Biomechanical and microstructural characteristics of a collagen film-based corneal stroma equivalent

The growth in refractive surgeries and corneal replacements has fueled interest in the development of a tissue-engineered cornea. This study characterizes the microstructure and biomechanical properties of film-based corneal stroma equivalents over time in culture. The increased collagen density in the films was hypothesized to result in improved mechanical properties both initially and over time. The microstructure of the film-based stromal equivalent was examined using atomic force microscopy and scanning electron microscopy; the mechanical properties, relaxed modulus, and ultimate tensile strength were quantified using uniaxial tensile testing. The dense, film-based stromal equivalent had a lamellae-like microstructure, which was notably different than the porous structure of sponges used previously. Seeded human corneal stromal fibroblasts remained on the surface of the film rather than migrating into the film and produced fibers of extracellular matrix with diameters of 35-75 nm. After an initial decrease during hydration, the relaxed modulus and ultimate tensile strength for fully hydrated collagen films were 0.4 +/- 0.2 MPa and 0.3 +/- 0.1 MPa, respectively. The mechanical properties of cell-seeded films mimicked those of control films. While further studies are needed to quantify the optical properties, the dense, lamellae-like structure of collagen films is a feasible scaffold for the development of tissue-engineered stroma.     

Tissue engineering of corneal stromal layer with dermal fibroblasts: phenotypic and functional switch of differentiated cells in cornea

Previously, we successfully engineered a corneal stromal layer using corneal stromal cells. However, the limited source and proliferation potential of corneal stromal cells has driven us to search for alternative cell sources for corneal stroma engineering. Based on the idea that the tissue-specific environment may alter cell fate, we proposed that dermal fibroblasts could switch their phenotype to that of corneal stromal cells in the corneal environment. Thus, dermal fibroblasts were harvested from newborn rabbits, seeded on biodegradable polyglycolic acid (PGA) scaffolds, cultured in vitro for 1 week, and then implanted into adult rabbit corneas. After 8 weeks of implantation, nearly transparent corneal stroma was formed, with a histological structure similar to that of its native counterpart. The existence of cells that had been retrovirally labeled with green fluorescence protein (GFP) demonstrated the survival of implanted cells. In addition, all GFP-positive cells that survived expressed keratocan, a specific marker for corneal stromal cells, and formed fine collagen fibrils with a highly organized pattern similar to that of native stroma. However, neither dermal fibroblast-PGA construct pre-incubated in vitro for 3 weeks nor chondrocyte-PGA construct could form transparent stroma. The results demonstrated that neonatal dermal fibroblasts could switch their phenotype in the new tissue environment under restricted conditions. The functional restoration of corneal transparency using dermal fibroblasts suggests that they could be an alternative cell source for corneal stroma engineering.     

胶原酶二步消化法分离培养牛角膜基质成纤维细胞*

背景：获得纯度高、活性强、生物学特性更接近体内状态的角膜基质成纤维细胞是角膜损伤后修复研究的基础。 目的：探索体外培养牛角膜基质成纤维细胞的新方法。 方法：用Ⅰ型胶原酶对牛角膜基质层进行二步消化分离后制成细胞悬液转入培养瓶中培养,取生长良好的细胞进行传代培养。采用锥虫蓝染色检测消化分离后细胞即刻存活率;倒置相差显微镜动态观察细胞的生长状态;波形蛋白免疫组织化学染色鉴定角膜成纤维细胞;MTT法检测细胞增殖情况;取处于对数生长期的细胞,绘制细胞生长曲线并计算细胞群倍增时间。 结果与结论：在体外成功分离培养牛角膜基质成纤维细胞,显微镜下观察及波形蛋白免疫组织化学染色后证实所培养的细胞为角膜基质成纤维细胞。锥虫蓝染色,细胞即刻成活率达93.5%。细胞生长曲线近似“S”形,其群体倍增时间为        38.70 h。说明二步消化法细胞培养技术简便、经济、高效,为原代培养角膜基质成纤维细胞提供了有效渠道。 关键词：角膜;成纤维细胞;细胞原代培养;细胞形态;生物学特性;消化法 doi:10.3969/j.issn.1673-8225.2012.07.015     

