Thomas Horbett - Academia.edu (original) (raw)
Papers by Thomas Horbett
Journal of Colloid and Interface Science, Aug 1, 1988
J Biomater Sci Polym Ed, 1995
Fibrinogen adsorbed to polymeric surfaces and then allowed to reside on the surface while it is k... more Fibrinogen adsorbed to polymeric surfaces and then allowed to reside on the surface while it is kept in a buffer solution for a period of time (the 'residence time') undergoes postadsorptive changes that decrease its SDS elutability, displaceability by plasma, polyclonal antifibrinogen binding, and ability to support platelet adhesion (summarized in Chinn et al. J. Biomed. Mater. Res. 26, 757 (1992)). In order to better understand the nature of the changes in adsorbed fibrinogen, the binding of ten different monoclonal antifibrinogen molecules to fibrinogen adsorbed from plasma to Biomer and several other surfaces has been measured after increasing residence time in buffer. Three of the monoclonal antibodies used bind to sequences that have been implicated in platelet binding to fibrinogen. One of these (M1) binds to the C-terminal region of the gamma chain (402-411), another (R1) binds to the N-terminal region of the A alpha chain containing an RGDF sequence (95-98), and the third (R2) binds to the C-terminal region of the A alpha chain containing an RGDS sequence (572-575). Two other antibodies (P1 and K4) also bind to the C-terminal region of the gamma chain (373-385 and 392-406, respectively). Five other antibodies that bind to other regions in fibrinogen were also used. Two of the antibodies (K4 and P1) are also known to be sensitive to conformational changes in the fibrinogen molecule. The binding of the various antibodies changed with residence time in ways that were highly dependent on the particular antibody. The binding of some antibodies was very stable with respect to residence time, others rose with time, some declined with residence time and one appears to pass through a maximum. However, none of the changes in antibody binding were nearly as fast as has been observed for the changes in platelet binding reported previously. Binding to the platelet binding region near the gamma chain C-terminal region either did not change with residence time (M1), increased with residence time (K4), or else decreased more slowly than observed for platelets (P1). Binding of the antibodies to the RGD sequences near the N-terminus of the A alpha chain (95-98) was very low initially but increased with residence time, while the binding to the RGD sequence near the C-terminus of the A alpha chain (572-575) increased slightly at short residence times but then declined substantially after longer residence times. Thus, the changes in the expression of the putative platelet binding domains do not correlate with the declines in platelet binding to plasma preadsorbed Biomer.(ABSTRACT TRUNCATED AT 400 WORDS)
Journal of Biomedical Materials Research Part a, Dec 15, 2010
The role of complement C3 in mediating adhesion of monocytes to plasma deposited tetraglyme surfa... more The role of complement C3 in mediating adhesion of monocytes to plasma deposited tetraglyme surfaces was studied. Although fibrinogen (Fg) is usually considered the main factor in mediating phagocyte attachment, plasma deposited PEO-like tetraethylene glycol dimethyl ether (tetraglyme) coatings that have ultra-low Fg adsorption (< 10 ng/cm 2 ) from low concentration solutions and low monocyte adhesion in vitro still show high phagocyte adhesion after short implantations and later become encapsulated when tested in vivo. To test whether higher Fg adsorption under in vivo conditions could explain the higher in vivo reactivity, we again measured the resistance of tetraglyme films to Fg adsorption. We found a surprising and previously unreported increased amount of adsorbed Fg on tetraglyme surfaces from higher concentration protein solutions. However, monocyte adhesion to tetraglyme did not markedly increase despite the increased Fg adsorption. We thus suspected proteins other than Fg must be responsible for the increased in vivo reactivity. We found that on tetraglyme pre-adsorbed with C3-depleted serum, monocyte adhesion was greatly reduced as compared to samples adsorbed with normal serum. Addition of exogenous pure C3 to the serum used to pre-adsorb the surfaces restored monocyte adhesion to tetraglyme coatings. While Fg clearly plays an important role in mediating monocyte adhesion to tetraglyme surfaces, the results show an additional role for adsorbed C3 in monocyte adhesion.
J Biomater Sci Polym Ed, 1992
A direct enzyme-linked immunosorbent assay (ELISA), using a polyclonal anti-fibrinogen conjugated... more A direct enzyme-linked immunosorbent assay (ELISA), using a polyclonal anti-fibrinogen conjugated to horseradish peroxidase, was used to detect fibrinogen adsorption from blood plasma to ten different materials. Adsorption was also measured with [125I]-fibrinogen. The materials studied included glass, Biomer, Immulon I, and a series of hydroxyethylmethacrylate (HEMA) and ethylmethacrylate (EMA) co-polymers. For all the materials studied, the results from the ELISA technique closely paralleled those obtained using [125I]-fibrinogen. The cross-reactivity of the antibody with proteins other than fibrinogen was generally small. Both experimental methods detected the presence of a maximum in fibrinogen adsorption (as a function of the plasma dilution) to the more hydrophobic materials. For all but two HEMA/EMA co-polymers, a linear correlation between the ELISA and [125I]-fibronogen measurements was indicated by inspection of cross plots as well as by a statistical test.
