If the cells are attached to each other and to elements of the ECM, and if the attachments are maintained, then these points of attachment in an expanded tissue move causing the cells to extend. This extension of cells, increases the surface area of cytoplasmic extensions and decreases the area of the cell body. Therefore, when point counting analysis is performed, a greater surface density ratio is obtained for soaked skin than for virginal control tissue. Unlike the surface density ratio, no significant difference was measured between the fibroblasts width/length ratios of virginal control group and Kreb's soaked tissue. The finding of an increase surface density ratio and a constant width/length ratio between soaked tissue and virginal control is not contradictory,  87  because these are relative measures.  We did not measure absolute cell  volumes and their changes in this work, which would have helped to clarify this issue further.  5.4 Effect of stretching on fibroblast morphology  Retraction of fibroblast cytoplasmic extensions has been observed in fibroblasts grown on collagen gels after the lattices were subjected to external loading (Eastwood etal., 1998; Grinnell, 1994; Mochitate et al., 1991; Tomasek ef al., 1992).  The surface density ratio close to zero (<0.05) obtained for  stretched skin samples (Figure 4.5) indicates that retraction of fibroblast extensions also occurs in tissue in vitro for an elongation of 20%. The stretching of skin doesn't affect only the fibroblast extensions, but also the shape of the cell bodies. The calculation of a fibroblast width/length ratio indicated that fibroblast bodies round up when the skin was subjected to stretching. An average width/length ratio of 0.62±0.17 was obtained for the skin samples stretched in Kreb's buffer, compared to a value of 0.31 ±0.10 for skin only soaked in the Kreb's solution (Figure 4.6). Statistical analysis showed that the width/length ratio of skin stretched in the physiological solution was significantly (p<0.001) higher than the ratio of soaked-only samples. The closer the width/length ratio is to one the "rounder" are the fibroblasts since a ratio of one indicates a circle. The morphological changes observed in fibroblasts in tissue in vitro may be of importance in studies of some pathological conditions.  Fibroblast 88  attachments have been suggested to play a role in the increased negativity of interstitial fluid pressure measured in some acute inflammatory reactions (Koller et al., 1993; Reed ef al., 1997). This hypothesis is based in part on in vitro experiments (Grinnell, 1994; Rubin ef al., 1995) with fibroblasts grown on collagen gels in which it has been demonstrated that cell attachments and detachments are modulated by different growth factors and inflammatory agents. Thus, via an increase or decrease in the number of fibroblast attachments and cell contraction the interstitial pressure could be raised or reduced. Fibroblast morphological changes observed in this project result from physical rather than chemical stimuli; fibroblasts "round up" following stretching. This occurrence is consistent with loss of attachment between fibroblasts and ECM. Under these conditions of loss of attachment, fibroblasts will be unable to support mechanical stresses in the tissue. Therefore, the materials such as proteoglycans and hyaluronan which, due to their oncotic nature create expansive forces in the interstitium, are confined by smaller restraining forces, with a subsequent lowering of interstitial fluid pressure as described by Reed et al. (1997) and Koller et al. (1993). That is, the results of this work corroborate in a qualitative sense the hypothesis and experimental findings of Reed ef al. (1997) and Koller et al. (1993).  5.5 Effect of C48/80 on fibroblast  morphology  When the cellular surface density ratio and the width/length ratio of skin soaked in Kreb's buffer (figures 4.5 and 4.6) and skin soaked in Kreb's 89  containing C48/80 (figures 4.5 and 4.6) were compared, no statistical differences were found. Since it was verified (section 4.1.2) that mast cell degranulation occurred in tissues samples soaked in Kreb's buffer containing C48/80, it is justified to say that mast cell degranulation alone did not affect the relative surface area of fibroblast cytoplasmic extensions and the shape of the cell bodies. No significant differences were found in these ratios between skin stretched in Kreb's buffer with or without C48/80. That is, the retraction of fibroblast cytoplasmic extensions and rounding up of cell bodies was similar with or without addition of compound 48/80. When mast cell degranulation occurs in an intact tissue, chemical agents such as histamine are released and an inflammatory reaction is produced within the tissue.  Edema formation has been observed concomitantly with this  inflammation (Koller et al., 1993; Woie et al., 1993) and the hypothesis was that the edema was a result of breaking of fibroblast adhesions in the dermis of skin. This hypothesis was supported by the measured increased negativity of interstitial fluid pressure obtained after mast cell degranulation (Koller et al., 1993; Woie etal., 1993). If C48/80 was provoking the detachment of fibroblasts, the cytoplasmic extensions should retract and there should be a difference in the surface area covered by fibroblast extensions between skin soaked in the presence of C48/80 and skin soaked in Kreb's buffer only. But the surface density results show that this was not the case. The fibroblast bodies should also begin to round up if the attachments were broken by contact with C48/80,  90  but the results of the width/length ratio indicated that the cell bodies are not rounder in skin soaked in C48/80 than in skin soaked in plain Kreb's buffer. The hypothesis is not supported by the morphological results obtained with tissues which are soaked-only. However, when tissues were stretched in Kreb's solution with and without C48/80, the results for the surface density ratio and the width/length ratio of fibroblasts both showed "rounding up" of fibroblasts. It is possible that these two sets of results can be explained by a two-staged process related to disruption of fibroblast adhesion followed by "rounding up". For example, fibroblast adhesions may break under the influence of C48/80, but it may take mechanical stimulation to induce the retraction of cytoplasmic extensions and "rounding up" of cell bodies. That could explain the absence of a difference between the ratios calculated for skin soaked in C48/80 and skin soaked in Kreb's buffer. The rate at which fibroblasts round up