CD34+ corneal stromal cells are bone marrow-derived and express hemopoietic stem cell markers

Previous studies have suggested that corneal stromal keratocytes express the CD34 antigen. We wished to investigate CD34 antigen expression in normal mouse cornea using dual- and triple-staining techniques. Whole-mount preparations of mouse and rat corneas were examined with confocal microscopy using single, dual, or triple immunostaining to study their morphology, phenotype, and distribution. Single-cell suspensions from normal mouse corneas were also prepared and analyzed by flow cytometry. After short-term culture of corneal stromal explants, nonadherent cells were harvested and cytospins were prepared and stained for different markers.Combined staining for F-actin and leukocyte differentiation markers clearly showed that the corneal stroma contains a population of CD45(+) resident bone marrow-derived cells, whereas most cells were CD45-F-actin(+) keratocytes. A significant proportion (two thirds) of CD45(+) cells in the normal corneal stroma expressed CD34(+), whereas no CD45(-) cells (i.e., keratocytes) coexpressed CD34. In addition, CD34(+) cells were CD11c(-) and CD11b(+). Fewer than 10% of the CD34(+) cells also coexpressed Sca-1(+), but no CD34(+) cells coexpressed major histocompatibility complex (MHC) class II(+). In contrast, the remaining population of CD45(+)CD34(-) cells in the corneal stroma expressed CD11b, MHC class II(+) but not CD11c and were found mostly in the anterior and peripheral part of stroma. These cells are in intimate contact with corneal keratocytes, which stained only for F-actin and were negative for all leukocyte markers. Very few CD45(+) cells expressed the B220 marker, suggesting a plasmacytoid dendritic cell phenotype. Flow cytometry analyses confirmed the morphometric data showing that 68% of CD45(+) cells coexpress CD34 and CD11b, whereas 22% are CD11b(+)CD34(-).We conclude that the normal mouse cornea contains two populations of bone marrow-derived leukocytes, both of which are distinct from stromal keratocytes. The larger population resembles CD34(+) hemopoietic stem cells, whereas the smaller population are CD34(-)CD11b(+) MHC class II(+) macrophages. A very small percentage comprises plasmacytoid dendritic cells.     

Bioreactor design for cornea tissue engineering: Material-cell interactions

Several materials were evaluated for potential use in a bioreactor system for a tissue-engineered cornea. Two types of cytotoxicity tests were performed using human corneal stromal fibroblasts: a 24 h cytotoxicity test based on the ASTM standard F813-01 and a 7 days growth inhibition test. It was determined that culture configuration, autoclaving and materials surface preparation were all important factors influencing cell viability. Poly(etheretherketone) and titanium–6Al–4V were found to be the most appropriate materials for use in a corneal bioreactor system. Furthermore, poly(oxymethylene) copolymer and poly(tetrafluoroethlylene) are not safe for use with human corneal fibroblasts after autoclaving.     

Microstructural Characteristics of Extracellular Matrix Produced by Stromal Fibroblasts

The overall objective of this investigation was to characterize the extracellular matrix deposited by the stromal fibroblasts as a function of time in culture and matrix microstructure. Stromal fibroblasts were seeded onto collagen matrices and cultured for up to 5 weeks. The collagen matrices contained collagen fibrils with an average diameter of 215 ± 20 nm. When cultured on a collagen film, an average fibril diameter of 62 ± 39 nm was observed for single layer films with only slight variations with time in culture, and after 1 week of culture between two film layers 67 ± 47 nm fibrils were observed after 1 week. When the film surface was molded into 1 and 2 μm microgrooves, the initial average fibril diameter of the extracellular matrix was 73 ± 21 and 73 ± 31 nm respectively. When cultured on a collagen sponge, an average fibril diameter of 107 ± 20 nm was initially observed and decreased to 47.5 ± 17 nm after 1 week in culture. For cells cultured on a collagen sponge, Western blotting showed an increase in myofibroblast phenotype expression with time in culture. Shifts in phenotype were less distinct for cells cultured on collagen films. The microstructure, rather than geometry, of the matrix substrate appeared to influence the newly synthesized extracellular matrix and cell phenotype.     