Journal of Biomedical Materials Research, Oct 1, 1995
Images of the Twenty-First Century. Proceedings of the Annual International Engineering in Medicine and Biology Society, 1989
A computer model of a glucose-sensitive membrane device that includes an additional flux of oxyge... more A computer model of a glucose-sensitive membrane device that includes an additional flux of oxygen from the insulin reservoir to alleviate the oxygen depletion problem has been developed. The response of the membrane with the additional flux of oxygen shows a larger pH drop and a greater sensitivity to glucose than in the case where there is no flux of oxygen from the insulin reservoir. The sensitivity to glucose is shown to increase to 200 mg% glucose as compared to the no-flux limit of 100 mg% with the pH drop increasing as well
Journal of Laboratory and Clinical Medicine
To study mechanisms of complex thrombus formation in vivo, and to compare the relative antithromb... more To study mechanisms of complex thrombus formation in vivo, and to compare the relative antithrombotic effects of anticoagulants and antiplatelet agents, a model was developed in baboons. Segments of collagen-coated tubing followed by two sequentially placed expansion chambers exhibiting disturbed flow patterns were exposed to native blood under laminar flow conditions. The device was incorporated for 1 hour into an exteriorized arteriovenous shunt in baboons under controlled blood flow (20 ml/min). Morphologic evaluation by scanning electron microscopy showed that thrombi associated with collagen were relatively rich in platelets but thrombi in the chambers were rich in fibrin and red cells. Deposition of indium 111-labeled platelets was continuously measured with a scintillation camera. Platelet deposition increased in a linear (collagen-coated segment) or exponential (chambers 1 and 2) fashion over time, with values after 40 minutes averaging 24.1 +/- 3.3 x 10(8) platelets (collagen segment), 16.7 +/- 3.4 x 10(8) platelets (chamber 1), and 8.4 +/- 2.4 x 10(8) platelets (chamber 2). Total fibrinogen deposition after 40 minutes was determined by using iodine 125-labeled baboon fibrinogen and averaged 0.58 +/- 0.14 mg in the collagen segment, 1.51 +/- 0.27 mg in chamber 1, and 0.95 +/- 0.25 mg in chamber 2. Plasma levels of beta-thromboglobulin (beta TG), platelet-factor 4 (PF4), and fibrinopeptide A (FPA) increased fourfold to fivefold after 60 minutes of blood exposure to the thrombotic device. Platelet deposition onto the collagen segment, chamber 1, and chamber 2 was linearly dependent on the circulating platelet count. Platelet accumulation in chamber 1 and chamber 2 was also dependent on the presence of the proximal collagen segment. An anticoagulating dose of standard heparin decreased platelet deposition in the chambers (p less than 0.05) but did not decrease deposition onto the collagen segment. Although beta TG and PF4 levels remained elevated after the administration of standard heparin, the elevation in plasma FPA was interrupted. Further evidence that the thrombotic process was dependent on platelets was provided by the finding that prostaglandin I2 at high concentration (35 ng/ml) decreased platelet deposition onto the collagen segment and in chambers 1 and 2, decreased beta TG and PF4 release, and reduced FPA formation. The combination of standard heparin and PGI2 produced the most potent inhibition of platelet thrombus formation and prevented the increases in plasma PF4, beta TG and FPA.
Investigative Ophthalmology & Visual Science
The outgrowth of corneal epithelial cells onto a polymeric substrate is expected to be the primar... more The outgrowth of corneal epithelial cells onto a polymeric substrate is expected to be the primary event in the epithelialization of a synthetic corneal graft. To study the effects of polymer surface properties on corneal epithelial cell outgrowth, a quantitative in vitro cell outgrowth assay was done. Polymers with systematic variations in hydroxyl content were used as outgrowth substrates. These polymers were characterized by electron spectroscopy for chemical analysis for elemental surface composition and by captive air-bubble contact angle for surface wettability. Circular corneal buttons were punched from excised rabbit corneas, placed onto these substrates, and incubated in a hormonally enriched culture medium. Outgrowth of the epithelial cells was allowed to proceed onto the substrates for 4 days. The outgrowth areas were measured, and an outgrowth index was calculated for each substrate by comparing it with tissue culture polystyrene substrate. The highest outgrowth generally occurred on substrates with intermediate wettabilities (captive air-bubble contact angles of approximately 45-75°); it was less on substrates of lower or higher wettabilities. Protein coatings of albumin, immunoglobulin G (IgG), fibronectin, and culture medium were found to lower the wettabilities of native substrates. Albumin and IgG precoating were shown to reduce epithelial cell outgrowth; fibronectin precoating was shown to improve outgrowth on most substrates. These results suggest that epithelial cell outgrowth is influenced by both substrate and protein interactions. Invest Ophthalmol
Journal of Molecular Recognition, 1996
Two peptides from the ligand-binding site of the platelet receptor GPIIb/IIIa, residues 296-306 o... more Two peptides from the ligand-binding site of the platelet receptor GPIIb/IIIa, residues 296-306 of GPIIb, designated B12 by D'Souza et al. (1991), and 300-311 of GPIIb, designated G13 by Taylor et al., (1992), as well as two control peptides, designated C14 and C20, were adsorbed to treated polystyrene substrates. Fibrinogen adsorption to the peptide-coated substrates was characterized. The specificity of I-125 labeled fibrinogen binding to the peptide-coated substrates was investigated by measuring the amount of fibrinogen adsorbed to each substrate and the inhibition of fibrinogen binding by RGDS peptide, bovine serum albumin, a divalent ion chelator (ethylene diamine tetra-acetic acid disodium salt), unlabeled fibrinogen and the B12 peptide. The results show that non-specific binding of fibrinogen to the B12-coated substrate is predominant under most conditions. Binding of monoclonal antibodies to fibrinogen adsorbed to the peptide coated substrates was characterized. The failure of several antibodies to bind fibrinogen adsorbed to the B12 substrate suggested that adsorption of fibrinogen to the B12-coated substrate alters its conformation relative to fibrinogen adsorbed to the bare substrate.
