All PhD Theses

J.A.T.C. Parker

Microtextured biomedical polymers and growth factors in an attempt to improve soft tissue healing

16-01-2003

A scientific essay in Medical Sciences

DOCTORAL THESIS defended in public on 16th of January 2003

SUMMARY

Wound healing is a complex repair process, which follows disturbances in tissue integrity, due to injury. Tissue injury can be acute or become chronic. Acute injury is the result of trauma, and a wound can become chronic because of certain conditions that prevent the normal wound healing response from closing the wound. Examples of such systemic- or local conditions are: metabolic disorders, ischemia or malnutrition. Disturbances in wound healing can be manifested in a variety of ways. Apart form insufficient wound approximation, when the wound does close, it can contract and cause both cosmetic- and functional disturbances. Also growth can be impaired. The current studies were concerned with the prevention of the derailment of the wound healing response by developing methods to manipulate wound healing. The specific aim was to fundamentally alter tissue healing around implants by improving soft tissue regeneration and limiting the inflammatory reaction, scar tissue formation and wound contraction. This aim can be fulfilled when the fibrous tissue capsules are thinner with superior organisation of the connective tissue components, and the presence of inflammatory cells in the capsules of interface is scarce or absent. In chapter 1, background information is given concerning wound healing, transforming growth factor-beta, polymer scaffolds that can be used as implants, surface microtexturing and contact guidance. In the following chapters, studies are described that address the problems and hypotheses that were formulated in the introductory section. The research was specifically concerned with the design of the implant (problems 1-3), the mechanical properties (1, 2), tissue adhesion to the implant (1, 2) and the anchoring of the implant in tissue (1-3). In addition, the research was concerned with the tissue reaction on the interface (1-3, 5) and the thickness, degree of cellularity and (geometrical) constitution of the capsule (connective tissue organisation) (1-3, 5). Finally, the loading and controlled release of a functional growth factor from implants in vitro and in vivo (4) was investigated, and the presence of indicators of wound contraction in the capsule (5). In this summary, this research is briefly discussed on a point-by-point basis.Does microtexturing of subcutaneous (biodegradable) membranes lead to:A) Superior connective tissue organisationB) A thinner capsuleC) Reduced inflammatory reactionD) Superior tissue attachment?

In chapter 2, this problem was addressed by evaluating the tissue reaction around implants with different surface topography. Coin-shaped silicone and poly-l-lactic acid (PLA) implants were made with double-sided parallel micro-grooves (depth 1.0 µm, width 10.0 µm), random roughness on a micrometer scale, and smooth control implants. These implants were inserted into subcutaneous pockets created on the flanks of goats. After 1, 3 or 12 weeks, goats were sacrificed, the implants retrieved and histologically processed. Light microscopic evaluation revealed the formation of fibrous tissue capsules around all implant materials. The PLA did not visibly degrade during the study period. Histomorphometrical analyses were performed on capsule thickness, capsule quality and the implant-tissue interface quality. Capsules around PLA implants showed significant better capsule quality, compared with silicone implants. Compared to smooth implants, capsule around micro-grooved implants were thicker, but capsules around roughened implants were thinner. However, randomly roughened implant surfaces generally elicited a stronger and more prolonged inflammatory reaction compared to smooth- and micro-grooved implant surfaces. It was concluded that the application of micro-grooves or random surface roughness to polymer implants apparently does not have beneficial effects on peri-implant tissue healing.Does enhanced tissue adhesion of subcutaneous microtextured (biogradable) polymer implants lead to:A) Superior connective tissue organisationB) A thinner capsuleC) Reduced inflammatory reactionD) Superior tissue attachment?From in vitro studies it is known that a plasma-treatment can enhance cell spreading. Similar effects can be observed after pre-treatment of the surface with a protein coating, to mediate cell modifications in a three-month experiment in a goat model. We made silicone and poly-l-lactic acid implants with double-sided parallel micro-grooves (depth 1.0 µm, width 10.0 µm), a random surface roughness, or a smooth surface. Implants either received a radio-frequency glow discharge (RFGD) treatment, a fibronectin (Fn) pre-coating, or no pre-treatment. Subsequently, they were inserted into subcutaneous pockets created on the flanks of goats for 1, 3 or 12 weeks. Histological analysis showed that a fibrous tissue capsule had formed around all implants. Histomorphometrical analysis was performed on capsule thickness, capsule quality and the implant-tissue interface quality. Fn-treated surfaces showed a considerable early inflammatory reaction. Besides this, RFGD treatment or Fn pre-coating did not further influence any of the measured parameters. In conclusion, pre-treatment of polymer implant surfaces with Fn or RFGD treatment did not significantly influence tissue reaction around implants with micro-grooved, roughened or smooth surfaceDoes immobilization of microtextured (biodegradable) polymer implants lead to:A) Superior connective tissue organisationB) A thinner capsuleC) Reduced inflammatory reaction?