Development and testing of a human collagen graft material

Human Type I collagen was extracted from placenta using pepsin and salt fractionation. The collagen was characterized by SDS-PAG electrophoresis dispersed in acidic medium, freeze-dried, and cross-linked in an 0.25% glutaraldehyde solution pH 4.5 for 2 days. After washing for 7 days and freeze drying the resultant collagen sponge was tested with regard to mechanical, physical, enzymatic degradation properties and biological responses. The modulus of elasticity was found to be 289 +/- 10 g/mm2 and the sponge was insoluble in water, buffered saline, or tissue culture medium over a period of 6 weeks with swelling occurring at less than 5% of volume. The sponge had a high fluid binding capacity, amounting to 56 +/- 5 mL tissue culture medium per gram of dry weight. Bacterial collagenase produced slow degradation of the sponge with complete disappearance by 24 h only when high concentrations (200 units enzyme per mg of the collagen sponge) were used. Cytotoxicity studies using human gingival and periodontal ligament fibroblasts revealed less than 5% apparent cytotoxicity or proliferation. Subcutaneous implantation was followed by resorption and vascularization over a period of 6-8 weeks. It was concluded that the collagen sponge prepared from human Type I collagen has potential as a graft material in oral surgical procedures.     

Dexamethasone inhibits interleukin-1beta-induced corneal neovascularization: role of nuclear factor-kappaB-activated stromal cells in inflammatory angiogenesis

Dexamethasone, a synthetic corticosteroid, is widely used as a potent anti-inflammatory drug in various diseases including corneal angiogenesis. However, dexamethasone's impact on interleukin (IL)-1beta-dependent inflammatory angiogenesis is unknown. Here, we show that dexamethasone inhibits IL-1beta-induced neovascularization and the expression of the angiogenesis-related factors, vascular endothelial growth factor-A, KC, and prostaglandin E(2) in the mouse cornea 2 days after IL-1beta implantation. IL-1beta caused IkappaB-alpha phosphorylation in corneal stromal cells but not in infiltrated CD11b(+) cells 2 days after IL-1beta implantation. In contrast, both cell types were positive for phosphorylated IkappaB-alpha 4 days after IL-1beta implantation. Dexamethasone significantly inhibited IkappaB-alpha phosphorylation 2 and 4 days after IL-1beta implantation. Furthermore, dexamethasone inhibited IL-1beta-induced expression of vascular endothelial growth factor-A, KC, and prostaglandin E(2), and signaling of nuclear factor (NF)-kappaB in corneal fibroblasts in vitro. A selective NF-kappaB inhibitor attenuated IL-1beta-induced corneal angiogenesis. These findings suggest that NF-kappaB activation in the corneal stromal cells is an important early event during IL-1beta-induced corneal angiogenesis and that dexamethasone inhibits IL-1beta-induced angiogenesis partially via blocking NF-kappaB signaling.     

Ex vivo large-scale generation of human platelets from cord blood CD34+ cells

In the present investigation, we generated platelets (PLTs) from cord blood (CB) CD34(+) cells using a three-phase culture system. We first cultured 500 CB CD34(+) cells on telomerase gene-transduced human stromal cells (hTERT stroma) in serum-free medium supplemented with stem cell factor (SCF), Flt-3/Flk-2 ligand (FL), and thrombopoietin (TPO) for 14 days. We then transferred the cells to hTERT stroma and cultured for another 14 days with fresh medium containing interleukin-11 (IL-11) in addition to the original cytokine cocktail. Subsequently, we cultured the cells in a liquid culture medium containing SCF, FL, TPO, and IL-11 for another 5 days to recover PLT fractions from the supernatant, which were then gel-filtered to purify the PLTs. The calculated yield of PLTs from 1.0 unit of CB (5 x 10(6) CD34(+) cells) was 1.26 x 10(11) - 1.68 x 10(11) PLTs. These numbers of PLTs are equivalent to 2.5-3.4 units of random donor-derived PLTs or 2/5-6/10 of single-apheresis PLTs. The CB-derived PLTs exhibited features quite similar to those from peripheral blood in morphology, as revealed by electron micrographs, and in function, as revealed by fibrinogen/ADP aggregation, with the appearance of P-selectin and activated glycoprotein IIb-IIIa antigens. Thus, this culture system may be applicable for large-scale generation of PLTs for future clinical use.     