is determined by the breaking rate of cell adhesions and the rate at which fibroblast extensions retract. If the rate at which cell attachments break is the limiting phase in the process of fibroblasts "rounding up" and if cell adhesions are broken only by contact with C48/80, then it may be possible to observe a morphological difference in fibroblasts between skin samples fixed after only a few stretching sinusoidal cycles for C48/80 exposed and unexposed tissues. To investigate this hypothesis, skin samples were stretched for five sine wave cycles (instead of 40) in Kreb's buffer with and without addition of C48/80 and then fixed for morphological analysis. This was only a preliminary test and  91  thus only one skin sample was stretched in plain Kreb's buffer and one sample was stretched in C48/80 and 5 micrographs were made per tissue sample. Results suggest that the retraction of cytoplasmic extensions after 5 sine wave cycles is more pronounced in skin stretched in buffer with C48/80 than without C48/80. That is a smaller surface density ratio was calculated for skin stretched in presence of C48/80 (0.15±0.15) than for skin stretched in Kreb's buffer only (0.41 ±0.27) (Figure 4.9). Fibroblast bodies appear also to be "rounder" after 5 sinusoidal cycles in tissue samples stimulated mechanically in C48/80 than in Kreb's buffer alone. This was indicated by a closer width/length ratio to 1 obtained for skin stretched in Kreb's buffer containing C48/80 (0.51 ±0.11) than in plain Kreb's buffer (0.36±0.13) (Figure 4.10). Since no statistical analysis was performed on these data the differences in ratios between C48/80 exposed and unexposed tissues cannot be proven to be significant. However, the hypothesis that C48/80 breaks cell adhesions alone and that the rate at which fibroblast adhesions break is an important phase in the process of fibroblasts "rounding up" is supported by these preliminary morphological results.  5.6 Effect of stretching on skin tension  The results of the qualitative biomechanical experiments suggest that the tension of a skin sample recorded during stretching has a tendency to decrease with the number of sine wave cycles within a sinusoidal sequence and also with the number of sequences (Figure 4.11). Some tension recovery is also observed between the sequences. The general trend observed in the biomechanical 92  experiments is similar to the results observed by Vogel (Vogel et al., 1985; Vogel, 1993). In collagen gels, it has been shown that fibroblasts generate a tension throughout the gel via cell adhesions (Brown et al., 1998; Delvoye et al., 1991; Eastwood et al., 1994) and it is believed that a similar tension is maintained by fibroblasts in the dermis (Delvoye et al., 1991; Reed etal., 1997). Therefore, if these cell attachments are broken, the tension in the tissue should decrease. The hypothesis that cell adhesions influence the mechanical response of a tissue is supported by the decrease in tension observed as a result of biomechanical experiments along with the "rounding up" of fibroblasts demonstrated in the morphological study. If fibroblast adhesions influence the tension in the experimental protocols as used here, then the retraction of cytoplasmic extensions and "rounding up" of cells must be progressive since the decrease in tension is gradual with the number of sinusoidal waves and sequences. To investigate this hypothesis, additional tests were performed; skin samples were fixed after one and four sinusoidal sequences instead of the seven sequences as reported in the main body of the results. Surface density ratio measurements showed an important retraction of fibroblast cytoplasmic extensions during the first stretching sequence followed by a progressive retraction of extensions as the number of sequences increases; the ratio value decreased with an increased in the number of sequences (Figure 4.7). Surface density ratios of 0.40+0.15, 0.16±0.14 and 0.039+0.070 were obtained respectively for skin stretched in Kreb's buffer for 1,  93  4 and 7 sinusoidal sequences. An increased of the "rounding up" of fibroblast bodies with the number of stretching sequences was also observed (Figure 4.8), but the difference between the width/length ratios was less obvious than with the surface density ratios.  Width/length ratios of 0.49±0.17, 0.50+0.15 and  0.62+0.17 were calculated respectively for skin stretched for 1, 4 and 7 sequences.  These fibroblast morphological changes observed after various  number of stretching sequences are consistent with the variation of skin tension observed in the biomechanical experiments. However, because of the small number of samples analyzed for these experiments and because of the magnitude of the standard deviations, these results must be considered as preliminary. To investigate further the role of fibroblast adhesions with respect to tissue tension, biomechanical experiments were performed on previously frozen skin samples. By freezing skin samples, it was expected to kill the cells and therefore fibroblast activity would not have any effect on recovery of tension during stretching experiments. Results from two experiments showed that no difference could be graphically observed in tension trends between fresh skin and previously frozen skin (Figures 4.15 and 4.16). In these experiments, skin was stretched each hour for a 3 hour period in Kreb's buffer. A decrease in tension was observed with the number of sine wave cycles and with the number of sequences for both, fresh and previously frozen tissues. A similar amount of tension recovery as a percentage of the first sequence was also observed in both types of skin samples, i.e., fresh and previously frozen tissues (Figure 4.17  94  and 4.18). If, as assumed, the cells were dead in the previously frozen skin samples, these results do not support the hypothesis that cell adhesion is in part responsible for tension changes. However, a different pattern of tension was obtained for one experiment with a previously frozen skin sample (Figure 4.19). In this test, the tension of the first stretching sequence decreased with the number of sine wave cycles similarly to the trend observed in tissue with living cells, but instead of a progressive decrease in tension with the number of sinusoidal sequences, a constant tension was attained after the second sequence. The decrease of tension with the number of sine wave cycles in the first sequence may be the result of the breaking of existing cell adhesions. The rapid lost of tension and the near absence of tension recovery, observed in this experiment with hypothetical^ dead