Journal of Controlled Release, 1993
Explicit expressions for the hydronium ion transport have been added to an existing theoretical m... more Explicit expressions for the hydronium ion transport have been added to an existing theoretical model describing the steady-state and transient behavior of glucose-sensitive membranes. The glucose sensitive membrane is a hydrogel containing pendant amine groups and immobilized glucose oxidase creating a membrane which swells in response to changes in glucose concentration. The extent of membrane swelling and the time required for swelling depend upon the pH within the membrane. The membrane pH is a function of the rate of production of gluconic acid from glucose as well as the transport rate of the hydronium ion. Although the steady-state membrane (pH) was predicted by the new model to be lower than that predicted by the previous model, the ( pH) versus glucose curves produced by the two models were similar. The predicted steady-state membrane (pH) was found by the new model to be most affected by changes in the buffer concentration and diffusivity. Contrary to findings of the previous model, the steady-state (pH) was unaffected by the membrane's amine content. However, the amine content of the membrane was the most important factor affecting the transient behavior of the membrane (pH) . The time to reach steady-state with an amine content typically used for a glucose-sensitive membrane was in the order of hours. To achieve response times in the range of minutes rather than hours, both the model and experimental observations show that the concentration of fixed titrateable groups must be minimized.
Journal of Controlled Release, 1992
A theorstical model has been developed to evaluate possib!e designs for au insulin delivery syste... more A theorstical model has been developed to evaluate possib!e designs for au insulin delivery system which is responsive to glucose. The system is based upon a glucose sensitive hydroge! containing immobilized glucose oxidase and catalase. The requirement of molecular oxygen for the enzymic reactions ordinarily limits the response of the system to levels of glucose (less than 50 mg%) below the patho-physiological range (50-1000 mg%). The thearetical model developed dencribcs the tzansiant as weii as steady state diffusion and reaction of glucose, oxygen and &conic acid within the macroporous hydrogel for various designs chosen to alleviate the oxygen limitation. The designs are evaluated in terms of the average pH change within the gel in response to increasing concentrations of gIucase. The model predicts the oxygen limitation that has been seen experimentally for a thin ghxose sensitive membrane (GSM ). The modeling studies have also revealed four device configurations that overccune this limitation. Each design uses silicone rubber to create additional pathways for oxygen delivery to the gel interior. A thin GSM loaded with insulin and placed over a silicone rubber drum, sealed to form a pseous reservoir, was found to be sensitive to glucose concentrations in the O-500 rug% range. A second design using a silicone rubber tube tilled with GSM (also preloaded with insulin), exhibited glucose sensitivity in the O-400 mg% range. A combined design, with a central gaseous reservoir bounded by plugs of GSM in a silicone rubber tube, retained a linear response even in the 300-500 mgO/a range. Finally a fourth design, with an insulin loaded GSM sandwiched between two circular tilms of silicone rubber sheeting, was sensitive in the O-200 mg% range. Further considerations of response time, insulin loading and delivery capacity are also discussed. and action. Current clinical trials to test whether good metabolic control can prevent the long term complications of diabetes include intensified conventional therapy with multiple daily injections and continuous subcutaneous insulin infusion with external or implanted pumps [ 1,2].
Journal of Colloid and Interface Science, 1996
plasma bulk composition, contact time, and the relative sur-Adsorption of bovine fibrinogen from ... more plasma bulk composition, contact time, and the relative sur-Adsorption of bovine fibrinogen from dilute plasma and human face activities of the plasma proteins (4, 5). For example, serum albumin (HSA) from buffered HSA solution to low-temperadsorption of fibrinogen to many substrates initially inature isotropic (LTI) pyrolytic carbon (Pyrolite) and polycreases with adsorption time, but quickly decreases, i.e., etherurethane urea (Biomer) was measured. Whereas fibrinogen passes through a maximum at early times of blood contact adsorption to Biomer passed through a maximum at intermediate . Further, maximal fibrinogen adsorption to many polyplasma dilution (a typical Vroman peak), maximal adsorption to meric materials occurs from intermediate plasma dilution
Journal of Colloid and Interface Science, 1989
The adsorption of baboon fibrinogen from buffer to glass, silicone rubber (SR), polyethylene (PE)... more The adsorption of baboon fibrinogen from buffer to glass, silicone rubber (SR), polyethylene (PE), and polystyrene (PS) was measured after 2 h at 37°C using ~25I-labeled fibrinogen. Adsorption was greatest on PS ( ~550 ng/cm 2) followed by PE (~480 ng/cm2), SR (~440 ng/cm2), and finally glass ( ~240 ng/cm2). The strength of attachment of fibrinogen to these materials was also examined by measuring the elutability of preadsorbed fibrinogen molecules by undiluted plasma as a function of adsorption time. Much of the fibrinogen adsorbed for only 1 min to SR, PE, and PS was displaced by plasma whereas only a small fraction of the fibrinogen adsorbed for 1 h could be displaced from these materials. On glass, practically all of the preadsorbed fibrinogen was displaced from the surface by plasma, independent of the adsorption time. The conversion or transition of adsorbed fibrinogen molecules from a weakly bound (displaceable) to tightly bound (nondisplaceable) state occurred most rapidly on PS followed by SR and PE. Estimates for the fibrinogen transition and displacement rate constants were evaluated from experimental data. A model describing fibrinogen adsorption from plasma was developed and solved analytically. The model predicts maxima in fibrinogen adsorption from plasma as both a function of contact time and plasma dilution on each of the materials studied and so appears to account for the presence of a Vroman effect as well as its variation with substrate surface chemistry.