Chick embryo fibroblasts senscence in vitro: pattern of cell division and life span as a function of cell density.

The pattern of cell division, ageing and death of cultured chick embryo fibroblasts inoculated at a wide range of cell densities is described. Cell populations seeded at densities of 2.1 x 10-3 to 3.1 x 10-4 cells per cm-2 double between 39 and 17 times respectively while their life span at all densities remains 57 puls or minus 3 days. Thus, under the experimental conditions employed, the life span of cultured embryonic chick cells is a function of calendar rather than mitotic time. We also find that the duration of phase III and the generation time of the cells during this period are proportional to the density of the culture. Moreover, the division rate of cells inoculated at low densities (4.2-31.0 x 10-2 cells per cm-2) remains constant throughout their entire lifetime. It is suggested that the decline in the proliferative capacity of chick cells ageing in vitro is a function of both their density and their age.     

Fabrication and in vitro testing of polymeric delivery system for condensed DNA

Polyethylenimine (PEI) was combined with plasmid DNA and freeze dried following the addition of sucrose as a lyoprotectant and pore-forming agent. Freeze-dried PEI DNA condensates were dry mixed with granular polylactideglycolic acid (PLGA) then compression molded and sponged to encapsulated PEI DNA. A measurement of the elastic modulus indicated that 91 wt% sucrose substituted for 95 wt% sodium chloride as a porogen, resulting in PLGA sponges with a mechanical modulus of 100 kPa. The PEI DNA was retained (80%) within PLGA sponges prepared with sucrose during the leaching and subsequent 2-week release studies, whereas sodium chloride PLGA sponges caused the premature release (100%) of PEI DNA within 2 days. In vitro gene transfer studies with PEI DNA PLGA sponges established that adherent and infiltrating fibroblasts expressed reporter gene for 15 days compared with the short, 3-day expression mediated by direct gene of PEI DNA on cells in culture. The results demonstrate an approach to encapsulate condensed DNA in a PLGA sponge for the purpose of retaining DNA within the matrices and creating efficient gene transfer during tissue engineering.     

Effect of a dextran derivative associated with TGF-beta1 or FGF-2 on dermal fibroblast behaviour in dermal equivalents

Dextran derivatives that mimic the action of heparin have been shown to protect heparin-binding growth factors, such as Transforming Growth Factor-beta1 (TGF-beta1) and Fibroblast Growth Factor-2 (FGF-2). The aim of this study was to investigate the effect of LS21 DMCBSu, a dextran derivative which contains methylcarboxylate, benzylamide and sulfate groups, both by itself and when combined with TGF-beta1 and FGF-2, on the behaviour of fibroblasts. Two systems were assessed: a monolayer culture and three-dimensional collagenous matrices (dermal equivalent). Polymeric biomaterial LS21 DMCBSu and LS21 DMCBSu associated with either TGF-beta1 or FGF-2, were added to the monolayer culture on day 3. After 7 days of culture the number of cells was determined. Two treatments were carried out on the dermal equivalents: 9 days of treatment from day 0 to day 9 of culture and 9 days of treatment from day 21 to day 30 of culture for the premature and the mature dermal equivalents respectively. In the monolayer culture, the bioactive polymer produced a slight increase in fibroblast growth (10% with 10 microg/ml of LS21 DMCBSu) and promoted the stimulating effect of the growth factors on cell growth. In the premature dermal equivalents growth was stimulated by 20% when 10 microg/ml LS21 DMCBSu was added. The dextran derivative mixed with TGF-beta1 slightly inhibited the growth effect of the growth factor in the dermal equivalents. The functionalized dextran with FGF-2 enhanced the stimulating effect of the growth factor in the premature dermal equivalent. A significant increase in cell growth was observed with the fibroblasts treated with the FGF-2 LS21 DMCBSu mixture and FGF-2 (51% and 40%, respectively). However, none of the described treatments affected the cell growth in the mature dermal equivalent. Furthermore, the dextran derivative had no effect on dermal contraction under these experimental conditions (3D culture).     

