December 1999
Volume 40, Issue 13
Free
Cornea  |   December 1999
Failure to Activate Transcription Factor NF-κB in Corneal Stromal Cells (Keratocytes)
Author Affiliations
  • Jeffery R. Cook
    From the Vision Research Laboratories of New England Medical Center and the Departments of Ophthalmology, and Anatomy and Cellular Biology, Tufts University School of Medicine, Boston, Massachusetts.
  • Mehernosh K. Mody
    From the Vision Research Laboratories of New England Medical Center and the Departments of Ophthalmology, and Anatomy and Cellular Biology, Tufts University School of Medicine, Boston, Massachusetts.
  • M. Elizabeth Fini
    From the Vision Research Laboratories of New England Medical Center and the Departments of Ophthalmology, and Anatomy and Cellular Biology, Tufts University School of Medicine, Boston, Massachusetts.
Investigative Ophthalmology & Visual Science December 1999, Vol.40, 3122-3131. doi:https://doi.org/
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jeffery R. Cook, Mehernosh K. Mody, M. Elizabeth Fini; Failure to Activate Transcription Factor NF-κB in Corneal Stromal Cells (Keratocytes). Invest. Ophthalmol. Vis. Sci. 1999;40(13):3122-3131. doi: https://doi.org/.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. Freshly isolated cultures of corneal stromal cells (keratocytes) are incompetent to synthesize the tissue remodeling proteinase, collagenase, in response to agents such as cytochalasin B (CB) or phorbol myristate acetate (PMA), which are strong stimulators of collagenase expression in subcultured fibroblasts of all types, including those from corneal stroma. Incompetence is due to failure to activate an autocrine interleukin (IL)-1α feedback loop required to mediate cell response. The goal of the present study was to investigate the mechanism for this failure.

methods. A cell culture model of freshly isolated corneal stromal cells and subcultured stromal fibroblasts from rabbits was used for these studies.

results. Competence to synthesize collagenase in response to CB was acquired as a differentiation property by corneal stromal cells placed in culture, and did not require subculture. Competence acquisition correlated with transition to a fibroblastic spindle shape, assembly of actin stress fibers, and the acquired capacity to collapse in response to CB. It was demonstrated that competence could be more precisely defined as the capacity to express IL-1α in response to IL-1, making possible activation of the feedback loop. Investigation into the signaling pathway for IL-1α expression in response to IL-1 revealed a requirement for reactive oxygen species and activity of the transcription factor nuclear factor (NF)-κB. Importantly, freshly isolated stromal cells were found to be relatively incompetent to activate NF-κB in comparison to subcultured stromal fibroblasts.

conclusions. Failure to activate NF-κB explains incompetence for expression of IL-1α in corneal stromal cells. Because NF-κB regulates many cell functions with potential to disturb corneal structure, including expression of inflammatory, stress, and degradative proteinase genes; protection against apoptosis; and cell replication; this seems likely to be an important mechanism protecting corneal stasis and preserving function.