Journal of Colloid and Interface Science, 1986
The adsorption of proteins to polymers is typically irreversible. Even the use of detergents does... more The adsorption of proteins to polymers is typically irreversible. Even the use of detergents does not elute all the adsorbed protein from all polymers. The fundamental reasons for such apparently tight protein binding are not well understood, so a study of several aspects of elution was undertaken. A method for examining the interaction of proteins at the protein/polymer interface using SDS elutability as a measure of the protein-surface interaction strength was developed. The effects of polymer type, elution agent, elution conditions, protein type, protein concentration, sample age, and storage temperature on elutability were examined. The results show that protein elutability from polymers decreases slowly over a period of days at 4°C but proceeds much more rapidly at elevated temperatures. The results indicate that protein denaturation may be responsible for both the initial incomplete elution and the decreases in SDS elutability of fibrinogen and albumin from polymers with time.
Journal of Colloid and Interface Science, 1990
ABSTRACT
Journal of Colloid and Interface Science, 1988
ABSTRACT
Journal of Biomedical Materials Research, 1990
Changes in the fibrinogen molecule after its adsorption to biomaterials may be important in deter... more Changes in the fibrinogen molecule after its adsorption to biomaterials may be important in determining blood compatibility. Previously, postadsorptive transitions were detected by measuring the elutability of adsorbed proteins with surfactant solutions. The elutability decreased with increased residence time, suggesting that protein-surface interactions increased with residence time. In this study, we have examined the effects of polymer structure and composition, chain mobility, and hydrophobicity on the postadsorptive transitions of fibrinogen. Glassy, rigid polymers showed high fibrinogen adsorption, regardless of whether the polymer was hydrophilic or hydrophobic. However, the binding strength (as measured by elutability) was much lower on hydrophilic polymers and oxygen-containing hydrophobic polymers. Short-term transitions, requiring 2 h or less after adsorption, were observed only on hydrophobic polymers that contained no oxygen. More gradual transitions were observed on hydrophobic polymers containing oxygen, but only after a lag time of 1-4 h.
Journal of Biomedical Materials Research, 1988
The interaction of cells with solid surfaces is important in many settings, including the respons... more The interaction of cells with solid surfaces is important in many settings, including the response of tissue to implanted materials. Protein adsorption to the surface plays a critical role in controlling cell interactions with surfaces. However, few comprehensive studies of both cell behavior and protein adsorption in complex protein mixtures (e.g., serum) have been done so the connection between these events is not well understood. In particular, methods to systematically perturb both protein adsorption and cell behavior in order to understand their relationship have been lacking. To induce changes in cell and protein behavior, the effects of serum dilution and substrate surface chemistry were studied. Surface chemistry was varied by using a series of polymers and copolymers of hydroxyethyl methacrylate (HEMA) and ethylmethacrylate (EMA) varying in their hydrophobic/hydrophilic balance. Large changes in cell spreading and fibronectin adsorption were observed when either serum concentration or polymer type was varied. The spreading of 3T3 cells in serum was found to be well correlated with the amount of fibronectin adsorption to the substrates. Attachment was not correlated with fibronectin adsorption, especially on glass preadsorbed with diluted serum. For 3T3 cells and perhaps other cells that have a receptor for a protein which is present in the medium, the amount of adsorption of this protein to the substrate appears to be a critical factor controlling cell interactions with the substrate.
Journal of Biomedical Materials Research, 2001
Monocytes and macrophages play critical roles in inflammatory responses to implanted biomaterials... more Monocytes and macrophages play critical roles in inflammatory responses to implanted biomaterials. Monocyte adhesion may lead to macrophage activation and the foreign body response. We report that surface chemistry, preadsorbed proteins, and adhesion time all play important roles during monocyte adhesion in vitro. The surface chemistry of tissue culture polystyrene (TCPS), polystyrene, Primaria, and ultra low attachment (ULA) used for adhesion studies was characterized by electron spectroscopy for chemical analysis. Fibrinogen adsorption measured by (125)I-labeled fibrinogen was the lowest on ULA, higher on TCPS, and the highest on polystyrene or Primaria. Monocyte adhesion on protein preadsorbed surfaces for 2 h or 1 day was measured with a lactate-dehydrogenase method. Monocyte adhesion decreased over time. The ability of preadsorbed proteins to modulate monocyte adhesion was surface dependent. Adhesion was the lowest on ULA, higher and similar on TCPS or polystyrene, and the highest on Primaria. Monocyte adhesion on plasma or fibrinogen adsorbed surfaces correlated positively and linearly to the amount of adsorbed fibrinogen. Preadsorbed fibronectin, immunoglobulin G, plasma, or serum also promoted adhesion compared with albumin preadsorbed or uncoated surfaces. Overall, biomaterial surface chemistry, the type and amount of adsorbed proteins, and adhesion time all affected monocyte adhesion in vitro.
