An in vitro cytotoxicity study of aldehyde-treated pig dermal collagen

The cytotoxicity of aldehyde-treated collagen was assayed by measuring 3H-thymidine incorporation in adult human skin fibroblasts grown in tissue culture for 1 or 3 days in the presence of pig dermal collagen cross-linked with formaldehyde or glutaraldehyde. A comparison was also made with collagen preparations washed for 2 weeks either at 15 degrees throughout or partly at 15 degrees and partly at 37 degrees. Collagen treated with both formaldehyde and glutaraldehyde proved increasingly toxic with increase in the concentrations of aldehyde used. While the maximum toxic effect was observed after 1 day culture in formaldehyde-treated collagen, with thymidine uptake ranging from 4-48% of control values with 5-0.1% formaldehyde and a 15 degrees wash, the toxic effect of glutaraldehyde treatment increased with longer exposure and at 3 days thymidine uptake ranged from 3-40% of control values with 0.05-0.001% glutaraldehyde and washing at 15 degrees. Washing partly at 37 degrees significantly reduced toxicity, the differences in thymidine uptake as compared with washing at 15 degrees alone ranging from 34-50% with 1 and 0.3% formaldehyde respectively in 1 day cultures and from 14-37% with 0.02 and 0.005% glutaraldehyde in 3 day cultures. While fibroblasts actively grew and migrated when seeded on non-cross-linked collagen, only limited cell survival occurred on aldehyde-treated collagen.     

Effect of aldehyde cross-linking on human dermal collagen implants in the rat.

The fate of s.c. implants of fibrous, trypsin-purified human dermal collagen and collagen cross-linked with formaldehyde and glutaraldehyde has been studied in rats. Dermal collagen, and untreated skin implants, underwent resorption associated with a pronounced round-cell reaction. While collagen implants cross-linked with solutions of 0.04% and 0.08% formaldehyde became reduced in size, those cross-linked with 1% formaldehyde and 0.01%, 0.02% and 0.04% glutaraldehyde, although undergoing some collagen remodelling, retained their original size over the 25-week period of study. At 5 weeks the aldehyde cross-linked implants showed their greatest cellularity, reaching a lower, more stable cell population by 18 weeks. More round cells were seen at 5 weeks, particularly after formaldehyde cross-linking, than a later times when few were present. The results indicate that aldehyde-stabilized preparations of heterograft dermal collagen could have applications in the repair of tissue defects in man.     

Effect of aldehyde cross-linking on human dermal collagen implants in the rat

The fate of s.c. implants of fibrous, trypsin-purified human dermal collagen and collagen cross-linked with formaldehyde and glutaraldehyde has been studied in rats. Dermal collagen, and untreated skin implants, underwent resorption associated with a pronounced round-cell reaction. While collagen implants cross-linked with solutions of 0.04% and 0.08% formaldehyde became reduced in size, those cross-linked with 1% formaldehyde and 0.01%, 0.02% and 0.04% glutaraldehyde, although undergoing some collagen remodelling, retained their original size over the 25-week period of study. At 5 weeks the aldehyde cross-linked implants showed their greatest cellularity, reaching a lower, more stable cell population by 18 weeks. More round cells were seen at 5 weeks, particularly after formaldehyde cross-linking, than a later times when few were present. The results indicate that aldehyde-stabilized preparations of heterograft dermal collagen could have applications in the repair of tissue defects in man.     

Inhibition of gingival fibroblast growth by Bacteroides gingivalis.

Human gingival fibroblasts were exposed in culture to cell extracts of different black-pigmented Bacteroides species, and their growth was monitored by determining thymidine uptake and counting cells. Of the Bacteroides species tested (B. gingivalis, B. asaccharolyticus, and B. intermedius), B. gingivalis gave the extract with the strongest inhibitory effect on fibroblast thymidine uptake. Linear inhibition reaching 80% of the control level was obtained with a dose of 100 micrograms of B. gingivalis extract protein per ml. The effect of B. asaccharolyticus resembled that of B. gingivalis, but even at the highest dose tested B. intermedius had only a slight inhibitory effect. When fibroblasts were counted after 2- and 4-day exposures to B. gingivalis extracts, a clear depression in the number of fibroblasts was found. The effects of extracts obtained from early and late growth phases of B. gingivalis cultures were similar. A fraction of B. gingivalis consisting essentially of lipopolysaccharides (LPSs) was obtained by degrading the extract proteins with proteinase K. Silver staining of polyacrylamide gels revealed a LPS pattern with a molecular mass ranging from 37 to 60 kilodaltons. This LPS-rich fraction caused inhibition of thymidine uptake by gingival fibroblasts similar to that caused by the native extract alone. Thus, the inhibition of gingival fibroblast growth by B. gingivalis appeared to be LPS mediated. This inhibitory effect of B. gingivalis on oral fibroblast growth may be a virulence factor of this bacterium.     