The cornea focuses light on the retina, providing more than two thirds of the refractive power of the eye. 1 To perform this function, the cornea must remain perfectly transparent, and it must maintain a precise shape. The extracellular matrix (ECM) 1 of the corneal stroma undergoes an unusually slow rate of turnover and remodeling, and the stromal cells, or“ keratocytes” replicate at a very slow rate. 2 Therefore, corneal structure remains static unless injury occurs. Even then, corneal stasis may remain undisturbed, as corneal stromal cells are resistant to stimuli that would induce cells from other tissues to enter the cell cycle and initiate a fibrotic response. For example, simple surgeries such as radial keratotomy stimulate little or no stromal cell replication or new ECM deposition in the cornea of the human adult. 3 4 In addition, when the overlying epithelium is damaged or infected, corneal stromal cells rapidly undergo apoptosis, effectively precluding any fibrotic response. 5 These features of cornea may have evolved to limit undesirable changes in tissue structure because of normal renewal or repair processes, much as corneal immune privilege protects it against damaging inflammatory reactions. 6  
When injury to the stroma is extensive enough to initiate a fibrotic response, keratocytes located at the wound edge lose their quiescence and undergo transformation to the repair fibroblast phenotype. 7 8 These cells migrate into the wound area, proliferate, and deposit an opaque ECM characteristic of repair tissue. Transformation can be visualized on the molecular level as a reorganization of the actin cytoskeleton, with development of stress fibers and focal adhesion structures. 9 10 In addition, a battery of new genes is activated, including those encoding extracellular matrix components such as fibronectin, 9 the cell:ECM adhesion molecule, α5 integrin, 11 ECM-degrading matrix metalloproteinases (MMPs), 12 13 and inflammatory cytokines. 14 This same transition occurs when keratocytes are isolated from the corneal stroma and cultured in serum-containing medium; by the time these cells are subcultured, they have acquired the repair fibroblast phenotype according to the criteria cited above. 10 11 14 15 16 17 18 19 20 21 22  
We have held the view that a comparison of freshly isolated and subcultured corneal stromal cells might reveal mechanisms contributing to the maintenance of stromal stasis. In previous work, we demonstrated that freshly isolated cells differ from subcultures, or from wound fibroblasts, in their incompetence to synthesize collagenase in response to treatment with agents that stimulate remodeling of the actin cytoskeleton, such as phorbol myristate acetate (PMA) or cytochalasin B (CB). 14 19 20 We further reported that this incompetence was due to failure to activate an autocrine interleukin-1α (IL-1α) feedback loop, which is necessary to mediate cell response. IL-1α is a multifunctional cytokine that regulates expression of genes involved in inflammation and tissue remodeling and that also stimulates fibroblast replication. 23 Therefore, incompetence of freshly isolated stromal cells to express IL-1α might contribute to the maintenance of corneal stasis. For this reason, it seemed important to investigate the molecular mechanisms that make IL-1α expression possible in subcultured stromal cells or wound fibroblasts and to determine the nature of the block to IL-1α expression in stromal cells from the normal, uninjured tissue. Here we report our findings with regard to the role of transcription factor nuclear factor(NF)-κB in stromal cell competence to express IL-1α. Our identification of a deficiency in this regulatory factor in normal corneal stromal cells freshly isolated from the tissue suggests a novel mechanism for protection of corneal structure. 
Materials and Methods
Corneal Stromal Cell Culture and Treatment Reagents
Use of animals was in accordance with the ARVO Statement for Use of Animals in Ophthalmic and Vision Research. Stromal cells were isolated from the corneas of New Zealand White rabbits according to the method of Johnson–Muller and Gross. 24 Briefly, the epithelial and endothelial layers are separated from the corneal stromas by treatment with trypsin overnight at 4°C, then cells are freed from the stromal ECM by incubation in a solution of bacterial collagenase for 2 to 4 hours at 37°C. Cells were plated in Dulbecco’s modified Eagle’s medium containing 10% supplemented calf serum (HyClone, Logan, UT). These “freshly isolated” cells were used for an experiment within 24 hours after plating. Alternatively, cells were allowed to replicate in culture and then were removed from the plate by trypsinization and subcultured, as we have described previously 16 ; these “subcultured” cells were used on or before the fourth passage. For an experiment comparing freshly isolated and subcultured cells, each cell type was plated according to the same protocol, and treatments were performed at the same time. Plating protocols differed according to the needs of each experiment and are detailed in each of the subsections below. The following cell treatment reagents were used in this study: CB (Sigma, St. Louis, MO) at 5 μg/ml, PMA (Sigma) at 10−6 M, human recombinant IL-1α (R&D Systems, Minneapolis, MN) at 10 ng/ml, cycloheximide (CHX) (Sigma) at 5 μg/ml, pyrrolidine dithiocarbamate (PDTC) (Sigma) at 10 to 100 μM, rotenone (Sigma) at 10 and 40 μM, and H2O2 (Sigma) at 1 and 10 mM. 
Secreted Protein Analysis
For an experiment involving secreted protein analysis, freshly isolated or subcultured stromal cells were plated at subconfluent density in 24-well culture plates at 2.0 × 105 cells per well. The next day, the medium was changed to serum-free medium because the large quantities of albumin in serum interfere with subsequent electrophoretic resolution of proteins in the culture medium. Treatment reagents were then added to duplicate or triplicate wells to ensure reproducibility.[ 35S]-Methionine (New England Nuclear, Medford, MA) also was added at this time to the medium at 80 μCi/ml for biosynthetic-labeling. After 24 hours, medium containing radiolabeled, secreted cell proteins was collected. Samples containing equal TCA-precipitable CPMs were diluted in gel-loading buffer containing 2-mercaptoethanol and electrophoresed on 8% sodium dodecylsulfate (SDS)-polyacrylamide gels. Gels were autoradiographed to display radiolabeled proteins. The identity of specific protein bands as collagenase or stromelysin was determined by immunoprecipitation analysis, using sheep polyclonal antisera raised against the purified rabbit enzymes, as we have described in detail previously. 25  
RNA Analysis
To prepare freshly isolated cultures for an experiment involving RNA analysis, the stromal cells from eight corneas (approximately 7 × 106 cells) were plated in a“ 100-mm” culture dish (the actual surface diameter is 85 mm); this resulted in slightly subconfluent cultures. These cells were then used for an experiment within 24 hours. To prepare subcultures for an experiment, passaged cells were treated with trypsin, split 1:4 into 100-mm dishes, and allowed to multiply until they reached approximately 90% confluence. Before beginning an experiment, cell morphology was examined and photographed by phase contrast microscopy using an inverted microscope (Telaval 31; Zeiss, Thornwood, NY). To begin an experiment, the culture medium was changed, and then appropriate treatment reagents were added. Total RNA was isolated and analyzed by northern blot analysis as described. 19 20 Rabbit cDNA probes for collagenase, 26 the inflammatory cytokines IL-1α 27 and extractable nuclear antigen (ENA)-78, 14 and the acute phase protein, serum amyloid A3 (SAA3) 22 were labeled with 32P by random priming. 28 Loading equivalence between gel lanes was ascertained by probing for glyceraldehyde-3-phosphate dehydrogenase (GAPD) message with a human cDNA. 29  
Assay for Stromal Cell Competence