Journal of Colloid and Interface Science, Aug 1, 1988
J Biomater Sci Polym Ed, 1995
Fibrinogen adsorbed to polymeric surfaces and then allowed to reside on the surface while it is k... more Fibrinogen adsorbed to polymeric surfaces and then allowed to reside on the surface while it is kept in a buffer solution for a period of time (the 'residence time') undergoes postadsorptive changes that decrease its SDS elutability, displaceability by plasma, polyclonal antifibrinogen binding, and ability to support platelet adhesion (summarized in Chinn et al. J. Biomed. Mater. Res. 26, 757 (1992)). In order to better understand the nature of the changes in adsorbed fibrinogen, the binding of ten different monoclonal antifibrinogen molecules to fibrinogen adsorbed from plasma to Biomer and several other surfaces has been measured after increasing residence time in buffer. Three of the monoclonal antibodies used bind to sequences that have been implicated in platelet binding to fibrinogen. One of these (M1) binds to the C-terminal region of the gamma chain (402-411), another (R1) binds to the N-terminal region of the A alpha chain containing an RGDF sequence (95-98), and the third (R2) binds to the C-terminal region of the A alpha chain containing an RGDS sequence (572-575). Two other antibodies (P1 and K4) also bind to the C-terminal region of the gamma chain (373-385 and 392-406, respectively). Five other antibodies that bind to other regions in fibrinogen were also used. Two of the antibodies (K4 and P1) are also known to be sensitive to conformational changes in the fibrinogen molecule. The binding of the various antibodies changed with residence time in ways that were highly dependent on the particular antibody. The binding of some antibodies was very stable with respect to residence time, others rose with time, some declined with residence time and one appears to pass through a maximum. However, none of the changes in antibody binding were nearly as fast as has been observed for the changes in platelet binding reported previously. Binding to the platelet binding region near the gamma chain C-terminal region either did not change with residence time (M1), increased with residence time (K4), or else decreased more slowly than observed for platelets (P1). Binding of the antibodies to the RGD sequences near the N-terminus of the A alpha chain (95-98) was very low initially but increased with residence time, while the binding to the RGD sequence near the C-terminus of the A alpha chain (572-575) increased slightly at short residence times but then declined substantially after longer residence times. Thus, the changes in the expression of the putative platelet binding domains do not correlate with the declines in platelet binding to plasma preadsorbed Biomer.(ABSTRACT TRUNCATED AT 400 WORDS)
Journal of Biomedical Materials Research Part a, Dec 15, 2010
The role of complement C3 in mediating adhesion of monocytes to plasma deposited tetraglyme surfa... more The role of complement C3 in mediating adhesion of monocytes to plasma deposited tetraglyme surfaces was studied. Although fibrinogen (Fg) is usually considered the main factor in mediating phagocyte attachment, plasma deposited PEO-like tetraethylene glycol dimethyl ether (tetraglyme) coatings that have ultra-low Fg adsorption (< 10 ng/cm 2 ) from low concentration solutions and low monocyte adhesion in vitro still show high phagocyte adhesion after short implantations and later become encapsulated when tested in vivo. To test whether higher Fg adsorption under in vivo conditions could explain the higher in vivo reactivity, we again measured the resistance of tetraglyme films to Fg adsorption. We found a surprising and previously unreported increased amount of adsorbed Fg on tetraglyme surfaces from higher concentration protein solutions. However, monocyte adhesion to tetraglyme did not markedly increase despite the increased Fg adsorption. We thus suspected proteins other than Fg must be responsible for the increased in vivo reactivity. We found that on tetraglyme pre-adsorbed with C3-depleted serum, monocyte adhesion was greatly reduced as compared to samples adsorbed with normal serum. Addition of exogenous pure C3 to the serum used to pre-adsorb the surfaces restored monocyte adhesion to tetraglyme coatings. While Fg clearly plays an important role in mediating monocyte adhesion to tetraglyme surfaces, the results show an additional role for adsorbed C3 in monocyte adhesion.
J Biomater Sci Polym Ed, 1992
A direct enzyme-linked immunosorbent assay (ELISA), using a polyclonal anti-fibrinogen conjugated... more A direct enzyme-linked immunosorbent assay (ELISA), using a polyclonal anti-fibrinogen conjugated to horseradish peroxidase, was used to detect fibrinogen adsorption from blood plasma to ten different materials. Adsorption was also measured with [125I]-fibrinogen. The materials studied included glass, Biomer, Immulon I, and a series of hydroxyethylmethacrylate (HEMA) and ethylmethacrylate (EMA) co-polymers. For all the materials studied, the results from the ELISA technique closely paralleled those obtained using [125I]-fibrinogen. The cross-reactivity of the antibody with proteins other than fibrinogen was generally small. Both experimental methods detected the presence of a maximum in fibrinogen adsorption (as a function of the plasma dilution) to the more hydrophobic materials. For all but two HEMA/EMA co-polymers, a linear correlation between the ELISA and [125I]-fibronogen measurements was indicated by inspection of cross plots as well as by a statistical test.