cells, support the idea that cell adhesions play a certain role in tension changes during stretching. Because of the small number of experiments performed with previously frozen tissues and because of the divergence in the results, no conclusion can be drawn. More experiments are required to determine whether the changes in tension are different when stretching is performed on a previously frozen skin sample than when fresh skin is used. Biochemical tests (e.g ATP production) should be performed on skin samples previously frozen at -70°C to ensure that the cells are killed. The tissues should also be frozen for the same amount of time prior to the experiment because when a tissue is frozen at -70°C ice crystals grow in the tissue and this crystallization increases with time. In this project, the  95  two samples that showed no difference in tension trend were frozen for about two months compared to about two weeks for the other sample. Freezing may affect not just the cells but also the extracellular matrix and therefore another technique should be used to kill the cells (e.g. addition of cyanide in Kreb's buffer). In addition to a decrease in tension, biomechanical results show a certain recovery of tension after a time interval of 14.33 minutes (Figure 4.12). This recovery of tension as a percentage of the first sequence is partial and decreases with the number of stretching sequences until it reaches a certain constant in the last three sequences. Recovery of tension was also studied by Vogel (Vogel et al., 1985; Vogel, 1993). In his experiments, skin was stretched in vivo, i.e. the clips were attached to the animal, and the resting period was between 0.5 hour to 16 hours. At 16 hours the tissue reached full recovery. Since we have seen that fibroblasts undergo morphological changes in tissue, we can suspect that the recovery in tension observed in this work and also in Vogel's works is related, at least in part, to the reattachment of cells. In our study the period of rest between sequences was less than 15 minutes during which time cells can reattach with each other or with elements of the ECM.  5.7 Effect of C48/80 and stretching on skin tension  By qualitatively comparing the tension curves obtained for skin stretched in presence of C48/80 (Figure 4.13) to the ones obtained for skin stretched in Kreb's buffer only (Figure 4.11), no apparent difference in the trends related to 96  tension can be observed. As mentioned previously in the discussion, C48/80 induced an inflammatory type of reaction in tissue and it was expected to break fibroblast adhesions without requiring any mechanical stimulation.  With this  hypothesis in mind, it was expected to see a greater and/or more rapid decrease in the tension when tissue was stretched in Kreb's buffer containing C48/80. However, by calculating the tension of each sinusoidal sequence as a percentage of the tension of the first sequence, no significant difference can be seen with or without the addition of C48/80 in Kreb's buffer (Figure 4.14 and 4.12). These results suggest that stretching breaks all the fibroblast adhesion that exposure to C48/80 does. More work on the effect of C48/80 on mechanical response and specifically designed experiments are required to clarify this issue.  97  6. Conclusions The effect of sinusoidal stretching of fresh skin samples on both the mechanical response of the tissue and the morphology of the fibroblasts was investigated in vitro. These investigations were carried out on skin soaked in Kreb's buffer with and without addition of a mast cell secretagogue, C48/80. The stress-strain-time apparatus used in this project was not directly suitable for quantitative analysis for the study presented in this thesis; experiments had to be performed with short tissue samples and the force transducer induced a lot of noise in the tension signal.  Thus, qualitative  biochemical experiments were more acceptable for the goals of this project. An investigation of fibroblast viability demonstrated that stretching damaged the cells.  A significant number of apparently dead and/or dying  fibroblasts were observed in skin samples stimulated mechanically. The death of fibroblasts may be related to lost of attachments with the extracellular matrix and absence of extracellular signals as well as to various factors related to experimental conditions, such as the amount of stretching. To analyze the morphological changes of fibroblasts, surface density and width/length ratios were calculated using electron micrographs.  The surface  density ratio for skin samples stretched in Kreb's (0.04±0.07) was significantly less (p<0.001) than the ratio for soaked-only skin samples (1.45±0.69). Thus, the fibroblast cytoplasmic extensions retract when skin is subjected to sinusoidal stretching.  Fibroblast bodies also undergo changes in shape when skin is  stretched; they "round up". A width/length ratio closer to 1 and significantly 98  higher (p<0.001) was measured for skin stretched in Kreb's buffer (0.62±0.17) than for soaked-only tissues (0.31 ±0.10).  These morphological changes in  fibroblasts support the idea proposed by Reed et al. (1997) that cell adhesions play a role in the control of the interstitial fluid pressure. To investigate the effect of stretching on the mechanical response of the tissue, the skin tension was recorded during stretching as a function of the number of sinusoidal waves and sequences. A gradual decrease of tension with the number of sine wave cycles was observed within a sequence and also with the number of sequences. This decrease in tension as a result of biomechanical experiments along with the "rounding up" of fibroblasts indicated in the morphological study corroborate the hypothesis that cell adhesions influence the mechanical response of a tissue. Some tension recovery was also observed in these biomechanical experiments between the sinusoidal sequences.  Since  fibroblasts undergo morphological changes in tissue, the recovery in tension is suspected to be related in part to the reattachment of cells. To further investigate the possible role of fibroblast attachments in the mechanical response of the tissue, skin samples were fixed after 1 and 4 sinusoidal sequences instead of 7 as reported for the main results of this project. Surface density ratios of these preliminary experiments demonstrated an important retraction of fibroblast cytoplasmic extensions during the first stretching sequence followed by a progressive retraction of extensions as the number of sequences increases. Width/length ratio calculations showed also an important "rounding up" of fibroblast bodies during the first sequence and this "rounding up"  99  increased slightly as the number of sinusoidal sequences increases. These results are consistent with the decreased in tension observed in the qualitative biomechanical experiments. In an additional attempt to demonstrate the effect of fibroblast adhesions on the variation of tension of the tissue, stretching was performed on previously frozen skin samples.  