Stromal cells freshly isolated from the cornea were plated in eight-chamber slides (Tissue-Tek; VWR Scientific, Boston, MA) at 104 cells/chamber. The assembly of actin filaments was followed in one set of cultures over a time course. To do this, cells in duplicate wells were fixed in 10% sodium phosphate-buffered formalin (pH 7.0) at 24-hour intervals, and actin filaments were stained with rhodamine isothiocyanate (RITC)-conjugated phalloidin (cat. no. R-415; Molecular Probes Inc., Eugene, OR), according to the manufacturers directions. Stained cells were viewed and photographed using a Zeiss Axiophot (Atlantex and Zeiler Instrument Corp., Avon, MA) equipped for epi-illumination. A second set of cultures was used for a time course analysis of competence for collagenase expression in response to CB. For this, duplicate culture wells were treated with CB for 24 hours, at 24-hour intervals. Monensin was added to cultures during the last 4 hours of CB treatment (1 μM final concentration), to block protein secretion. 30 Cells were then fixed, and indirect immunofluorescent antigen localization was performed using standard methods. 31 The primary antibody was the sheep polyclonal collagenase antiserum described in Secreted Protein Analysis, used at a 1:50 final dilution. The secondary antibody was fluorescein isothiocyanate–conjugated donkey anti-sheep IgG (used at a 1:50 final dilution; Jackson ImmunoResearch Laboratories,West Grove, PA). A set of cultures stained with secondary antibody only served as the control for antibody specificity. Six microscopic fields were photographed at random in each of the duplicate wells for each time point. The total cell number in each photograph and the number stained by specific antibody were counted. Statistical significance of differences between the number of collagenase-positive cells identified on each day was determined using the Student’s t-test. 
Transcription Factor Analyses
For electrophoretic mobility shift assay (EMSA), cells were plated and treated as for experiments using RNA analysis. The procedure for preparation of cleared nuclear lysates was essentially that of Schreiber et al. 32 Protease inhibitors were added to buffers just before use at the following concentration: 1 mM dithiothreitol, 1 mM polymethylsulfonyl fluoride, 1 mM benzamidine, 1 mg/ml aprotinin, 5 mM NaF, 10 mg/ml antipain, and 10 mg/ml leupeptin. Protein concentrations were determined by the Bio-Rad assay (Bio-Rad Laboratories, Hercules, CA) using bovine serum albumin as a standard. The lysates were stored at –80°C in 2-ml aliquots. To begin an experiment, lysates were thawed and protein equivalents of each cell lysate (5 to 10 μg) were aliquoted to individual tubes and incubated with 0.035 pmol of the appropriate radiolabeled, double-stranded oligonucleotide probe at room temperature for a total of 30 minutes. The following probes were used (transcription factor binding sites underlined): NF-κB (5′-AGT TGA GGG GAC TTT CCC AGG C-3′) (Promega Corp., Madison WI), AP-1 (5′-CGC TTG ATG AGT CAG CCG GAA-3′) (Promega), an oligonucleotide (5′-GCA CTT GTA GCC ACG TAG CCA CGC CTA CTT-3′), corresponding to the E-box-like binding site in the human IL-1α gene 33 (Tufts University oligonucleotide facility). Oligonucleotides were 5′ end-labeled with[γ -32P]ATP. For oligonucleotide competition reactions, a 50-fold molar excess of unlabeled “cold” probe was added to the reaction before addition of the radiolabeled probe. For supershift analysis, affinity-purified rabbit polyclonal antibodies raised against peptides representing conserved, specific regions of NF-κB transcription factor subunits: NF-κB1/p50 (cat. no. sc-109), RelA/p65 (cat. no. sc-114), and c-Rel (cat. no. sc-272), or the unrelated transcription factor AP-2α (cat. no. sc-184) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibodies (usually 1 μl as recommended by the manufacturer) were added after incubation with the probe and allowed to incubate for an additional 45 minutes. At the end of the incubation times, reactions were loaded onto 4% nondenaturing polyacrylamide gels and electrophoresed with 0.5× Tris-borate-EDTA running buffer at room temperature for 3 hours. Gels were vacuum dried and exposed to x-ray film (Kodak X-OMAT AR; Eastman Kodak, Rochester, NY) overnight at –20°C. 
To inhibit NF-κB activity, subcultured stromal cells, plated as for EMSA analysis, were treated with a cell-permeable synthetic peptide, SN50 (Biomol, Plymouth Meeting, PA) at 50 μg/ml. The peptide contains the nuclear localization sequence of the NF-κB p50 subunit linked to the hydrophobic region (h-region) of the signal peptide of Kaposi fibroblast growth factor. The signal peptide confers cell permeability because of its hydrophobicity, whereas the nuclear localization sequence (amino acids 360–369) appears to interfere with a nuclear transport system used by NF-κB and other transcription factors. 34 An inactive analogue of the SN50 peptide, SN50M, was used as a control for nonspecific effects. 
Western Blot Analysis
To prepare cell lysates for western blot analysis, stromal cell cultures were washed twice with warm phosphate-buffered saline, lysed by scraping in hot 2× SDS-polyacrylamide gel electrophoresis (PAGE) sample buffer (containing 4% sodium dodecyl sulfate and 10%β -mercaptoethanol), and immediately boiled for 10 minutes Protein concentrations were determined by the Bio-Rad assay (Bio-Rad Laboratories, Hercules, CA) using BSA as a standard. The contents of protein equivalent lysate samples (usually 20 μg) were separated by SDS-PAGE under reducing conditions and transferred to nitrocellulose or Hybond membranes (Amersham Life Sciences, Arlington Heights, IL). Blots were probed with affinity-purified antibodies. Antibodies for NF-κB subunits p65 (cat. no. SA-171) and p50 (cat. no. SA-170) were purchased from Biomol. 
Results
Freshly Isolated Stromal Cells Acquire Competence to Synthesize Collagenase in Response to CB by Differentiation in Culture
Viewed by phase contrast microscopy the day after plating (Fig. 1A ), freshly isolated corneal stromal cells exhibit a round or crescent shape, and they are relatively small with a transparent cytoplasm. However, when these cells are examined again after subculture, they have a typical fibroblast spindle shape, and their cytoplasm has become highly refractile (Fig. 1A) . These changes are reflective of the reorganization of the actin cytoskeleton. 18 As we have previously shown, 20 the subcultured phenotype is also characterized by competence to synthesize collagenase and the related MMP, stromelysin, in response to agents that stimulate remodeling of the actin cytoskeleton, such as CB or PMA (Fig. 1B)
The differences between freshly isolated stromal cells and their subcultures with respect to competence for collagenase synthesis might be due to their transformation in culture, similar to the transformation they undergo in the corneal wound. However, because primary stromal cells undergo many replications before subculture, it was also possible that the change was due to selection for a subcomponent of the cell population with a growth advantage under the culture conditions. To distinguish between these possibilities, freshly isolated cells were plated in replicate culture wells and their phenotype was monitored daily to determine whether transformation could occur without subculturing. Results of a representative experiment are depicted in Figure 2 and summarized in Table 1 . Rhodamine isothiocyanate (RITC)-phalloidin staining of cells in culture for 24 hours (Fig. 2A) revealed a poorly developed actin cytoskeleton with a wispy, weblike structure. However, stress fibers had begun to develop in cells viewed 2 days after plating (Fig. 2B) and were fully developed by day 3 (Fig. 2C) . The cells had also become spindle-shaped by day 3, appearing identical with subcultured cells (cf. Fig. 2C and 1A ). No further changes in the actin cytoskeleton were observed on day 4 (data not shown). Progress of these cells toward competence for collagenase expression in response to CB was scored by using immunofluorescent localization to identify individual cells synthesizing collagenase and then by quantifying the percentage of these cells in the total population (Table 1) . Before day three, none of the cells scored competent in this assay; however, on day 3, 30% of the cell population exhibited competence. Because intracellular collagenase concentrates within secretory cell compartments, collagenase-positive cells could be identified easily by their brightly fluorescent perinuclear spots (Fig. 2F) . Besides the change in cell shape, two additional changes were correlated with the burst of competence acquisition. The first was in cell replication: the number of cells in a culture well doubled daily through day 3, and then replication ceased as cells became confluent. The second involved a change in response to CB: day 1 and 2 cultures did not collapse in response to CB (Figs. 2D , 2E) , even though actin filaments were disrupted (cf. Fig. 2B and 2E ). In contrast, treatment of day 3 cultures caused the collapse in cell-shape characteristic of fibroblast response to CB (Fig. 2F) . Nevertheless, the percentage of collagenase-positive cells was still increasing on day 4 (Table 1) , indicating that these changes, even if necessary, are not sufficient. Of more importance to our purposes was the rapidity of competence acquisition in relation to the rate of cell replication. This reveals that competence is not acquired by selection for a specific cell type. In fact, competence acquisition does not require that cells even be subcultured, but instead appears to be an individual response by each cell to being placed in serum-containing culture. 