Smooth muscle differentiation in cultured human breast gland stromal cells

We analyzed in cultures from the human breast the potential of stromal cells resembling fibroblasts to undergo smooth muscle differentiation. The cellular components of the breast tissue from 10 biopsies were disaggregated by collagenase digestion and further purified by differential centrifugation into suspensions of single cells and intact blood vessels. These two fractions of stromal cells were plated in culture and their phenotypic traits analyzed within 24 hours. During this time, the blood vessel fraction gave rise to stromal cells with smooth muscle differentiation as judged immunocytochemically using monoclonal antibodies to alpha-/gamma-muscle actins, to alpha-smooth muscle actin, to type IV collagen, and to laminin. Furthermore, the cells of this fraction resembled smooth muscle cells based on 2D gel electrophoresis and immunoblotting determination of isoactin content. After 24 hours in culture, the single-cell fraction consisted of an almost pure population of cells not exhibiting smooth muscle differentiation but rather resembling fibroblasts. Maintenance of fibroblast-like cells without smooth muscle differentiation was possible for more than 14 days on chemically defined medium. These cells remained quiescent, as measured by cell quantification and immunoreactivity to the proliferation-associated antigen, Ki-67. Growth of these cells could be stimulated by adding serum at any time during the experimental period. Single-cell fractions from seven biopsies were allowed to grow exponentially in the presence of serum for up to 10 days, and the kinetics of smooth muscle differentiation were monitored immunocytochemically and biochemically. These experiments showed that alpha-smooth muscle actin synthesis was induced in 10 to 80% of the fibroblast-like cells after 4 to 11 days in culture. Both the final number of alpha-smooth muscle actin-positive cells and the onset of synthesis varied with the initial seeding density. Dose-response experiments (at constant cell density) revealed that serum exerted maximal effect at concentrations above 10%. It was therefore concluded that elements of smooth muscle differentiation may arise in non-smooth muscle stromal cells taken directly from human breast tissue.     

The need for modification of the polyvinyl sponge model of connective tissue growth. Histoligic and biochemical studies in the rabbit.

Quantitative histopathologic and biochemical comparisons were made between polyvinyl sponge capsular and sponge tissue in the rabbit on different days after subcutaneous implantation. Up to 9 days the predominant cell type in the capsular tissue is the fibroblast and in the sponge it is the neutrophil. During this time period the sponge tissue shows lower rates of (14C) proline and (14C) cytidine incorporation and lower rates of total (14C) collagen synthesis than the surrounding capsule. Different gel electrophoretic patterns of isolated radioactive proteins are found in sponge and capsule at 6 days. These biochemical differences appear to be related to the small number of fibroblasts, relative to granulocytes present in sponges during the first 9 days after implantation. It is suggested that future biochemical investigations of the early phase of connective tissue reactions (first 9 days) in this model utilize sponge capsular tissue within 1 cm of the sponge edge instead of the sponge and its contents.     

Mechanical properties evolution of a PLGA-PLCL composite scaffold for ligament tissue engineering under static and cyclic traction-torsion in vitro culture conditions

This study aims to investigate the in vitro degradation of a poly(L-lactic-co-glycolic acid)-poly(L-lactic-co-?-caprolactone) (PLGA-PLCL) composite scaffold’s mechanical properties under static culture condition and 2?h period per day of traction-torsion cyclic culture conditions of simultaneous 10% uniaxial strain and 90° of torsion cycles at 0.33?Hz. Scaffolds were cultured in static conditions, during 28?days, with or without cell seeded or under dynamic conditions during 14?days in a bioreactor. Scaffolds’ biocompatibility and proliferation were investigated with Alamar Blue tests and cell nuclei staining. Scaffolds’ mechanical properties were tested during degradation by uniaxial traction test. The PLGA-PLCL composite scaffold showed a good cytocompatibility and a high degree of colonization in static conditions. Mechanical tests showed a competition between two process of degradation which have been associated to hydrolytic and enzymatic degradation for the reinforce yarn in poly(L-lactic-co-glycolic acid) (PLGA). The enzymatic degradation led to a decrease effect on mechanical properties of cell-seeded scaffolds during the 21st days, but the hydrolytic degradation was preponderant at day 28. In conclusion, the structure of this scaffold is adapted to culture in terms of biocompatibility and cell orientation (microfiber) but must be improved by delaying the degradation of it reinforce structure in PLGA.     