Lysine-enhanced glutaraldehyde crosslinking of collagenous biomaterials.

Crosslinking of collagenous biomaterials currently employs the use of glutaraldehyde. The putative enhancement of glutaraldehyde crosslinking by lysine was investigated in three model systems: bovine pericardium, collagen membranes, and bovine serum albumin. Repetitive sequential treatment of bovine pericardium with glutaraldehyde and lysine and finally with formaldehyde produced a matrix which, by the two criteria used (shrinkage temperature and urea/SDS soluble collagen), was shown to be more highly crosslinked than pericardium fixed in glutaraldehyde alone. Essentially the same results were obtained when membranes prepared from pepsin-soluble pericardial collagen were subjected to sequential glutaraldehyde and lysine treatments, reaching shrinkage temperatures of more than 90 degrees C. Heart valves prepared from lysine-enhanced glutaraldehyde crosslinked bovine pericardium were tested in vitro in an accelerated fatigue tester and have been shown to behave satisfactorily after 300 million cycles. These additional crosslinks proved to be stable in saline at 37 degrees C. Studies on bovine serum albumin attempted to get an insight into the mechanisms of lysine enhancement of glutaraldehyde crosslinking by treating sequentially albumin with glutaraldehyde and lysine and analysis of the products by gel filtration and SDS-PAGE. These studies suggest that free amino groups exposed by proteins are initially reacted with glutaraldehyde and then bridged by the diamino compound (lysine) producing more extensive intermolecular crosslinking than glutaraldehyde alone.     

Enhancement of human lymphocyte responses to phytomitogens in vitro by incubation at elevated temperatures.

Incubation of human lymphocytes with PHA or Con A at 39 degrees C for 3 days caused a consistent, statistically significant, increase in [3H]thymidine uptake, compared with cultures incubated at 37 degrees C. The responses were also significantly increased after 2 days of incubation, suggesting that hyperthermia not only enhanced, but also caused an earlier onset of the mitogen response.     

Spontaneous rosette formation by mouse L cells.

Mouse fibroblasts form spontaneous rosettes with sheep erythrocytes and with red cells from several other species. They do not form C3 or Fc rosettes, and they do not bind small amounts of aggregated gammaglobulin. Spontaneous rosette formation is maximum at 37 degress and minimum at 4 degrees. The reaction is complete in 30 min. Antisera to fibroblast membrane components do not inhibit rosette formation. Rosetting is inhibited by fixation of the cells with formaldehyde or glutaraldehyde as well as by killing with phenol or by prolonged incubation with sodium azide. The rosette reaction is inhibited by vinblastine but not by cytochalasin B. From these data on the kinetics, temperature and drug sensitivities it is suggested that T cells and fibroblasts form spontaneous rosettes by different mechanisms.     

Pathology and cell proliferation induced by intra-nasal instillation of aldehydes in the rat: comparison of glutaraldehyde and formaldehyde