The results of these biomechanical experiments were  inconclusive. Two out of three experiments demonstrated no difference between the tension trend observed with fresh skin and with previously frozen tissues and one experiment showed a completely different pattern in tension with no recovery in tension. More experiments are required. No effect of C48/80 was observed on fibroblasts morphology and on tissue tension. The hypothesis that the edema formation, caused by subsequent mast cell degranulation, is a result of the breaking of cell attachments was not supported by the morphological results; no significant difference was found in the surface density and width/length ratio measurements between soaked-only C48/80 exposed and unexposed tissues. No effect of C48/80 was observed on the tissue tension trend as a result of biomechanical experiments. Similar percentages of recovery in tension were measured for skin stretched with and without addition of C48/80 in Kreb's buffer. It was concluded that stretching breaks all the fibroblast adhesions that exposure to C48/80 does.  100  7. Recommendations To the best of our knowledge, the relationship between morphological changes in fibroblasts and the mechanical response of the skin haven't been studied in tissue before. Therefore, the project presented in this thesis was mostly a pilot study. The results obtained from this project have shown that some modifications should be introduced to the stress-strain-time apparatus for future experiments and that specified design experiments are required to better understand the role of fibroblast attachments in mechanical response as well as the effect of C48/80 on cell attachments. Electronic and design modifications should be performed on the stressstrain-time apparatus such that a piece of skin of at least 1 cm length can be used for experimentation. Furthermore, a less sensitive, and at the same time robust, force transducer should be placed in the apparatus. These two changes would allow the investigator to perform quantitative analyses and would reduce the error in the interpretation of biomechanical results. In an attempt to understand the death of fibroblasts in stretched tissues, sinusoidal stretching should be performed at a smaller elongation, for example 5%. If dead cells are still found in tissues at such a small strain then it would indicate that the amount of stretching is not the cause of death of fibroblasts. It would be also interesting to investigate the viability of cells in skin stretched in vivo. To perform the stretching in vivo would required a modification of the stress-strain-time apparatus so that the clips, that hold the tissue sample, could be placed on the back of the animal. Another solution would be to buy a new 101  apparatus that allow this kind of in vivo measurement. Different surface density ratios were measured for the virginal control group and for the skin soaked in Kreb's buffer. We suspect that the penetration of the bathing solution in soaked tissues increased the surface area of fibroblast cytoplasmic extensions and decreased the surface area of fibroblast bodies. To clarify this issue, the absolute cell volumes and their changes should be calculated. More experiments with skin stretched for one and four sinusoidal sequences should be performed such that a progressive retraction of fibroblast extensions could be statistically proven. These experiments are important for the investigation of the role that cell adhesions play in the mechanical response of the tissue. Experimentation with previously frozen tissues is another method that may be used to investigate the relation between fibroblast attachments and mechanical response of tissues. The experiments performed in this project with previously frozen skin samples were inconclusive; divergent results were obtained. Since there was evidence of interaction between cell adhesions and tissue tension in one test, more experiments with previously frozen tissues should be performed. Additional care should be taken to freeze the specimens such that only the cells are affected by the freezing process. Preliminary tests on skin samples stretched for only 5 sine waves in Kreb's buffer with and without C48/80 indicated that there was a difference in the retraction of fibroblast extensions and "rounding up" of fibroblast bodies between  102  the two treatment groups.  It would be interesting to do more of these  experiments and see if the difference between C48/80 exposed and unexposed tissues is statistically significant. These experiments could help to determine if C48/80 alone breaks fibroblast attachments. Another experimentation that could be designed to investigate the effect of C48/80 on cell adhesions would be to study fibroblast morphology in skin samples stretched at a smaller strain (e.g. 5% instead of 20%).  At a smaller elongation stretching may not break all  fibroblast attachments and thus, if C48/80 alone can break cell adhesions, "rounding up" of fibroblasts should be more pronounced in skin stretched in Kreb's buffer containing C48/80 than in Kreb's buffer only.  103  8. References Adams, J. C, "Cell adhesion - spreading frontiers, intricate insights", Trends in Cell Biology, 7(3) :107-110, 1997 Alberts, B; Bray, D.; Lewis, J.; Raff, M.; Roberts, K.; Watson, J. D., Molecular biology of the cell, Second edition, Garland Publishing, Inc., New York & London, pp. 791-836,1989 Barocas, V. H.; Tranquillo, R. T., "An Anisotropic Biphasic Theory of TissueEquivalent Mechanics: The Interplay Among Cell Traction, Fibrillar Network Deformation, Fibril Alignment, and Cell Contact Guidance", J. of Biomechanical Engineering, 119:137-145, 1997 Barocas, V. H.; Moon, A. G.; Tranquillo, R. T., "The Fibroblast-Populated Collagen Microsphere Assay of Cell Traction Force-Part 2: Measurement of the Cell Traction Parameter", J. of Biomechanical Engineering, 117:161-170, 1995 Belkoff, S. M.; Haut, R. C, "A structural model used to evaluate the changing microstructure of maturing rat skin", J. of Biomechanics, 24(8):711-720, 1991 Bell, E.; Ivarsson, B.; Merril, C, "Production of a tissue-like structure by contraction of collagen lattices by human fibroblasts of different proliferative potential in vitro", Proc. Natn. Acad. Sci. U.S.A., 76:1274-1278,1979 Bert, J. L.; Pearce, R. H., "The interstitium and microvascular exchange", Handbook  of  Physiology.  