Effect of substrate composition and alignment on corneal cell phenotype

Corneal blindness is a significant problem treated primarily by corneal transplants. Donor tissue supply is low, creating a growing need for an alternative. A tissue-engineered cornea made from patient-derived cells and biopolymer scaffold materials would be widely accessible to all patients and would alleviate the need for donor sources. Previous work in this lab led to a method for electrospinning type I collagen scaffolds for culturing corneal fibroblasts ex vivo that mimics the microenvironment in the native cornea. This electrospun scaffold is composed of small-diameter, aligned collagen fibers. In this study, we investigate the effect of scaffold nanostructure and composition on the phenotype of corneal stromal cells. Rabbit-derived corneal fibroblasts were cultured on aligned and unaligned collagen type I fibers ranging from 50 to 300?nm in diameter and assessed for expression of α-smooth muscle actin, a protein marker upregulated in hazy corneas. In addition, the optical properties of the cell-matrix constructs were assessed using optical coherence microscopy. Cells grown on collagen scaffolds had reduced myofibroblast phenotype expression compared to cells grown on tissue culture plates. Cells grown on aligned collagen type I fibers downregulated α-smooth muscle actin protein expression significantly more than unaligned collagen scaffolds, and also exhibited reduced overall light scattering by the tissue construct. These results suggest that aligned collagen type I fibrous scaffolds are viable platforms for engineering corneal replacement tissue.     

Establishment of a rat long-term culture expressing the osteogenic phenotype: dependence on dexamethasone and FGF-2

Rat stromal bone-marrow cells cultured in the presence of dexamethasone, ascorbic acid, beta-glycerophosphate, and fibroblast growth factor-2 (FGF-2) express the osteogenic phenotype (Pitaru et al., J. Bone Miner. Res. 8:919-929, 1993). The purpose of this study was to establish a long-term homogeneous culture expressing the osteogenic phenotype. The cultures were routinely passaged every 5 days in the absence or presence of either or both dexamethasone and FGF-2, and the cumulative doubling number and the expression of the osteogenic phenotype were determined. Cultures treated with dexamethasone (10(-7) M) ceased proliferation and only upon addition of FGF-2 (3 ng/ml) was a spontaneous immortalization achieved, as expressed by sustained proliferation for about 1 year, with a doubling time of 22 h and more than 300 doublings in 72 passages. Both FGF-2 and dexamethasone are required and act synergistically to maintain cell propagation, alkaline phosphatase expression, and osteocalcin secretion; however, protein content was FGF-2 dependent and the mineralization was dexamethasone dependent. Repetitive single-cell cloning tested the homogeneity and stability of the cells expressing the osteogenic phenotype in these long-term cultures. It was shown that 25% to 50% of subclones derived from clones with an osteogenic phenotype do not further express the osteogenic phenotype. In conclusion, we have established a spontaneously immortalized dexamethasone- and FGF-2-dependent rat stromal bone-marrow-derived long-term culture expressing the osteogenic phenotype. The cultures tend to lose the osteogenic phenotype, and dexamethasone supports the long-term preservation of the osteogenic phenotype.     

干燥脱水保存鸵鸟脱细胞角膜基质载体材料的细胞毒性

背景：陕西省眼科研究所通过前期研究发现鸵鸟角膜具有开发成为人角膜材料替代品的优势。 目的：判断鸵鸟角膜基质支架材料对细胞的潜在毒性作用。 方法：采用细胞培养方法,将干燥脱水法保存的鸵鸟脱细胞角膜基质载体支架制备成浸提液与L-929细胞共同培养,采用MTT比色法评价其作用1,2,3 d,对细胞生长和增殖的影响。 结果与结论：干燥脱水法保存的鸵鸟脱细胞角膜基质载体支架浸提液组培养1,3,5 d,对实验细胞毒性反应级别均为1级。参照《中华人民共和国国家标准GB/T16886.5-2003》进行细胞毒性试验,供试品组见少量细胞呈圆形,疏松贴壁,无胞浆内颗粒,偶见细胞溶解。结合MTT法分析结果表明,采用干燥脱水法保存鸵鸟脱细胞角膜基质载体材料,细胞毒性为1级,为合格材料。