In chapter 4, we compared the tissue reaction to smooth and micro-grooved implants at different implantation sites. We hypothesized that subperiosteally less mobility is to be expected between an implant and surrounding tissue, which can lead to a more subdued tissue response. In addition, we hypothesized that a similar effect can be reached when substrata are equipped with micro-grooves. Poly-l-lactic acid smooth or micro-grooved surfaces (width 2 / 10 µm, depth µm) were implanted subperiosteally on the frontal bone of the skull and subcutaneously in the flanks of goats for 2, 4 and 12 weeks. After sacrifice, implants and surrounding tissue were histologically processed. Light microscopical and histormorphometrical evaluation on capsule thickness, capsule quality and implant tissue interface was performed. In addition, we stained for α-smooth muscle actin, collagen and CD-68 expression. All implants were surrounded by a fibrous capsule. Capsules around subperiosteal implants were more matured than around subcutaneous implants. In time, capsule thickness significantly decreased around subperiosteal implants, but increased around subcutaneous implants. The applied surface topography did not influence the tissue reaction significantly. Also, nowhere differences were found in the presence of collagen or α-smooth muscle actin. The interfacial cells around all implants frequently showed staining for the monocyte-macrophage marker CD-68. We concluded that in this model, decreased mobility of an implant relative to the surrounding tissue did positively influence the peri-implant tissue response, but the applied surface topography did not.Can microtextured (biodegradable) membranes be used for controlled release of TGF-β 3:A) In vitroB) In vivo (subcutaneous)?

Transforming Growth Factor beta 3 (TGF-β 3) has been under investigation with the objective to improve wound healing. Yet, little experimental knowledge exists about applications of TGF-β 3 in implantology and tissue engineering. In chapter 5, the release kinetics and bioactivity of TGF-β 3 were determined, when released form microtextured silicone and poly-l-lactic acid (PLA) surfaces in vitro and in vivo. We loaded surfaces with 100 ng TGF-β 3. An in vitro assay showed that TGF-β 3 was released in a burst-like manner. Released TGF-β 3 was capable of inhibiting the proliferation of mink lung epithelial cells, indicating that released TGF-β 3 had remained at least partly active. Subsequently, an in vivo experiment (1h – 3 days) was performed using implants loaded with TGF-β 3. In crysections, TGF-β 3 activity was assessed with an in situ bioassay. We found that active TGF-β 3 was released up to 24 hours. Furthermore, released TGF-β 3 could be detected up to 320 µm from the implant. On the basis of these observations, we conclude that TGF-β 3 loaded onto microtextured polymer membranes remains functional when released in vitro and in vivo and therefore, may represent an alternative for introducing a growth factor into a wound to achieve long-term and long-range biological effects.Does loading of subcutaneous microtextured (biodegradable) membranes lead to:A) Superior connective tissue organisationB) A thinner capsuleC) Reduced inflammatory reaction?In both normal and disturbed wound healing, the generation of large, contracting scars can raise serious functional and cosmetic problems. A possible strategy to minimize or avoid the generation of scar tissue surrounding an implant is to apply TGF-β 3 to the implant. This was done in an in vivo experiment, as described in chapter 6.

0, 1 or 2.5 µg of TGF-β 3 was freeze-dried onto poly-l-lactic acid micro-grooved substrates (width 10 µm, depth 1 µm) and implanted subcutaneously in the back of rats for 2 and 8 weeks. After sacrifice, implants and surrounding tissue were histologically processed. Light microscopical and histomorphometrical evaluation of capsule thickness, capsule quality and implant-tissue interface was performed. In addition, we stained for α-smooth muscle actin (SMA), collagen and ED-1 (a monocyte-macrophage marker). All implants were surrounded by a fibrous capsule. Capsules of the implants loaded with 1 or 2.5 µg of TGF-β 3 showed a significantly higher quality. This meant that capsules were more matured compared to implants without TGF-β 3. However, no significant differences were found in the thickness of the capsules or in the quality of the interface. After 2 weeks, significantly more α-SMA was detected in capsules of the implants, with 2.5 µg of TGF-β 3, compared to 1 µg or no TGF-β 3. After 8 weeks, significantly more α-SMA expression was detected around all implants with 1 µg of TGF-β 3, compared to implants with 2.5 µg or no TGF-β 3. In conclusion, the use of micro-grooved PLA substrates with TGF-β 3 did not lead to an overall improvement of peri-implant tissue healing. 