The relative toxicities of formaldehyde and glutaraldehyde to the rat nasal epithelium were determined following intra-nasal instillation of aqueous solutions of these compounds into one nostril of male Fischer 344 (F-344) rats. Lesions identical in appearance to those resulting from acute inhalation exposure to formaldehyde were induced by both compounds in a concentration-dependent manner. Treatments included India ink or 1 M methylene blue (for instillation deposition studies); sterile saline (vehicle control); 40, 200, 400, and 800 mM formaldehyde; and 10, 20, and 40 mM glutaraldehyde. Dye-treated rats were sacrificed immediately, and nasal passages were examined to determine the localization of instilled materials. Three days after treatment, all other animals received a single ip injection of 5-bromo-2'-deoxyuridine 2 hr prior to sacrifice, and the nasal passages were prepared for histopathology and cell proliferation studies. While sterile saline and 10 mM glutaraldehyde induced no significant epithelial changes, 20 and 40 mM glutaraldehyde induced extensive lesions in the treated side of the nose. Aldehyde-induced lesions included inflammation, epithelial degeneration, respiratory epithelial hypertrophy, and squamous metaplasia in association with marked increases (3-8-fold) in labeling index for both compounds. Formaldehyde induced similar lesions but required concentrations of 200 mM or more to elicit a toxic response. Thus, glutaraldehyde is approximately an order of magnitude more toxic to the nasal epithelium than formaldehyde. These studies also indicate that the nose is very resistant to the aldehydes studied, requiring instillation of millimolar concentrations before toxic responses occurred.     

Characterization of collagen gel solutions and collagen matrices for cell culture

The influence of glutaraldehyde as a crosslinking agent to increase the strength of collagen matrices for cell culture was examined in this study. Collagen solutions of 1% were treated with different concentrations (0-0.2%) of glutaraldehyde for 24 h. The viscoelasticity of the resulting collagen gel solution was measured using dynamic mechanical analysis (DMA), which demonstrated that all collagen gel solutions examined followed the same model pattern. The creep compliance model of Voigt-Kelvin satisfactorily described the change of viscoelasticity expressed by these collagen gel solutions. These crosslinked collagen gel solutions were freeze-dried to form a matrix with a thickness of about 0.2-0.3 mm. The break modulus of these collagen matrices measured by DMA revealed that the higher the degree of crosslinking. the higher the break modulus. The compatibility of fibroblasts isolated from nude mouse skin with these collagen matrices was found to be acceptable at a cell density of 3 x 10(5) cells/cm2 with no contraction, even when using a concentration of glutaraldehyde of up to 0.2%.     

Calcification of subcutaneously implanted type I collagen sponges. Effects of formaldehyde and glutaraldehyde pretreatments.

Although collagen-containing implants are widely used in various surgical applications, there has been relatively little attention paid to the possibility that this type of biomaterial may undergo pathologic calcification which could compromise its function. The present study reports for the first time the calcification of a series of implants of purified collagen sponges prepared with graded degrees of aldehyde-induced cross-linkages (assessed by shrinkage-temperature, wetting time, and collagenase digestibility). Type I collagen sponges were pretreated with either glutaraldehyde (0.1% to 2.0% aqueous solution, for 5-180 minutes) or formaldehyde (as vapors for 15 minutes to 15 hours), and implanted subcutaneously for 21 days in weanling rats. Although specimens not pretreated with either aldehyde reagent and the formaldehyde sponges pretreated for 15 minutes were resorbed without evidence of calcification, all other aldehyde-pretreated implants mineralized. The degree of calcification did not correlate with extent of cross-linking. Formaldehyde-pretreated implants calcified more extensively (Ca2+ = 87.8 +/- 2.8 micrograms/mg, mean +/- standard error of the mean; n = 58) than did glutaraldehyde-pretreated implants (Ca2+ = 40.9 +/- 1.4 micrograms/mg; n = 52). It is concluded that both glutaraldehyde- and formaldehyde-pretreated Type I collagen sponges calcify after subdermal implantation in young rats. Although aldehyde pretreatment of Type I collagen sponge implants is a prerequisite for their eventual mineralization, the threshold level of aldehyde-induced cross-linking required to potentiate their maximal pathologic calcification is low.     

Serum effects of mitogenic reactivity in subjects with systemic lupus erythematosus, rheumatoid arthritis and scleroderma. Technical considerations and lack of correlation with anti-lymphocyte antibodies.