The  Cardiovascular  System.  Microcirculation.  Bethesda, MD: Am. Physiol. Soc, sect. 2, vol. IV, pt. 1, chapt. 12, 521-547, 1984 Brown, R. A.; Prajapati, R.; McGrouther, D. A.; Yannas, I. V.; Eastwood, M., "Tensional homeostasis in dermal fibroblasts: mechanical responses to mechanical loading in three-dimensional substrates", J. of Cellular Physiology, 175:323-332, 1998 Brown, R. A.; Talas, G.; Porter, R. A.; McGrouther, D. A.; Eastwood, M., "Balanced mechanical forces and microtubule contribution to fibroblast contraction", J. of Cellular Physioloy, 169:439-447, 1996 Champion, R. H.; Gillman, T.; Rook, A. J.; Sims, R. T., An introduction to the biology of the skin, Blackwell Scientific publications Oxford and Edinburgh, 1970 Chapuis, J. F.; Agache, P., "A new technique to study the mechanical properties of collagen lattices", J. Biomechanics, 25(1 ):115-120, 1992  104  Clark, E. A.; Brugge, J. S., "Integrins and signal transduction pathways: the road taken", Science, 268(5208) :233-239, 1995 Colin, H. D., "Biomechanical Properties of Dermis", Journal Of Investigative Dermatology, 79:17-20,1982 Cook, T. H., "Mechanical properties of human skin with aging", Aging and the skin, edited by A. K. Balin and A. M. Kligman, Raven Press, New York, pp. 205225, 1989 Cua, A. B.; Wilhelm, K.-P.; Maibach, H. I., "Elastic properties of human skin: relation to age, sex and anatomical region", Arch. Dermatol. Res., 282: 283288, 1990 Delvoye, P.; Wiliquet, P.; Leveque, J.-L; Nusgens, B. V.; Lapiere, C. M., "Measurement of mechanical forces generated by skin fibroblasts embedded in a three-dimensional collagen gel", Journal of Investigative Dermatology, 97(5):898-902, 1991 Delvoye, P.; Colige, A.; Lapiere, C. M., "Isometric traction developed by fibroblasts in a collagen gel: a model for pharmacological studies?", Journal of Investigative Dermatology, 91:393,1988 Eastwood, M.; Mudera, V. C ; McGrouther, D. A.; Brown, R. A., "Effect of precise mechanical loading on fibroblast populated collagen lattices: morphological changes", Cell Motility and the Cytoskeleton, 40:13-21,1998 Eastwood, M.; Porter, R.; Khan, U.; McGrouther, G.; Brown, R., "Quantitative Analysis of Collagen Gel Contractile Forces Generated by Dermal Fibroblasts and the Relationship to Cell Morphology", Journal of Cellular Physiology, 166:33-42, 1996 Eastwood, M.; McGrouther, D. A.; Brown, R. A., "A culture force monitor for measurement of contraction forces generated in human dermal fibroblast cultures: Evidence for cell-matrix mechanical signalling", Biochim. Biophys. Acta, 1201:186-192,1994 Edwards, C ; Marks, R., "Evaluation of Biomechanical Properties of Human Skin", Clinics in Dermatology, 13:375-380,1995 Foutz, T. L.; Stone, E. A.; Abrams, C. F. Jr., "Effects of freezing on mechanical properties of rat skin", Am J Vet Res, 53(5)788-792, 1992 Foutz, T. L; Abrams, C. F. Jr,; Stone, E. A.; Thrall, D. E., "Characterization of the non-linear loading curve of rat skin", Frontiers Med. Biol. Engng., 6(3):187-197, 1994 105  Fung, Y. C , Biomechanics: mechanical, properties of living tissue, 2  ed., New  York, Springer-Verlag, 1994 Glauert, A. M., Fixation, dehydration and embedding of biological specimens, North-Holland Biomedical Press, Amsterdam, The Netherlands, 1975 Grinnell, F., "Mini-Review on the Cellular Mechanisms of Disease", The Journal of Cell Biology, 124:401 -404, 1994  Harris, A. K.; Stopak, D.; Wild, P, "Fibroblast traction as a mechanism for collagen morphogenesis", Nature, 290(19):249-251, 1981 Hay, E. D., Organization and fine structure of epithelium and mesenchyme in the developing chick embryo, Epithelial-Mesenchymal Ineractions, Fleischmajer and  Billingham, 1968, pp. 31-55 Horwitz, A. F., "Integrins and Health", Scientific American, pp. 68-75, May 1997 Hougen, J. O., "Experiences and experiments with process dynamics", Chemical Engineering Progress Monograph Series, 60(4): 13-33,1964 Hougen, J. Ov, Walsh, R. A., "Pulse testing method", Chemical progress,  57(3):69-79,1961  Engineering  Koller, M.-E.; Woie, K.; Reed, R. K., "Increased negativity of interstitial fluid pressure in rat trachea after mast cell degranulation", J. Appl. Physiol., 74(5) :2135-2139, 1993 Mansour, J. M.; Davis, B. R., Srour, M.; Theberge, R., "A method for obtaining repeatable measurements of the tensile properties of skin at low strain", Biomechanics, 26(2) :211-216,1993  McConkey, D. J.; Orrenius, S., "Signal transduction pathways to apoptosis", Trends in cell biology, 4(10):370-375, 1994  Meredith, J. E.; Fazeli, B.; Schwartz, M. A., "The extracellular matrix as a cell survival factor", Molecular biology of the cell, 4:953-961, 1993 Meredith, J. E.; Schwartz, M. A., "Integrins, adhesion and apoptosis", Trends in cell biology, 7(4) :146-150, 1997 Mochitate, K.; Pawelek, P.; Grinnell, F., "Stress relaxation of contracted collagen gels: Disruption of actin filament bundles, release of cell surface fibronectin, and down regulation of DNA and protein synthesis", Exp. Cell. Res., 193:198-207, 1991  106  Montagna, W.; Parakkal, P. F., The structure and function of skin, 3 edition, New York, Academic Press, 1974 rd  Mridha, M.; Odman, S.; Oberg, P. A., "Mechanical pulse wave propagation in gel, normal and oedematous tissues", J. Biomechanics, 25(10):1213-1218, 1992 Nakagawa, S.; Pawelek, P.; Grinnell, F., "Extracellular matrix organization modulates fibroblast growth and growth factor responsiveness", Exp. Cell. Res., 182(2):572-582, 1989 Navajas, D.; Maksym, G. N.; Bates, J. H. T., "Dynamic viscoelastic nonlinearity of lung parenchymal tissue", J. Appl. Physiol., 79(1):348-356, 1995 Oxlund, H.; Manschot, J.; Viidik, A., "The role of elastin in the mechanical properties of skin", J. Biomechanics, 21 (3):213-218,1988 Oxlund, H.; Andreassen, T. T., "The roles of hyaluronic acid, collagen and elastin in the mechanical properties of connective tissues", J. Anal, 1 3 1 : 611-620, 1980 Pereira, J. M.; Mansour, J. M.; Davis, B. R., "Dynamic measurement of the viscoelastic properties of skin", J. Biomechanics, 24(2):157-162, 1991 Potter, D. D.; Furshpan, E. J.; Lennox, E. S., "Connections between cells of the developing squid as revealed by electrophysiological methods", Proc. Nat. Acad. Sci. U.S.A., 55:328-336, 1966 Potts, R. O.; Chrisman, D. A. Jr.; Buras, E. M. Jr., "The dynamic mechanical properties of human skin in vivo", J. Biomechanics, 16(6):365-372,1983 Purslow, P. P.; Wess, T. J.; Hukins, D. W. L., "Collagen orientation and molecular spacing during creep and stress-relaxation in soft connective tissues", Journal of Experimental Biology, 201(1 ):135-142, 1998 Reed, R. K.; Woie, K.; Rubin, K., "Integrins and control of interstitial fluid pressure", News Physiol. Sci., 12, February, pp. 42-48,1997 Reith, A.; Mayhew, T. M., Stereology and morphometry in electron microscopy: problems and solutions, An ultrastructural pathology publication series, Hemisphere publishing corporation, USA, 1988 Rubin, K.; Sundberg, C ; Ahlen, K.; Reed, R. K., "Integrins: transmembrane links between the extracellular matrix and the cell interior", Interstitium, connective tissue and lymphatics, Editors: Reed, R. K., McHale, N. G., Bert, J. L., Winlove, P., Laine, G. A., Portland Press: London UK, chapt. 