Freshly Isolated Cells Are Incompetent to Synthesize IL-1α in Response to IL-1
As discussed in the introduction, collagenase gene response to CB or PMA is not direct, but must be mediated through an IL-1α positive feedback loop; freshly isolated cells cannot activate the feedback loop in response to CB or PMA, unlike subcultured cells. 19 20 To activate this positive feedback loop, cells must be able to respond to two different stimulators: CB or PMA acts as the initiating stimulus, but cells must subsequently be able to respond to the IL-1α they synthesized if feedback amplification is to occur. Northern blot analysis was performed to learn whether IL-1 could induce expression of IL-1α in freshly isolated cells (Fig. 3) . Expression of IL-1α was stimulated to easily detectable levels in subcultured cells after 2 hours of IL-1 treatment, and this expression was maintained at the 24-hour time point. In contrast, upregulation occurred only minimally in freshly isolated cells at the 2-hour time point, and no expression was observed at the 24-hour time point. The capacity for IL-1 induction of another inflammatory cytokine, ENA-78, also was impaired in freshly isolated cells. In contrast, expression of a third cytokine, SAA3, was stimulated in response to IL-1 in both cell types. Collagenase expression could not be seen in either cell type after 2 hours of IL-1 treatment, but this is because the induction kinetics for collagenase are slow 17 ; the 24-hour time point revealed that, like SAA3, expression of collagenase could be induced by IL-1 in freshly isolated cells as well as in subcultures. These results suggest that IL-1 signal transduction in subcultured cells must bifurcate along two separate pathways at some point before activation of the genes examined and that the pathway for activation of IL-1α and ENA-78 is nonfunctional in freshly isolated cells. Competence, therefore, can be further defined as the capacity to activate a set of genes in response to IL-1, which includes the inflammatory cytokines, IL-1α and ENA-78. 
Reactive Oxygen Species and Transcription Factor NF-κB: Role in Regulation of the IL-1α Gene by IL-1
It is well known that IL-1 can activate signal transduction pathways that ultimately stimulate DNA-binding activity of transcription factors AP-1 and NF-κB. 35 36 The requirement of an AP-1 binding site in the collagenase promoter for response to IL-1 has been well documented; however, there are no known binding sites in the collagenase promoter for NF-κB. 37 In contrast, the sequence of the 5′ flanking DNA of the human IL-1α gene 33 contains potential binding sites for NF-κB and AP-1. The contribution of these binding sites to IL-1α promoter activity has not been determined, to date. Nevertheless, the differences between the IL-1α and collagenase promoters suggested the hypothesis that NF-κB defines a pathway for regulation of the IL-1α gene by IL-1 that is separate from the pathway regulating the collagenase gene. This hypothesis was investigated in a series of experiments with subcultured stromal cells and is summarized in Figure 4
Figure 4A shows an experiment to determine whether IL-1 induces NF-κB DNA-binding activity. EMSA revealed NF-κB DNA-binding activity in nuclear lysates within 30 minutes of treatment with IL-1. Characterization of the DNA-binding complexes is shown in Figure 4B . The complexes that formed with nuclear lysates prepared from IL-1–treated stromal cells (−) were similar to those formed using a HeLa cell lysate control, but no retardation of the probe occurred in the absence of nuclear lysate (neg). The inducible protein–DNA complex was competed slightly by a 10-fold (1:10) molar excess of unlabeled NF-κB probe and almost entirely by a 50-fold excess (1:50); however, a 50-fold excess of nonspecific probe (NSB) did not affect the complex formation. The constitutive complex appears to be nonspecific because there was no competition from specific or nonspecific cold probe. Antibodies to NF-κB proteins p50 (NF-κB1) or p65 (RelA) each supershifted a distinct subcomponent of the inducible complex. In contrast, an antibody to the NF-κB protein c-Rel did not affect the EMSA profile, nor did an antibody to the unrelated transcription factor AP-2. NF-κB binds to DNA as a homo- or heterodimer, however, the epitope for antibody binding can be masked in one subunit of a dimer. Thus, it seems likely that the complex of faster mobility represents p50:p50 dimers and the complex of slower mobility represents p50:p65 heterodimers; these are the complexes between these two subunits typically found. 38 We conclude that IL-1 treatment of subcultured stromal cells induces DNA-binding complexes containing at least two NF-κB proteins, p50 and p65. 
It is reported that NF-κB is prevented from binding DNA in unstimulated cells by sequestration in the cytoplasm bound to an inhibitory component of the IκB family; activation involves release of NF-κB from IκB; this exposes a nuclear localization signal, enabling it to translocate to the nucleus and bind to its recognition sequence in the promoter region of specific genes. 39 40 Reactive oxygen species (ROS) have been identified as common second messengers, initiating release of NF-κB from I-κB. 41 Immunofluorescent localization experiments in subcultured stromal cells demonstrated that induction of DNA-binding by IL-1 treatment was correlated with translocation of p65 from the cytoplasm to the nucleus of the cell (data not shown). NF-κB activation by IL-1 in corneal stromal cell subcultures could be blocked by addition of the free radical scavenger PDTC (Fig. 4C , left panel). This also blocked induction of IL-1α mRNA in response to IL-1 (Fig. 4C , right panel), but did not block induction of collagenase expression, as assayed by measuring synthesized protein levels (data not shown). Addition of a cell-permeable peptide (SN50) with the capacity to interfere with NF-κB activation inhibited IL-1α mRNA induction in response to IL-1, whereas an inactive peptide of similar structure (SN50M) had no activity (Fig. 4D) . Together, these results indicate a requirement of ROS and NF-κB for expression of IL-1α, but not collagenase, in response to IL-1. These findings define a distinct signaling pathway for IL-1α expression in subcultured stromal cells, which is different from the pathway regulating collagenase expression. 
Failure to Activate DNA Binding of Transcription Factor NF-κB: Role in Determining the Incompetent Phenotype