Methodological problems which affect the assessment of humoral effects on mitogenic reactivity include: (1) the source and concentration of serum used to support cell cultures; (2) the day to-day variability of inhibitory effects and (3) the specific activity of [3H]thymidine added to the culture. These problems were alleviated by addition of half concentration (7-5%) of pooled normal human serum to all cultures, the intoruction of anti-lymphocyte serum as a suitable internal control for monitoring the suppressability of lymphocytes and a reduction of specific activity of the [3H]thymidine to 1-3 C2/mM. Inhibitory factors were loosely bound to the lymphocyte surface and eluted off after incubation at 37 degrees C for 1 hr. Cells from twenty-five subjects and paired controls were cultured simultaneously in medium containing either 15% normal human serum (NHS) or 7-5% patient and 7-5% NHS. The cells were stimulated with various dilutions of phytohaemagglutinin, Con A or pokeweed mitogen. Lupus serums suppressed the reactivity of autologous lymphocytes to PHA and pokeweed mitogen. Serums from subjects with RA and scleroderma did not significantly inhibit blastogenesis of autologous lymphocytes. One-half of the lupus serums significantly inhibited the reactivity of homologous lymphocytes to two of three mitogens. Only one of eight scleroderma serums and none of twelve RA serums and none of twelve RA serums had this effect. All patients serums were examined for antilymphocyte antibodies by microcytotoxicity and immunofluorescent techniques. These antibodies were usually found in SLE, and were often observed in subjects with rheumatoid arthritis but not scleroderma. A firm relationship between serum suppressors of lymphocyte blastogenesis and anti-lymphocyte antibodies was not found.     

Microwave irradiation and cross-linking of collagen.

In a multifactorial experiment, dermal sheep collagen was treated in diluted glutaraldehyde solutions, 70% ethyl alcohol, Cialit 1:5000, and distilled water for 1, 3 and 5 min, respectively, in combination with microwave irradiation at different temperature settings. The shrinkage temperature indicating the degree of cross-linking achieved was then determined. Treatment in 0.65% glutaraldehyde with microwave irradiation setting for 60 degrees C resulted in the maximum shrinkage temperature within 1 min, whilst at the lower setting of 50 degrees C, the maximum shrinkage temperature for both glutaraldehyde solutions is only reached after 5 min. Neither microwave irradiation by itself, Cialit or ethyl alcohol induce cross-linking of collagen fibres. These findings are relevant for implant studies.     

Evaluation of the genotoxic potential of glutaraldehyde

The cytotoxic and genotoxic effects of glutaraldehyde were studied in vitro in the human TK6 lymphoblast cell line and in primary cultures of rat hepatocytes. TK6 lymphoblasts were exposed to glutaraldehyde for 2 hr in serum-free GSH-free media. Cytotoxic effects were observed at concentrations as low as 10 microM with only 10% cell survival at 20 microM. Alkaline elution studies indicated that glutaraldehyde-induced DNA-protein crosslinking increased linearly over the concentration range from 0 to 25 microM. Glutaraldehyde-induced mutations were assessed at the thymidine kinase locus over the same concentration range and reached a plateau at 10 microM of about six times the background mutant frequency. At equivalent levels of DNA-protein crosslinks and cytolethality, glutaraldehyde was mutagenic at approximately a one-seventh lower concentration than the rodent nasal carcinogen formaldehyde (Craft et al.; Mutation Research 176:147-155, 1987). Glutaraldehyde induced a marginal increase in unscheduled DNA synthesis in the in vitro hepatocyte DNA repair assay, but only at the two highest concentrations of 50 and 100 microM, indicating the induction of some DNA excision-repair activity. These data demonstrate that glutaraldehyde exhibits DNA-reactive genotoxic activity that may involve, at least in part, DNA-protein crosslinking in these cell culture models. These findings suggest the need to examine the potential carcinogenic activity of glutaraldehyde in appropriate inhalation studies.     

An examination of the biologic response to injectable, glutaraldehyde cross-linked collagen implants