Journal of Biomedical Materials Research, Oct 1, 1995
Images of the Twenty-First Century. Proceedings of the Annual International Engineering in Medicine and Biology Society, 1989
A computer model of a glucose-sensitive membrane device that includes an additional flux of oxyge... more A computer model of a glucose-sensitive membrane device that includes an additional flux of oxygen from the insulin reservoir to alleviate the oxygen depletion problem has been developed. The response of the membrane with the additional flux of oxygen shows a larger pH drop and a greater sensitivity to glucose than in the case where there is no flux of oxygen from the insulin reservoir. The sensitivity to glucose is shown to increase to 200 mg% glucose as compared to the no-flux limit of 100 mg% with the pH drop increasing as well
Journal of Laboratory and Clinical Medicine
To study mechanisms of complex thrombus formation in vivo, and to compare the relative antithromb... more To study mechanisms of complex thrombus formation in vivo, and to compare the relative antithrombotic effects of anticoagulants and antiplatelet agents, a model was developed in baboons. Segments of collagen-coated tubing followed by two sequentially placed expansion chambers exhibiting disturbed flow patterns were exposed to native blood under laminar flow conditions. The device was incorporated for 1 hour into an exteriorized arteriovenous shunt in baboons under controlled blood flow (20 ml/min). Morphologic evaluation by scanning electron microscopy showed that thrombi associated with collagen were relatively rich in platelets but thrombi in the chambers were rich in fibrin and red cells. Deposition of indium 111-labeled platelets was continuously measured with a scintillation camera. Platelet deposition increased in a linear (collagen-coated segment) or exponential (chambers 1 and 2) fashion over time, with values after 40 minutes averaging 24.1 +/- 3.3 x 10(8) platelets (collagen segment), 16.7 +/- 3.4 x 10(8) platelets (chamber 1), and 8.4 +/- 2.4 x 10(8) platelets (chamber 2). Total fibrinogen deposition after 40 minutes was determined by using iodine 125-labeled baboon fibrinogen and averaged 0.58 +/- 0.14 mg in the collagen segment, 1.51 +/- 0.27 mg in chamber 1, and 0.95 +/- 0.25 mg in chamber 2. Plasma levels of beta-thromboglobulin (beta TG), platelet-factor 4 (PF4), and fibrinopeptide A (FPA) increased fourfold to fivefold after 60 minutes of blood exposure to the thrombotic device. Platelet deposition onto the collagen segment, chamber 1, and chamber 2 was linearly dependent on the circulating platelet count. Platelet accumulation in chamber 1 and chamber 2 was also dependent on the presence of the proximal collagen segment. An anticoagulating dose of standard heparin decreased platelet deposition in the chambers (p less than 0.05) but did not decrease deposition onto the collagen segment. Although beta TG and PF4 levels remained elevated after the administration of standard heparin, the elevation in plasma FPA was interrupted. Further evidence that the thrombotic process was dependent on platelets was provided by the finding that prostaglandin I2 at high concentration (35 ng/ml) decreased platelet deposition onto the collagen segment and in chambers 1 and 2, decreased beta TG and PF4 release, and reduced FPA formation. The combination of standard heparin and PGI2 produced the most potent inhibition of platelet thrombus formation and prevented the increases in plasma PF4, beta TG and FPA.
Investigative Ophthalmology & Visual Science
The outgrowth of corneal epithelial cells onto a polymeric substrate is expected to be the primar... more The outgrowth of corneal epithelial cells onto a polymeric substrate is expected to be the primary event in the epithelialization of a synthetic corneal graft. To study the effects of polymer surface properties on corneal epithelial cell outgrowth, a quantitative in vitro cell outgrowth assay was done. Polymers with systematic variations in hydroxyl content were used as outgrowth substrates. These polymers were characterized by electron spectroscopy for chemical analysis for elemental surface composition and by captive air-bubble contact angle for surface wettability. Circular corneal buttons were punched from excised rabbit corneas, placed onto these substrates, and incubated in a hormonally enriched culture medium. Outgrowth of the epithelial cells was allowed to proceed onto the substrates for 4 days. The outgrowth areas were measured, and an outgrowth index was calculated for each substrate by comparing it with tissue culture polystyrene substrate. The highest outgrowth generally occurred on substrates with intermediate wettabilities (captive air-bubble contact angles of approximately 45-75°); it was less on substrates of lower or higher wettabilities. Protein coatings of albumin, immunoglobulin G (IgG), fibronectin, and culture medium were found to lower the wettabilities of native substrates. Albumin and IgG precoating were shown to reduce epithelial cell outgrowth; fibronectin precoating was shown to improve outgrowth on most substrates. These results suggest that epithelial cell outgrowth is influenced by both substrate and protein interactions. Invest Ophthalmol
Journal of Molecular Recognition, 1996
Two peptides from the ligand-binding site of the platelet receptor GPIIb/IIIa, residues 296-306 o... more Two peptides from the ligand-binding site of the platelet receptor GPIIb/IIIa, residues 296-306 of GPIIb, designated B12 by D'Souza et al. (1991), and 300-311 of GPIIb, designated G13 by Taylor et al., (1992), as well as two control peptides, designated C14 and C20, were adsorbed to treated polystyrene substrates. Fibrinogen adsorption to the peptide-coated substrates was characterized. The specificity of I-125 labeled fibrinogen binding to the peptide-coated substrates was investigated by measuring the amount of fibrinogen adsorbed to each substrate and the inhibition of fibrinogen binding by RGDS peptide, bovine serum albumin, a divalent ion chelator (ethylene diamine tetra-acetic acid disodium salt), unlabeled fibrinogen and the B12 peptide. The results show that non-specific binding of fibrinogen to the B12-coated substrate is predominant under most conditions. Binding of monoclonal antibodies to fibrinogen adsorbed to the peptide coated substrates was characterized. The failure of several antibodies to bind fibrinogen adsorbed to the B12 substrate suggested that adsorption of fibrinogen to the B12-coated substrate alters its conformation relative to fibrinogen adsorbed to the bare substrate.