3,1995  107  Ruoslahti, E.; Reed, J. C, "Anchorage dependence, integrins, and apoptosis", Cell, 77:477-478, 1994 Selye, H., The mast cells, Washington, Butterworths, pp. 157-165,1965 Silver, F. H., Biological Materials: structure, mechanical properties and modeling of soft tissues, New York University Press, 1987 Stopak, D.; Harris, A. K., "Connective tissue morphogenesis by fibroblast traction", Developmental biology, 90:383-398, 1982 Tomasek, J. J.; Haaksma, C. J.; Eddy, R. J.; Vaughan, M. B., "Fibroblast contraction occurs on release of tension in attached collagen lattices: Dependency on an organized actin cytoskeleton and serum", Anat. Rec, 232:359-368, 1992 Tong, P.; Fung, Y-C, "The stress-strain relationship for the skin", Biomechanics, 9:649-657, 1976 Van Kooten, T. G.; Schakenraad, J. M.; Van der Mei, H. C ; Busscher, H. J., "Development and use of a parallel-plate flow chamber for studying cellular adhesion to solid surfaces", Journal of Biomedical Materials Research, 26:725738, 1992 Vernon, R. B.; Angello, J. C; Iruela-Arispe, M. L; Lane, T. F.; Sage, E. H., "Reorganization of basement membrane matrices by cellular traction promotes the formation of cellular networks In vitro", Laboratory Investigation, 66(5):536547, 1992 Vogel, H. G., "Mechanical measurements of skin", Acta Dermato-Venereologica, No. S185 SI5, pp. 39-43, 1994 Vogel, H. G., "In Vivo Recovery of Repeatedly Strained Rat Skin after Systemic Treatment with Desmotropic Drugs", Skin Pharmacol., 6:103-110, 1993 Vogel, H. G., "Mechanical properties of rat skin with aging", Aging and the skin, edited by A. K. Balin and A. M. Kligman, Raven Press, New York, pp. 227-275, 1989 Vogel, H. G.; Denkel, K., "In Vivo Recovery of Mechanical Properties in Rat Skin After Repeated Strain", Arch. Dermatol. Res., 277:484-488, 1985 Vogel, H. G., "Repeated relaxation and determination of the isorheological point in skin strips of rats as influenced by maturation and aging", Bioengineering and the Skin, 1:321-335, 1985  108  Vogel, H. G., "Age dependence of viscoelastic properties in rat skin. Directional variations in stress-strain and hysteresis experiments", Bioeng Skin, 2:136-155, 1983 Vogel, H. G., "Age dependence of mechanical parameters in rat skin following repeated strain", Aktuel Gerontol, 8[11]:601-618,1981 Vogel, H. G.; Hilgner, W, "Influence of Age and of Desmotropic Drugs on the Step Phenomenon Observed in Rat Skin", Arch. Derm. Res., 264:225-241, 1979 Vogel, H. G.; Hilgner, W., "Analysis of the Low Part of Stress-Strain Curves in Rat Skin. Influence of Age and Desmotropic Drugs", Arch. Derm. Res., 258:141150, 1977 Vogel, H. G., "Strain of rat skin at constant load (creep experiments). Influence of age and desmotropic agents.", Gerontology, 23:77-86,1977 Wan Abas, W. A. B., "Stress stabilisation behaviours in skin under small tensile loads in vitro", Bio-Medical Materials and Engineering, 5(2):59-63, 1995 Wang, L., "Dynamic Mechanical Response of Airway Mucosal Membrane", Ph.D. Thesis, University of British Columbia, 1997 Wijn, P., "The alinear viscoelastic properties of human skin in vivo for small deformations", Ph.D. Thesis, University Nijmegen, 1980 Woie, K.; Koller, M.-E.; Heyeraas, K. J.; Reed, R. K., "Neurogenic inflammation in rat trachea is accompanied by increased negativity of interstitial fluid pressure", Circulation Research, 73(5):839-845,1993 Wyllie, A. H.; Kerr, J. F. R.; Currie, A. R., "Cell death: the significance of apoptosis", International Review of Cytology, 68:251-306,1980  109  Appendix A: Linearity and calibration of the force transducer  Example of records on the calibration and linearity of the force transducer.  Figure A.1 Calibration of the force transducer with standard weights  2500 -j  Actual weight (mg)  110  Appendix B: Records of point counting Example of number of line intercepts and number of hits recorded for point counting analysis are shown in this appendix. The numbers recorded for 5 micrographs are presented for each treatment group. The number of line intercepts for fibroblast extensions and fibroblast bodies is presented in the following tables as well as the surface density ratio (denoted by E/B in the examples). The number of hits was counted for fibroblast extensions, fibroblast bodies, collagen fascicles, interstitial fluid space and cells other than fibroblasts (or other structural components of the dermis such as blood vessels). The relative proportion of each of these elements in the micrograph was calculated by dividing the number of hits of each element by the total number of hits counted in the micrograph. The resulting percentage for each element is shown in the following tables under the sign "%".  111  Example of recorded numbers for point counting for the Virginal  Treatment:  Virginal Control  Treatment:  Control  group  Virginal Control  M i c r o g r a p h #2  M i c r o g r a p h #1 E/B  Line Intercepts: - Fibroblasts body (B)  50  - Fibroblasts extensions (E)  64  1.28  %  Hits:  E/B  Line Intercepts: - Fibroblasts body (B)  41  - Fibroblasts extensions (E)  46  %  Hits:  - Fibroblasts body  88  12.9  - Fibroblasts extensions  14  2.1  1.12  - Fibroblasts body  60  8.8  - Fibroblasts extensions  12  1.8  549  80.5  564  82.7  - Fluid s p a c e  20  2.9  - Fluid s p a c e  39  5.7  - Other cells  11  1.6  - Other cells  7  1  - Collagen fascicles  Treatment:  Virginal Control  Treatment: E/B  - Fibroblasts body (B)  43  - Fibroblasts extensions (E)  25  - Collagen fascicles - Fluid s p a c e - O t h e r cells  Treatment:  Virginal Control  Micrograph  #5  0.581  0.55  %  - Fibroblasts extensions  501  73.6  - Collagen fascicles  63  9.3  - Fluid s p a c e  155  23.1  0  0  - Other cells  0  0  - Fibroblasts body  87  12.9  5  0.74  425  63.2  E/B  - Fibroblasts extensions (E)  41  - Other cells  30  0.44  1.24  %  Hits:  - Fluid s p a c e  - Fibroblasts extensions (E) Hits:  3  33  - Fibroblasts extensions  55  16.7  - F i b r o b l a s t s b o d y (B)  - Collagen fascicles  - F i b r o b l a s t s b o d y (B)  114  Line Intercepts:  - Fibroblasts body  E/B  Line Intercepts:  %  Hits: - Fibroblasts extensions  Virginal Control  M i c r o g r a p h #4  M i c r o g r a p h #3 Line Intercepts:  - Fibroblasts body  - Collagen fascicles  42  6.2  14  2.1  591  87.3  30  4.4  0  0  11  Example of recorded numbers for point counting for skin soaked in Kreb's buffer  Treatment: Soaked in Kreb's Micrograph #1 Line Intercepts: . - Fibroblasts body (B) - Fibroblasts sheets (E) Hits: - Fibroblasts body - Fibroblasts sheets - Collagen fascicles - Fluid space - Other cells  Treatment: Soaked in Kreb's Micrograph #3 Line Intercepts: - Fibroblasts body (B) - Fibroblasts sheets (E) Hits: - Fibroblasts body - Fibroblasts sheets - Collagen fascicles - Fluid space - Other cells  Treatment: Soaked in Kreb's Micrograph #5 Line Intercepts: - Fibroblasts body (B) - Fibroblasts sheets (E) Hits: - Fibroblasts body - Fibroblasts sheets - Collagen fascicles - Fluid space - Other cells  25 29  E/B 1.16 %  34 9 291 355 0  50 56  4.9 1.3 42.2 51.5 0  E/B 1.12 %  83 16 479 83 23  52 39 96 8 543 36 7  12.1 2.3 70 12.1 3.4  Treatment: Soaked in Kreb's Micrograph #2 Line Intercepts: E/B - Fibroblasts body (B) 25 - Fibroblasts sheets (E) 51 2.04 % Hits: - Fibroblasts body 47 6.8 - Fibroblasts sheets 13 1.9 - Collagen fascicles 533 77.5 - Fluid space 95 13.8 - Other cells 0 0  Treatment: Soaked in Kreb's Micrograph #4 Line Intercepts: E/B - Fibroblasts body (B) 45 - Fibroblasts sheets (E) 51 1.13 Hits: % - Fibroblasts body 12.4 83 - Fibroblasts sheets 10 1.5 537 80.4 - Collagen fascicles - Fluid space 5.2 35 - Other cells 3 0.45  E/B 0.75 % 13.9 1.2 78.7 5.2 1  113  Example of recorded numbers for point counting for skin soaked and stretched at 20% in Kreb's buffer  Treatment: Stretched in Kreb's (20%)  Treatment: Stretched in Kreb's (20%)  Micrograph #1  Micrograph #2 E/B  Line Intercepts: - Fibroblasts body (B) - Fibroblasts sheets (E) Hits: - Fibroblasts body  - Fibroblasts body (B)  0  0 %  - Fibroblasts sheets (E) Hits:  0 %  - Collagen fascicles  0 452  65.8  - Fluid space - Other cells  155 8  22.6 1.2  - Fibroblasts body  - Fibroblasts sheets  0 510 72  0 73.7  - Fibroblasts sheets  - Collagen fascicles  10.4 1.3  0  10.5  14.6  9  33  72  101  - Fluid space - Other cells  E/B  Line Intercepts:  47  0  Treatment: Stretched in Kreb's (20%)  Treatment: Stretched in Kreb's (20%)  Micrograph #3 Line Intercepts:  Micrograph #4  - Fibroblasts body (B) - Fibroblasts sheets (E) Hits: - Fibroblasts body - Fibroblasts sheets - Collagen fascicles - Fluid space - Other cells  E/B 55 1  0.0182 %  - Fibroblasts sheets (E) Hits: - Fibroblasts body  100 1 494  14.5 0.14  95 0  13.8 0  71.6  26 0  0 %  69  - Fibroblasts sheets  0  - Collagen fascicles - Fluid space  409 211  - Other cells  0  - Fibroblasts body (B) .- Fibroblasts sheets (E)  25 0  Hits:  Treatment: Stretched in Kreb's (20%) Micrograph #5 E/B Line Intercepts: - Fibroblasts body (B)  E/B  Line Intercepts:  10 0 59.4 30.6 0  - Fibroblasts body - Fibroblasts sheets - Collagen fascicles - Fluid space - Other cells  0 %  63  9.1  0 488  0 70.7  . 139 0  20.1 0  Example of recorded numbers for point counting for skin soaked in C48/80  T r e a t m e n t : Soaked in C48/80  T r e a t m e n t : Soaked in C48/80  M i c r o g r a p h #1 Line Intercepts:  27 52  - Fibroblasts body (B) - Fibroblasts sheets (E) Hits:  -  Micrograph  E/B 1.93 %  Fibroblasts body Fibroblasts sheets Collagen fascicles Fluid space Other cells  81 11 475 104 6  12 1.6 70.2 15.4 0.89  T r e a t m e n t : Soaked in C48/80  - Fibroblasts body (B) - Fibroblasts sheets (E)  Hits:  -  Fibroblasts body Fibroblasts sheets Collagen fascicles Fluid space Other cells  21 24  - Fibroblasts body (B) - Fibroblasts sheets (E)  Hits:  -  Fibroblasts body Fibroblasts sheets Collagen fascicles Fluid space Other cells  -  Fibroblasts body Fibroblasts sheets Collagen fascicles Fluid space Other cells  32 6 317 329 0  Line Intercepts:  1.143 %  4.8 0.88 46.3 48.1 0  #5  Line Intercepts:  Hits:  40 70  E/B 1.75 %  39 4 583 67 0  M i c r o g r a p h #4  E/B  T r e a t m e n t : Soaked in C48/80 Micrograph  - Fibroblasts body (B) - Fibroblasts sheets (E)  5.6 0.58 84.1 9.7 0  T r e a t m e n t : Soaked in C48/80  M i c r o g r a p h #3 Line Intercepts:  #2  Line Intercepts:  45 29  E/B 0.644 %  93 8 467 66 52  13.6 1.2 68.1 9.6 7.6  - Fibroblasts body (B) - Fibroblasts sheets (E)  22 58  Hits:  -  Fibroblasts body Fibroblasts sheets Collagen fascicles Fluid space Other cells  59 11 483 133 5  E/B 2.64 % 8.5 1.6 69.9 19.2 0.72  Example of recorded numbers for point counting for skin soaked and stretched in C48/80  Treatment:  Stretched in C 4 8 / 8 0 ( 2 0 % )  Treatment:  Stretched in C 4 8 / 8 0 ( 2 0 % )  Micrograph  #1  Micrograph  #2  Line Intercepts: - Fibroblasts body (B) - Fibroblasts s h e e t s (E)  Line Intercepts:  E/B 22 5  - Fibroblasts body (B) 0.227  Hits:  - Fibroblasts s h e e t s (E)  E/B 62 2  0.0323  Hits:  %  %  45  6.6  - Fibroblasts s h e e t s  1  0.15  - Fibroblasts sheets  2  0.3  - Collagen fascicles  449  65.7  - C o l l a g e n fascicles  358  53.8  - Fluid s p a c e  188  27.5  - Fluid s p a c e  160  24.1  0  0  48  7.2  - Fibroblasts body  - O t h e r cells  - Fibroblasts body  - Other cells  Treatment:  Stretched in C 4 8 / 8 0 ( 2 0 % )  Treatment:  Micrograph  #3  Micrograph  Line Intercepts: - Fibroblasts body (B) - Fibroblasts s h e e t s (E)  51  Hits:  - Fibroblasts sheets (E)  E/B 50 1  Hits:  %  - Fibroblasts body  #4  - Fibroblasts body (B) 0.0392  14.6  Stretched in C 4 8 / 8 0 ( 2 0 % )  Line Intercepts:  E/B 2  97  0.02 %  58  8.5  - Fibroblasts s h e e t s  2  0.29  - Fibroblasts sheets  - Collagen fascicles  429  62.5  - Collagen fascicles  - Fluid s p a c e  184  26.8  - Fluid s p a c e  73  10.3  13  1.9  - Other cells  10  1.4  - O t h e r cells  Treatment:  Stretched in C 4 8 / 8 0 ( 2 0 % )  Micrograph  #5  Line Intercepts: - Fibroblasts body (B) - Fibroblasts s h e e t s (E)  E/B 30 1  Hits:  0.0333 %  63  9.2  - Fibroblasts s h e e t s  2  0.29  - Collagen fascicles  476  69.2  - Fibroblasts body  - Fluid s p a c e  65  9.4  - O t h e r cells  82  11.9  - Fibroblasts body  87 1 537  12.3 0.14 75.8  Appendix C: Statistical analyses In this appendix, details are provided on the two-way randomized Block Anova test performed for the statistical analysis of the surface density ratio and width/length ratio. The tissue sample was chosen as the blocking factor, i.e., the "experimental unit" for statistical analyses and the sample size was 9 (n=9). Surface  density ratio  Table C.1 Results of the statistical analysis of the surface density ratio  (df: degrees of freedom, F: F-distribution factor, Sig.: significance, H: hypothesis, E: error) Source Intercept C4880added Stretched C4880added*stretched Tissue  H E H E H E H E H E  Sum of Squares 22.958 0.939 0.115 1.372 20.288 1.372 8.302E-02 1.372 0.939 1.372  Mean • ,.;-VF:. Square 195.687 22.958 0.117 0.115 2.009 5.717E-02 20.288 354.893 5.717E-02 8.3027E-02 1.452 5.717E-02 0.117 2.052 5.717E-02  df 1 8 1 24 1 24 1 24 8 24  Sig. 0.000 0.169 0.000 0.240 0.083  Width/length ratio Table C . 