We next considered the hypothesis that a selective deficiency in NF-κB activity might contribute to the incompetent phenotype of freshly isolated stromal cells. To test this idea, EMSA was performed using equivalent protein aliquots of nuclear lysates prepared from freshly isolated or subcultured stromal cells. A representative experiment is shown in Figure 5 A. Low constitutive NF-κB DNA-binding activity was detectable in subcultured fibroblasts, but was negligible in freshly isolated cell lysates. Strikingly, PMA and IL-1 vigorously stimulated NF-κB DNA-binding activity in subcultured cells; however, PMA did not induce NF-κB DNA-binding activity and IL-1 was only slightly stimulatory in freshly isolated cells. In contrast, AP-1 DNA-binding activity was clearly constitutive in nuclear lysates made from both freshly isolated cells and subcultured stromal cells. PMA and IL-1 were stimulatory to variable degrees, depending on the starting constitutive level; however, both cell types attained an equivalent maximal stimulation. Binding to a third oligonucleotide derived from the IL-1α promoter, containing a palindromic sequence similar to an E-box, 37 did not change with IL-1 treatment, and binding was equivalent in lysates from freshly isolated or subcultured cells. Therefore, NF-κB DNA-binding activity was selectively deficient in freshly isolated stromal cells, as hypothesized. EMSA supershift analysis revealed that the composition of the NF-κB DNA-binding complexes in freshly isolated cells (Fig. 5B) was similar to subcultured cells (see previous Fig. 4B ); that is, the EMSA band was composed essentially of two subcomplexes: one containing p50 and one containing p65. There was simply less of these proteins in the binding complexes that formed using nuclear lysates from freshly isolated cells compared with subcultured cells. However, western blot analysis indicated equivalent amounts of p50 and p65 NF-κB protein in total cell lysates prepared from each cell type (Fig. 5C) . Therefore, the selective deficiency of NF-κB DNA-binding activity in freshly isolated cells is not due simply to a reduction in the amount of p50 and p65 and indicates that the deficiency lies in their activity. 
Discussion
This study investigates mechanisms underlying a phenotypic transition that corneal stromal cells undergo in culture which models the events after corneal injury in vivo. In this system, freshly isolated stromal cells are incompetent to synthesize the ECM remodeling enzyme, collagenase, in response to agents, such as CB or PMA, that easily stimulate collagenase expression in subcultured cells. In this study we characterize the transition to competence and demonstrate that it is a true cellular differentiation event that occurs as cells spend time in serum-containing culture. Using this model, we further investigate the molecular mechanisms determining competence and identify a role for the transcription factor NF-κB. 
Mechanism for Response to CB
In a previous publication, 20 we showed that incompetence to synthesize collagenase in response to CB or PMA is due to failure to activate an IL-1α autocrine loop necessary to mediate this response. To activate this positive feedback loop, cells must be able to respond to two different stimulators: CB or PMA acts as the initiating stimulus, but cells must subsequently be able to respond to the IL-1α they synthesized if feedback amplification is to occur. We show here that IL-1α expression is refractory to stimulation by IL-1 in freshly isolated cells, which would preclude feedback amplification of IL-1α levels. The mechanism whereby CB initiates activation of the autocrine loop was not addressed in our study, and it is possible that this step also is blocked. With regard to this, we are particularly intrigued by a change we observed that is associated with the initial burst in competence acquisition in our time course study: the capacity for CB-induced cell collapse. CB is one of a group of agents that alters cell shape and that also induces expression of collagenase. All these agents have in common the capacity to disrupt the actin cytoskeleton; agents that alter cell shape by their effects on other cytoskeletal components do not induce collagenase expression. These observations have been offered as evidence that actin disruption is the actual stimulator. 42 However, we show that actin disruption alone, is not sufficient for induction of collagenase expression. Possibly, the shape change resulting from cell collapse is also necessary; this might allow, for example, the physical displacement of an inhibitor away from positive elements of the signal transduction cascade. In turn, fully formed focal adhesions and stress fibers, with the corresponding isometric tension that these structures create, may be necessary for cell collapse. These structure probably form as a result of appearance of α5 integrin, the expression of which is induced when stromal cells are placed in serum-containing culture. 11 The appearance of this integrin makes possible the assembly of the α5β1 fibronectin receptor and allows attachment and spreading of cells on the fibronectin component of serum. The possible relationship between expression of α5 integrin and competence to express collagenase in response to CB will be interesting to explore. 
Differential Signaling Pathways for IL-1α and Collagenase
In contrast to their incompetence to express IL-1α, freshly isolated cells produce collagenase in response to IL-1 at the same levels as subcultured cells. The basis of this selectivity was found to reside in the use of different signal transduction pathways: IL-1 activates transcription of the gene for IL-1α via a pathway requiring ROS and transcription factor NF-κB, while activation of collagenase gene transcription does not rely on this pathway. Importantly, cells freshly isolated from the corneal stroma are relatively incompetent to respond to IL-1 by induction of NF-κB DNA-binding activity, even though they contain amounts of the relevant NF-κB proteins equivalent to subcultured cells. In contrast, maximal inducible DNA-binding activity for transcription factor AP-1 was equivalent in both cell types. These results reveal a deficiency in a signaling pathway in freshly isolated corneal stromal cells needed for expression of a specific set of genes that includes the gene for IL-1α. 
Placing stromal cells in serum-containing culture causes them to enter the cell cycle and the initial burst in acquisition of competence to respond to CB, as observed in our time course study, correlated with the attainment of cell confluence and a cessation of cell replication. Because active NF-κB is essential if cells are to express collagenase in response to CB, it may be relevant that NF-κB activation has been previously connected with entry or withdrawal from the cell cycle in a variety of systems. 43 44 45 46 47 Another change in cell phenotype that we observed to correlate with the initial burst in competence acquisition was the transition to a fibroblastic spindle shape and the assembly of actin stress fibers. Again, these are events associated with activation of NF-κB in other cell types. 48 It may be interesting to follow up on both of these connections. It should be emphasized, however, that our study has examined competence for NF-κB activation, not the actual activation event. The factors that cause induction of NF-κB activity and the factors determining stromal cell competence for NF-κB activation may be quite different. 
Significance of Deficiency in NF-κB Activation for Corneal Stasis
This is the first study of which we are aware reporting incompetence to activate NF-κB in tissue stromal cells. NF-κB controls expression of a large group of stress, inflammatory, and remodeling genes, 40 and activity of NF-κB protects cells from undergoing apoptosis. 49 50 51 It is likely therefore that incompetence to activate NF-κB contributes broadly to maintaining corneal stasis, beyond its specific effect on the IL-1α autocrine loop. A question of immediate interest is whether incompetence to activate NF-κB, and the accompanying restrictions on gene expression and cell activities, is a unique property of the corneal stroma. Evidence in the literature suggests that other cell culture models besides our corneal stromal model may exhibit this property. For example, in a culture model of freshly isolated rat myometrial smooth muscle cells, serotonin can activate the IL-1α loop and collagenase expression only after cells have replicated in culture for several days and achieved confluence. 52 However, the corneal stroma is an unusually homogeneous tissue and lacks other cell types, such as tissue leukocytes or capillary endothelial cells, which are likely to be competent for IL-1α expression and NF-κB activation. 1 53 In this context, incompetence of stromal cells to activate NF-κB may be uniquely limiting for inflammatory and tissue remodeling events. Therefore, we suggest that incompetence of stromal cells to activate NF-κB may constitute a mechanism, such as immune privilege, which protects the structure of the corneal stroma. 
Figure 1.
 