Ultimately, the ambition of all wound healing investigators is to obtain a better understanding of the basic mechanisms which underly the wound healing process. The aim is to find a way to close a wound as quickly as possible, independent of size or cause, and with the lost- or injured tissue being regenerated i. e. replaced by tissue, which closely resembles the original tissue. In this thesis, experiments are described that form a contribution to this wound healing research. The prime research objective of these experiments was to develop a strategy for regulating wound healing, that can have a predictable impact on peri-implant tissue healing. The proposed strategy consisted of the application of different modifications to the surface of polymer implants. The modifications consisted of topographical alterations (chapters 2-4), plasma- or chemical pre-treatment (chapter 3), and the application of the growth factor TGF-β 3 (chapter 6). The hypothesized impact on tissue healing consisted of an improved soft tissue regeneration and inhibition of scar tissue formation, inflammatory reaction and wound contraction. Also, some fundamental mechanism were addressed. For instance, the influence on tissue reaction of movement at the interface between biomaterial and surrounding tissue (chapter 4). Further, the loading of TGF-β 3 on polymer implants was studied, and the in vitro release kinetics and activity, and in vivo activity and diffusion were characterised (chapter 5). A major finding in these studies was that the application of a microtexture to the surface of polymer implants, whether it be standardized or at random, in general did not lead to thinner or higher quality capsules with less inflammatory cells. It seems that the geometry of these textures, which has been reported to result topographical cues to fibroblasts in vitro, does not have a significant effect on capsule formation in vivo. A proposed explanation, is that the inflammatory cells which are present at the interface and are closely situated to the implant-surface, therefore sterically hinder the fibroblasts and prevent them from making contact with the texture, thus preventing contact guidance and alignment. Apparently, the long-standing suggestion, that micortexturing of implants has beneficial effects on peri-implant tissue healing, is false. Nevertheless, these results do not exclude other possible utilisations of the standardized microtextured surface pattern; such as for guided tissue regeneration of monolayers of for instance epithelial cells in percutaneous devices, or when a growth factor has to be loaded onto the surface of an implant. Also, experiments with different geometrical patterns of the standardized microtexture, may shed new light on this technology. Both pre-treatment of the surface of an implant with plasma or a fibronectin pre-coating, did not significantly aid in thinner, higher quality capsules with less inflammatory cells. Apparently, the use of methods that have been shown to enhance cel adhesion to implant materials in vitro, have no marked influence on the soft tissue reaction in vivo. Another long-standing issue, is the influence of movement of the implant relative to the surrounding tissue, on tissue reaction. In our model of relative immobility of the implant (superiostal location), capsules were of higher quality and became thinner in time, compared to our model of mobility (subcutaneous localisation). No differences in interface quality were observed. Movement between implant and surrounding tissue evidently leads to a more severe, less organised connective tissue reaction. Fixation of an implant relative to its surrounding tissue, may therefore prevent widening of the tissue-implant interface, infections, migration of the implant and in implant failure. A further part of our strategy to minimize or avoid the generation of scar tissue surrounding an implant, was to apply TGF-β 3 to the surface of microtextured implants. Initially, the release kinetics were characterized as being a burst-release, with the released TGF-β 3 was still active in tissue 24 h post-implantation, and showed diffusion through tissue. Thus, the described method may represent an alternative for introducing a growth factor into a wound to achieve long-term and long-range biological effects. Evaluation of the putative effects was performed in a model of peri-implant tissue healing. It was shown that implants with TGF-β 3 only partly improved tissue healing. This result however does not mean that application of TGF-β 3 to other wounds has no reducing effect on scar tissue. More insight has yet to be gathered, in the temporal and spatial levels of the various bioactive mediators, protein(ases) and extracellular matrix components, that are part of wound healing. In addition, more knowledge is needed concerning the presence-, growth factor production- and interactions of the cells that are part of wound healing. For instance, it remains to be determined whether a specifically directed growth factor which is applied to a wound, has to be released in one go- or has to be released slowly, to have the maximum desired effect. Depending on the required mechanism of action, a fitting method of release can be utilized.