2 Results of the statistical analysis of the width/length ratio  (df: degrees of freedom, F: F-distribution factor, Sig.: significance, H: hypothesis, Source Intercept C4880added Stretched C4880added*stretched Tissue  H E H E H E H E H E  Sum of Squares 7.650 4.4E-02 1.6E-04 34.2E-02 0.802 4.2E-02 3.3E-04 4.2E-02 4.4E-02 4.2E-02  df 1 8 1 24 1 24 1 24 8 24  "  Mean Square 7.650 5.5E-03 1.6E-04 1.8E-03 0.802 1.8E-03 3.3E-04 1.8E-03 5.5E-03 1.8E-03  F  Sig.  1383.0  0.000  0.094  0.762  457.99  0.000  0.188  0.668  3.159  0.014  117  Appendix D: Additional Graphs from Biomechanical Experiments Additional graphs as a result of biomechanical experiments are included in this appendix.  118  Appendix Dl: Graphs of tension for skin samples stretched in Kreb's buffer (time interval of 15 minutes between two stretching sequences)  Tension  (mgf)  1100.0 —.  ,l  1000.0 —  800.0 —\  700.0 —\  600.0 —  500.0 —  0.0  5.0  10.0  15.0  20.0  25.0  30.0  35.0  40.0  45.0  Sine wave cycle number  Figure D1.1 Tension as a function of sinusoidal waves and sequences for a skin sample stretched at an elongation of 20% in Kreb's buffer every 15 minutes for a period of 90 minutes  119  Tension (mgf)  3500.0 -  o  I I I I |  0  5  l l l l l l  10  —©—  Time zero, Y=-310*log(X)+2554, R 2=0.50  —A—  15 mins, Y=-156*log(X)+2123, R 2=0.19  —H  30 mins, Y=-169*log(X)+2108, R 2=0.34  —+—  45 mins, Y=-215*log(X)+2065, R 2=0.42  — *—  60 mins, Y=-113*log(X)+1796, R 2=0.19  —)K  75 mins, Y=-108*log(X)+1809, R 2=0.16  - •# -  90 mins, Y=-93*log(X)+1716, R 2=0.12  A  A  A  A  A  A  A  TT  l l l l l l  15  20  25  30  Sine wave cycle number  M i l l  35  40  45  Figure D1.2 Tension as a function of sinusoidal waves and sequences for a skin sample stretched at an elongation of 20% in Kreb's buffer every 15 minutes for a period of 90 minutes.  120  Tension  (mgf)  4000.0 Time zero, Y=-562*|og(X)+3538, R 2=0.67  O  3500.0  A  —A—  15 mins, Y=-357"log(X)+2629, R 2=0.48  —B  30 mins, Y=-171 *log(X)+2033, R 2=0.20  —+—  45 mins, Y=-114*log(X)+1779, R 2=0.10  —  60 mins, Y=-129*log(X)+1820, R 2=0.12  A  A  A  A  75 mins, Y=-95*log(X)+1690, R 2=0.09 A  90 mins, Y=-175*log(X)+1845, R 2=0.21 A  3000.0  2500.0  2000.0  1500.0 •  *  1000.0  500.0  A  TT 0  B  1  M l ! '  10  9  11  I I I I I I  15  20  +  It  1  •  A  I I I I I I I I I I I I  25  30  Sine wave cycle number  35  40  45  Figure D1.3 Tension as a function of sinusoidal waves and sequences for a skin sample stretched at an elongation of 20% in Kreb's buffer every 15 minutes for a period of 90 minutes.  121  Appendix D2: (mg/ml)  Graph of tension for skin samples stretched in C48/80  T e n s i o n (mgf)  1000.0 O  Time zero (Kreb's), Y=-70*log(X)+828, R 2=0.41 A  - A-  15 mins, Y=-22*log(X)+693, R 2=0.067  —0  30 mins, Y=-59*log(X)+693, R 2=0.42  A  A  +  900.0  45 mins, Y=-18*log(X)+615, R 2=0.054 A  - -k-  60 mins, Y=-31*log(X)+672, R 2=0.13  —)K  75 mins, Y=-40*log(X)+660, R 2=0.24  - -# -  90 mins, Y=-15*log(X)+603, R 2=0.055  A  A  A  800.0  700.0  600.0  +  500.0  *  *  *  ° *  •  400.0  M-l I  0  I  I I I 10  ^  I I  II I II 15  I  20  ° * *  II I II 25  n n  I  l"l  B* ~  +  1+  I II I I I 30  S i n e w a v e c y c l e number  35  I !  III I  40  I I  I  45  Figure D2.1 Tension as a function of sinusoidal waves and sequences for a skin sample stretched at an elongation of 20% in C48/80 (0.1 mg/ml) every 15 minutes for a period of 90 minutes. The first sinusoidal sequence (time zero) was performed in Kreb's buffer without addition of C48/80.  122  I  I  I I !  I I  Appendix D3: Graphs of tension for skin samples stretched in Kreb's (time interval of 1 hour between two stretching sequences)  buffer  Tension (mgf)  7000.0 o  6500.0 -  —©—  Time zero, Y=-747*log(X)+5983, R 2=0.78  —A—  1 hour, Y=-515*log(X)+5442, R 2=0.63  —B  2 hours, Y=-459*log(X)+4794, R 2=0.56  - +  A  A  A  -  3 hours, Y=-378*log(X)+4400, R 2=0.50 A  6000.0 -  5500.0 -  »o  o  5000.0  •  \"  4500.0 o  A A  4000.0  v+  A  o  o  no  •  3500.0  •  3000.0  •  2500.0  TT  11111  0  10  TT 15  I  20  I I  I I  I  25  •  TT 30  Sine wave cycle number  •  +  TT 35  40  TT 45  Figure D3.1 Tension as a function of sinusoidal waves and sequences for a skin sample stretched at an elongation of 20% in Kreb's buffer every hour for a 3 hour period.  123  T e n s i o n (mgf)  7000.0 o 6500.0 6000.0  -0—  Time zero, Y=-844*log(X)+5193, R 2=0.81  -A—  1 hour, Y=-513*log(X)+3888, R 2=0.81  -E3  2 hours, Y=-352*log(X)+3182, R 2=0.78  -+ -  3 hours, Y=-314*log(X)+2906, R 2=0.68  A  A  A  A  5500.0 5000.0 4500.0 4000.0 3500.0 3000.0  o O  O  o  o  o  """"®--^£>  2500.0 +  X •  -  E1-+  2000.0  A  n  n  '  U  I  0  I  I I I I I I I II  10  D  •  +  1500.0  o cb  + + I  I I  15  +  +  9  +  I I I I I I I I I ! ! I I I I I ! i I I I I I  20  25  30  35  40  Sine wave cycle number  Figure D3.2 Tension as a function of sinusoidal waves and sequences for a skin sample stretched at an elongation of 20% in Kreb's buffer every hour for a 3 hour period.  124  Appendix D4: Graph of tension for a previously frozen skin sample stretched in Kreb's buffer (time interval of 1 hour between two stretching sequences) Tension (mgf)  13000.0 — o  12000.0  — © —  Time xero, Y=-1342'log(X)+10473, R 2=0.f  —A—  1 hour, Y=-511*log(X)+9380, R 2=0.62  —H-  2 hours, Y=-646*log(X)+7860, R 2=0.72  A  A  A  - +  -  3 hours, Y=-818*log(X)+7504, R 2=0.86 A  11000.0  10000.0  9000.0  8000.0  7000.0  ^  + V  6000.0  + + + + +  5000.0  -Inn 4  - + ~ - -  +  4000.0  I I I I I I I I I I I I I II  0  10  O,  15  • + • n +• •  +  +  u  J .  +  + + - -.  +  I I I I I I I ! I I ! I I I I I •! j I I I I I I ! I I  20  25  30  Sine wave cycle number  35  40  45  Figure D4.1 Tension as a function of sinusoidal waves and sequences for a previously frozen skin sample stretched at an elongation of 20% in Kreb's buffer every hour for a 3 hour period.  125  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0058617/manifest

Comment

Related Items