Morphologic and molecular comparison of freshly isolated and subcultured stromal cells. (A) Phase contrast microscopic view of cells freshly isolated from the corneal stroma (top) and of cells subcultured through one passage (bottom). Both photographs were taken using the 43× objective. Bar, 0.4 mm. (B) Replicate culture wells of freshly isolated (top) or subcultured (bottom) stromal cells were left untreated (−) or treated with CB or PMA (treatments performed in duplicate). Shown is an autoradiograph depicting the profile of secreted 35S-methionine–labeled proteins after 24 hours of treatment, as displayed by PAGE. The arrowhead indicates the protein doublet representing procollagenase and prostromelysin as determined by immunoprecipitation analysis. The migration position of molecular size standards is indicated in kilodaltons.
Figure 1.
 
Morphologic and molecular comparison of freshly isolated and subcultured stromal cells. (A) Phase contrast microscopic view of cells freshly isolated from the corneal stroma (top) and of cells subcultured through one passage (bottom). Both photographs were taken using the 43× objective. Bar, 0.4 mm. (B) Replicate culture wells of freshly isolated (top) or subcultured (bottom) stromal cells were left untreated (−) or treated with CB or PMA (treatments performed in duplicate). Shown is an autoradiograph depicting the profile of secreted 35S-methionine–labeled proteins after 24 hours of treatment, as displayed by PAGE. The arrowhead indicates the protein doublet representing procollagenase and prostromelysin as determined by immunoprecipitation analysis. The migration position of molecular size standards is indicated in kilodaltons.
Figure 2.
 