Journal of Controlled Release, 1993
Explicit expressions for the hydronium ion transport have been added to an existing theoretical m... more Explicit expressions for the hydronium ion transport have been added to an existing theoretical model describing the steady-state and transient behavior of glucose-sensitive membranes. The glucose sensitive membrane is a hydrogel containing pendant amine groups and immobilized glucose oxidase creating a membrane which swells in response to changes in glucose concentration. The extent of membrane swelling and the time required for swelling depend upon the pH within the membrane. The membrane pH is a function of the rate of production of gluconic acid from glucose as well as the transport rate of the hydronium ion. Although the steady-state membrane (pH) was predicted by the new model to be lower than that predicted by the previous model, the ( pH) versus glucose curves produced by the two models were similar. The predicted steady-state membrane (pH) was found by the new model to be most affected by changes in the buffer concentration and diffusivity. Contrary to findings of the previous model, the steady-state (pH) was unaffected by the membrane's amine content. However, the amine content of the membrane was the most important factor affecting the transient behavior of the membrane (pH) . The time to reach steady-state with an amine content typically used for a glucose-sensitive membrane was in the order of hours. To achieve response times in the range of minutes rather than hours, both the model and experimental observations show that the concentration of fixed titrateable groups must be minimized.
Journal of Controlled Release, 1992
A theorstical model has been developed to evaluate possib!e designs for au insulin delivery syste... more A theorstical model has been developed to evaluate possib!e designs for au insulin delivery system which is responsive to glucose. The system is based upon a glucose sensitive hydroge! containing immobilized glucose oxidase and catalase. The requirement of molecular oxygen for the enzymic reactions ordinarily limits the response of the system to levels of glucose (less than 50 mg%) below the patho-physiological range (50-1000 mg%). The thearetical model developed dencribcs the tzansiant as weii as steady state diffusion and reaction of glucose, oxygen and &conic acid within the macroporous hydrogel for various designs chosen to alleviate the oxygen limitation. The designs are evaluated in terms of the average pH change within the gel in response to increasing concentrations of gIucase. The model predicts the oxygen limitation that has been seen experimentally for a thin ghxose sensitive membrane (GSM ). The modeling studies have also revealed four device configurations that overccune this limitation. Each design uses silicone rubber to create additional pathways for oxygen delivery to the gel interior. A thin GSM loaded with insulin and placed over a silicone rubber drum, sealed to form a pseous reservoir, was found to be sensitive to glucose concentrations in the O-500 rug% range. A second design using a silicone rubber tube tilled with GSM (also preloaded with insulin), exhibited glucose sensitivity in the O-400 mg% range. A combined design, with a central gaseous reservoir bounded by plugs of GSM in a silicone rubber tube, retained a linear response even in the 300-500 mgO/a range. Finally a fourth design, with an insulin loaded GSM sandwiched between two circular tilms of silicone rubber sheeting, was sensitive in the O-200 mg% range. Further considerations of response time, insulin loading and delivery capacity are also discussed. and action. Current clinical trials to test whether good metabolic control can prevent the long term complications of diabetes include intensified conventional therapy with multiple daily injections and continuous subcutaneous insulin infusion with external or implanted pumps [ 1,2].
Journal of Colloid and Interface Science, 1996
plasma bulk composition, contact time, and the relative sur-Adsorption of bovine fibrinogen from ... more plasma bulk composition, contact time, and the relative sur-Adsorption of bovine fibrinogen from dilute plasma and human face activities of the plasma proteins (4, 5). For example, serum albumin (HSA) from buffered HSA solution to low-temperadsorption of fibrinogen to many substrates initially inature isotropic (LTI) pyrolytic carbon (Pyrolite) and polycreases with adsorption time, but quickly decreases, i.e., etherurethane urea (Biomer) was measured. Whereas fibrinogen passes through a maximum at early times of blood contact adsorption to Biomer passed through a maximum at intermediate . Further, maximal fibrinogen adsorption to many polyplasma dilution (a typical Vroman peak), maximal adsorption to meric materials occurs from intermediate plasma dilution
Journal of Colloid and Interface Science, 1989
The adsorption of baboon fibrinogen from buffer to glass, silicone rubber (SR), polyethylene (PE)... more The adsorption of baboon fibrinogen from buffer to glass, silicone rubber (SR), polyethylene (PE), and polystyrene (PS) was measured after 2 h at 37°C using ~25I-labeled fibrinogen. Adsorption was greatest on PS ( ~550 ng/cm 2) followed by PE (~480 ng/cm2), SR (~440 ng/cm2), and finally glass ( ~240 ng/cm2). The strength of attachment of fibrinogen to these materials was also examined by measuring the elutability of preadsorbed fibrinogen molecules by undiluted plasma as a function of adsorption time. Much of the fibrinogen adsorbed for only 1 min to SR, PE, and PS was displaced by plasma whereas only a small fraction of the fibrinogen adsorbed for 1 h could be displaced from these materials. On glass, practically all of the preadsorbed fibrinogen was displaced from the surface by plasma, independent of the adsorption time. The conversion or transition of adsorbed fibrinogen molecules from a weakly bound (displaceable) to tightly bound (nondisplaceable) state occurred most rapidly on PS followed by SR and PE. Estimates for the fibrinogen transition and displacement rate constants were evaluated from experimental data. A model describing fibrinogen adsorption from plasma was developed and solved analytically. The model predicts maxima in fibrinogen adsorption from plasma as both a function of contact time and plasma dilution on each of the materials studied and so appears to account for the presence of a Vroman effect as well as its variation with substrate surface chemistry.