The biologic response to injectable, glutaraldehyde cross-linked, fibrillar collagen implants in the rat subcutaneous model was shown to be a function of the concentration of glutaraldehyde used for cross-linking. The collagen was prepared from bovine hide by pepsin solubilization and reconstituted as a fibrillar suspension of 35 mg collagen/mL. Fibrillar collagen implants cross-linked with glutaraldehyde concentrations equal to or less than 0.01% exhibited a response characterized by fibroblast invasion, neovascularization and little, if any, evidence of inflammation. Implants cross-linked with 0.1 and 1.0% glutaraldehyde elicited a foreign body/giant cell reaction and varying degrees of implant erosion. The interaction of human skin fibroblasts with 0.01% glutaraldehyde cross-linked collagen in vitro was found to be dependent on the culture conditions utilized to evaluate the interaction. When the ratio of cell culture media to collagen was 20:1, cell invasion of the cross-linked preparations was observed, whereas, when this ratio was reduced to 1:1, such interactions could not be detected. Noncross-linked preparations were colonized by cells regardless of the experimental conditions used. Studies of implants in both the rat and guinea pig subcutaneous models indicated that glutaraldehyde cross-linking concentrations as low as 0.0075% provided enhanced wet weight recovery (wet weight persistence) and resistance to biologic degradation (collagen persistence) as compared to noncross-linked fibrillar collagen preparations. These cross-linked implants also exhibited a greater degree of fibroblast infiltration and vascularization. Between 30 and 60 days, some degree of calcification developed in both collagen formulations implanted in rats and guinea pigs; however, the reaction occurred with greater frequency and intensity in cross-linked preparations in guinea pigs. Calcification in the guinea pig was followed by the appearance of focal areas of ossification.     

The response of human leucocyte cultures to stimulation by tuberculin and phytohaemagglutinin measured by the uptake of radioactive thymidine

A method is described for assessing the lymphocyte transformation of human leucocyte cultures induced by phytohaemagglutinin (PHA) and tuberculin purified protein derivative (PPD) by the measurement of the uptake of [¹⁴C]thymidine. The thymidine uptake of the cultures has been expressed as percentage of the total activity added. The mean basal uptake of twenty-two unstimulated cultures was 0·96%. The mean uptake of twenty-four PHA stimulated cultures was 19·5%, whilst the mean uptake of eleven PPD stimulated cultures from seven subjects, who were strongly positive on skin testing to tuberculin, was 20·1%. A rise of thymidine uptake was detected in five out of six tuberculin negative subjects (mean uptake 3·8%). Hydrocortisone hemisuccinate in final concentrations of 5 and 50 μg/ml usually reduced the thymidine uptake of unstimulated and PHA and PPD stimulated cultures, though the total cell count was reduced in only four out of ten cultures. An increase of thymidine from 1 to 64 μg/culture resulted in twoto three-fold increase in the uptake of thymidine; further increase of thymidine added to the cultures to 256 μg/culture resulted in a decrease of thymidine uptake.     

Mouse fibroblasts in long-term culture within collagen three-dimensional scaffolds: influence of crosslinking with diphenylphosphorylazide on matrix reorganization, growth, and biosynthetic and proteolytic activities

With the rapid development of tissue engineering and gene therapy, collagen-based biomaterials frequently are used as cell transplant devices. In this study we determined the behavior of mouse fibroblasts cultured for up to 6 weeks in control sponges treated by severe dehydration and used commercially as hemostatic agents and in two sponges (DPPA 2 and 3) crosslinked by diphenylphosphorylazide, a method developed in our laboratory. Growth capacity, biosynthetic and proteolytic activities, and matrix reorganization were followed over time in cultures and compared with similar data for fibroblasts in monolayer culture on plastic and in floating or attached collagen gels. Control sponges with and without seeded mouse fibroblasts showed rapid partial denaturation or contraction, weight loss, and severe calcification (13-18% Ca) after 6 weeks. In contrast, the crosslinked sponges showed only slightly decreased size and weight, and the calcification was inhibited (0.2% Ca) in the presence of cells. Mouse fibroblasts seeded on the crosslinked sponge surface at 50,000-200,000 cells/cm(2) progressively penetrated the matrix and proliferated to give the same constant cell density after 3 weeks (around 600,000 cells/sponge). A specific, two- to threefold decrease in collagen synthesis was observed between 1 and 3 or 6 weeks, due mainly to a decrease in the fraction secreted into the medium (25-30% instead of 45-50%). No collagenase 3 activity was detected in the culture medium under any condition or time whereas 25% gelatinase A was found by gelatin zymography to be in an active form in cultures within sponges as compared with less than 10% in monolayers and more than 50% in floating collagen gel. A small amount of gelatinase B was observed after 1 week in sponge cultures and was completely absent thereafter. These results show that the biosynthetic and proteolytic behavior of mouse fibroblasts cultured in crosslinked collagen scaffolds is different from that in monolayers or in floating collagen gels and more similar to that previously described in attached collagen gels.     