Characterization of stromal cell transformation in culture. Freshly isolated stromal cells were plated in replicate cell culture wells. The next day (day 1) and each subsequent day through day 4, a set of cultures was left untreated (A, B, C) or was treated with CB for 24 hours (D, E, F). (A, D) day 1, (B, E) day 2, (C, F) day 3. (A through C, E) RITC-phalloidin stain to visualize filamentous actin; (D, F) immunofluorescent stain for collagenase. Arrowheads in (F) indicate collagenase-positive cells. All photographs were taken using the 43× objective. Bar (A), 0.1 mm.
Figure 2.
 
Characterization of stromal cell transformation in culture. Freshly isolated stromal cells were plated in replicate cell culture wells. The next day (day 1) and each subsequent day through day 4, a set of cultures was left untreated (A, B, C) or was treated with CB for 24 hours (D, E, F). (A, D) day 1, (B, E) day 2, (C, F) day 3. (A through C, E) RITC-phalloidin stain to visualize filamentous actin; (D, F) immunofluorescent stain for collagenase. Arrowheads in (F) indicate collagenase-positive cells. All photographs were taken using the 43× objective. Bar (A), 0.1 mm.
Table 1.
 
Activation of Corneal Stromal Cells in Culture
Table 1.
 
Activation of Corneal Stromal Cells in Culture
% CL+No. of Cells/Field
Day 1014.3
Day 2023.2
Day 330.6 ± 6.950.0
Day 455.2 ± 9.545.8
Figure 3.
 
Freshly isolated stromal cells are incompetent to sustain an IL-1α autocrine feedback loop. Northern blot analysis of total RNA from untreated (−) or IL-1–treated cultures of freshly isolated or subcultured stromal cells, probed with radiolabeled rabbit cDNAs as indicated. Treatment was performed for 2 or 24 hours before RNA isolation. Blots were probed sequentially with cDNAs as indicated. Probing for GAPD served to indicate the relative equivalence of lane loading.
Figure 3.
 
Freshly isolated stromal cells are incompetent to sustain an IL-1α autocrine feedback loop. Northern blot analysis of total RNA from untreated (−) or IL-1–treated cultures of freshly isolated or subcultured stromal cells, probed with radiolabeled rabbit cDNAs as indicated. Treatment was performed for 2 or 24 hours before RNA isolation. Blots were probed sequentially with cDNAs as indicated. Probing for GAPD served to indicate the relative equivalence of lane loading.
Figure 4.
 