Journal of Colloid and Interface Science, 1986
The adsorption of proteins to polymers is typically irreversible. Even the use of detergents does... more The adsorption of proteins to polymers is typically irreversible. Even the use of detergents does not elute all the adsorbed protein from all polymers. The fundamental reasons for such apparently tight protein binding are not well understood, so a study of several aspects of elution was undertaken. A method for examining the interaction of proteins at the protein/polymer interface using SDS elutability as a measure of the protein-surface interaction strength was developed. The effects of polymer type, elution agent, elution conditions, protein type, protein concentration, sample age, and storage temperature on elutability were examined. The results show that protein elutability from polymers decreases slowly over a period of days at 4°C but proceeds much more rapidly at elevated temperatures. The results indicate that protein denaturation may be responsible for both the initial incomplete elution and the decreases in SDS elutability of fibrinogen and albumin from polymers with time.
Journal of Colloid and Interface Science, 1990
ABSTRACT
Journal of Colloid and Interface Science, 1988
ABSTRACT
Journal of Biomedical Materials Research, 1990
Changes in the fibrinogen molecule after its adsorption to biomaterials may be important in deter... more Changes in the fibrinogen molecule after its adsorption to biomaterials may be important in determining blood compatibility. Previously, postadsorptive transitions were detected by measuring the elutability of adsorbed proteins with surfactant solutions. The elutability decreased with increased residence time, suggesting that protein-surface interactions increased with residence time. In this study, we have examined the effects of polymer structure and composition, chain mobility, and hydrophobicity on the postadsorptive transitions of fibrinogen. Glassy, rigid polymers showed high fibrinogen adsorption, regardless of whether the polymer was hydrophilic or hydrophobic. However, the binding strength (as measured by elutability) was much lower on hydrophilic polymers and oxygen-containing hydrophobic polymers. Short-term transitions, requiring 2 h or less after adsorption, were observed only on hydrophobic polymers that contained no oxygen. More gradual transitions were observed on hydrophobic polymers containing oxygen, but only after a lag time of 1-4 h.
Journal of Biomedical Materials Research, 1988
The interaction of cells with solid surfaces is important in many settings, including the respons... more The interaction of cells with solid surfaces is important in many settings, including the response of tissue to implanted materials. Protein adsorption to the surface plays a critical role in controlling cell interactions with surfaces. However, few comprehensive studies of both cell behavior and protein adsorption in complex protein mixtures (e.g., serum) have been done so the connection between these events is not well understood. In particular, methods to systematically perturb both protein adsorption and cell behavior in order to understand their relationship have been lacking. To induce changes in cell and protein behavior, the effects of serum dilution and substrate surface chemistry were studied. Surface chemistry was varied by using a series of polymers and copolymers of hydroxyethyl methacrylate (HEMA) and ethylmethacrylate (EMA) varying in their hydrophobic/hydrophilic balance. Large changes in cell spreading and fibronectin adsorption were observed when either serum concentration or polymer type was varied. The spreading of 3T3 cells in serum was found to be well correlated with the amount of fibronectin adsorption to the substrates. Attachment was not correlated with fibronectin adsorption, especially on glass preadsorbed with diluted serum. For 3T3 cells and perhaps other cells that have a receptor for a protein which is present in the medium, the amount of adsorption of this protein to the substrate appears to be a critical factor controlling cell interactions with the substrate.
Journal of Biomedical Materials Research, 2001
Monocytes and macrophages play critical roles in inflammatory responses to implanted biomaterials... more Monocytes and macrophages play critical roles in inflammatory responses to implanted biomaterials. Monocyte adhesion may lead to macrophage activation and the foreign body response. We report that surface chemistry, preadsorbed proteins, and adhesion time all play important roles during monocyte adhesion in vitro. The surface chemistry of tissue culture polystyrene (TCPS), polystyrene, Primaria, and ultra low attachment (ULA) used for adhesion studies was characterized by electron spectroscopy for chemical analysis. Fibrinogen adsorption measured by (125)I-labeled fibrinogen was the lowest on ULA, higher on TCPS, and the highest on polystyrene or Primaria. Monocyte adhesion on protein preadsorbed surfaces for 2 h or 1 day was measured with a lactate-dehydrogenase method. Monocyte adhesion decreased over time. The ability of preadsorbed proteins to modulate monocyte adhesion was surface dependent. Adhesion was the lowest on ULA, higher and similar on TCPS or polystyrene, and the highest on Primaria. Monocyte adhesion on plasma or fibrinogen adsorbed surfaces correlated positively and linearly to the amount of adsorbed fibrinogen. Preadsorbed fibronectin, immunoglobulin G, plasma, or serum also promoted adhesion compared with albumin preadsorbed or uncoated surfaces. Overall, biomaterial surface chemistry, the type and amount of adsorbed proteins, and adhesion time all affected monocyte adhesion in vitro.