Thermostable alpha-amylase conjugated antibodies as probes for immunodetection in ELISA

Thermostable alpha-amylase from B. licheniformis has been conjugated with high efficiency to goat antibodies against human, mouse, and rabbit immunoglobulins to prepare second-step reagents which can be used in Enzyme Linked Immunosorbent Assays (ELISA). Various conjugation methods, such as one- and two-step glutaraldehyde coupling and cross-linking, using heterobifunctional reagents such as sulfosuccinimidyl-4-(N-maleimidomethyl)-cyclohexane-1-carbonate (sulfo-SMCC) and N-succinimidyl-S-acetylthioacetate (SATA), yielded active alpha-amylase labeled second antibodies. Such conjugates had molecular sizes ranging between 200-300 kDa. Filter sterilized solutions of conjugates, when stored at 37 degrees C for two weeks, retained 32% of their biological activity and were thermostable even after keeping for 1 h at 90 degrees C.     

Modified hematoxylin and eosin staining method for epoxy-embedded tissue sections

A modified hematoxylin and eosin staining method for glutaraldehyde fixed, osmium tetroxide postfixed, and epoxy resin (Medcast Resin) embedded tissue is described. Two microns thick sections of human and animal tissues are treated with 4% H2O2 for 4 minutes, washed, dryed, and then flooded with Gill's III hematoxylin in a moist chamber at 37 degrees C for 90 minutes. Subsequent steps are: washing; bluing in ammonia water for 1 minutes; washing again and staining in 1% eosin Y alcoholic solution for 5 minutes; rinsing in 100% ethanol; drying and mounting with Permont. This procedure is quick, easy and successful in demonstrating both color scale and quality similar to hematoxylin and eosin stains obtained in standard paraffin sections.     

Perlecan-stimulated nodules undergo chondrogenic maturation in response to rhBMP-2 treatment in vitro

The heparan sulfate proteoglycan, perlecan, is localized to hypertrophic chondrocytes in the growth plates of long bones. Mice mutants for perlecan display severe cartilage and skeletal defects. Previously, we demonstrated that C3H10T1/2 fibroblasts cultured on perlecan stimulated extensive formation of dense nodules reminiscent of embryonic cartilaginous condensations. These nodules stain intensely with Alcian blue, and antibodies specific for collagen type II and aggrecan; however, nodules do not express collagen type X, a marker of chondrogenic maturation. In this investigation, we tested the hypothesis that addition of rhBMP-2 to perlecan-induced nodules would promote chondrogenic maturation in vitro. C3H10T1/2 fibroblasts were seeded in Lab-Tek chambered "Permanox" slides uncoated or coated with perlecan (B&D, 5 microg/well), at a density of 2 x 10(5) cells/well. The cells were maintained in CMRL-1066 media supplemented with ascorbic acid, citrate, and pyruvate (50 ng/ml). C3H10T1/2 fibroblasts seeded on perlecan-coated wells began to condense and form cell aggregates within 15 min. On the third day postplating, the media was replaced and supplemented with or without rhBMP-2 (50 ng/ml, Genetics Institute). On day 6 of culture, microscopy revealed that rhBMP-2-treated cultures had significantly proliferated; however, untreated cultures had not. By day 12 of culture, confocal microscopy revealed that perlecan-stimulated nodules treated with rhBMP-2 express a late stage marker of chondrogenesis (collagen type X). Morphologically, cells expressing collagen type X in rhBMP-2-treated nodules appear larger in diameter, relative to cells not expressing collagen type X. Cells cultured on plastic and treated with rhBMP-2 did not form nodules, but attached and spread, yielding a high-density monolayer. In response to rhBMP-2 treatment, these cells also express collagen type X. However, the appearance of collagen type X occurs at a later time point relative to the appearance of collagen type X in perlecan-stimulated nodules. Thus, perlecan-stimulated nodules do mature at a faster rate when treated with rhBMP-2 relative to monolayer cells.