A signaling pathway incorporating ROS and NF-κB is used by IL-1 for stimulation of IL-1α gene expression in corneal stromal cell subcultures. (A) Subcultured stromal cells were treated with IL-1 and harvested for preparation of nuclear lysates 0 to 60 minutes later. Protein equivalent fractions of each lysate were used in EMSA with a probe containing a cannonical NFκB binding site. The migration position of the IL-1–inducible protein–DNA complex is indicated by an arrow. The constitutive complex is indicated by an arrowhead. (B) Subcultured stromal cells were treated with IL-1 and harvested for preparation of nuclear lysates after 2 hours. EMSA was performed with a probe containing a cannonical NF-κB binding site (−). Supershift analysis was performed with antibodies to the NF-κB proteins p50, p65, and c-rel. An AP-2 antibody served as a negative control. The migration position of the inducible protein–DNA complex is indicated by an arrowhead. The supershifted subcomplexes are indicated by arrows. Competition was performed with a 50-fold excess of “cold” nonspecific probe AP-1 (NSB) or with a 10- (1:10) or 50-fold (1:50) excess of “cold” NF-κB probe. Controls include a binding reaction without nuclear lysate (Neg) and a binding reaction with a purchased HeLa cell lysate. (C) Subcultured stromal cells were untreated (−) or pretreated with PDTC for 30 minutes at the indicated dose and then were treated with IL-1 in the continued presence of PDTC for 2 hours and harvested for EMSA (left panel). The migration position of the inducible NF-κB DNA-binding complex is indicated. A second set of cells was untreated (−) or pretreated with PDTC at 50 μM for 30 minutes and then were treated with IL-1 in the continued presence of PDTC for 2 hours and harvested for northern blot analysis (right panel). The northern blot analysis was probed with a cDNA for IL-α and then were stripped and reprobed with a cDNA for GAPD. (D) Cells were untreated (−), treated with NF-κB inhibitory peptide SN50, or treated with the inactive analogue SN50M for 15 minutes. Cells were then treated with IL-1 for 2 hours in the continued presence of the peptides and harvested for northern blot analysis. One blot shown was probed with a cDNA for IL-α, and a duplicate blot was probed with GAPD to ascertain equivalence of RNA loading.
Figure 4.
 
A signaling pathway incorporating ROS and NF-κB is used by IL-1 for stimulation of IL-1α gene expression in corneal stromal cell subcultures. (A) Subcultured stromal cells were treated with IL-1 and harvested for preparation of nuclear lysates 0 to 60 minutes later. Protein equivalent fractions of each lysate were used in EMSA with a probe containing a cannonical NFκB binding site. The migration position of the IL-1–inducible protein–DNA complex is indicated by an arrow. The constitutive complex is indicated by an arrowhead. (B) Subcultured stromal cells were treated with IL-1 and harvested for preparation of nuclear lysates after 2 hours. EMSA was performed with a probe containing a cannonical NF-κB binding site (−). Supershift analysis was performed with antibodies to the NF-κB proteins p50, p65, and c-rel. An AP-2 antibody served as a negative control. The migration position of the inducible protein–DNA complex is indicated by an arrowhead. The supershifted subcomplexes are indicated by arrows. Competition was performed with a 50-fold excess of “cold” nonspecific probe AP-1 (NSB) or with a 10- (1:10) or 50-fold (1:50) excess of “cold” NF-κB probe. Controls include a binding reaction without nuclear lysate (Neg) and a binding reaction with a purchased HeLa cell lysate. (C) Subcultured stromal cells were untreated (−) or pretreated with PDTC for 30 minutes at the indicated dose and then were treated with IL-1 in the continued presence of PDTC for 2 hours and harvested for EMSA (left panel). The migration position of the inducible NF-κB DNA-binding complex is indicated. A second set of cells was untreated (−) or pretreated with PDTC at 50 μM for 30 minutes and then were treated with IL-1 in the continued presence of PDTC for 2 hours and harvested for northern blot analysis (right panel). The northern blot analysis was probed with a cDNA for IL-α and then were stripped and reprobed with a cDNA for GAPD. (D) Cells were untreated (−), treated with NF-κB inhibitory peptide SN50, or treated with the inactive analogue SN50M for 15 minutes. Cells were then treated with IL-1 for 2 hours in the continued presence of the peptides and harvested for northern blot analysis. One blot shown was probed with a cDNA for IL-α, and a duplicate blot was probed with GAPD to ascertain equivalence of RNA loading.
Figure 5.
 
Deficiency of NF-κB DNA-binding activity in freshly isolated stromal cells. (A) Freshly isolated or subcultured cells were either left untreated (−) or treated with PMA or IL-1 for 2 hours. Cells were then collected to prepare nuclear lysates. EMSA analysis was performed using NF-κB, AP-1, or E-box probes. Arrows indicate the specific protein–DNA complexes formed with the NF-κB or AP-1 probe. Controls for each probe include a binding reaction without nuclear lysate (Neg) and a binding reaction with a purchased HeLa cell lysate. (B) EMSA was performed using nuclear lysates from freshly isolated stromal cells left untreated (−) or treated with IL-1 for 2 hours, using the NF-κB probe. Competition to ascertain specificity was performed with cold probe (Spec Inhib). The arrowhead indicates the specific, IL-1–inducible NF-κB complex. Arrows indicate the position of complexes shifted with antibodies (Ab) specific for the p50 or p65 subunits of NF-κB. An antibody for the unrelated transcription factor AP-2 served as a negative control. (C) Western blot of total cell lysates from freshly isolated or subcultured cells left untreated (−) or treated with IL-1 for 2 hours. Duplicate blots were probed with antibody to the p50 or the p65 subunits of NF-κB.
Figure 5.
 
Deficiency of NF-κB DNA-binding activity in freshly isolated stromal cells. (A) Freshly isolated or subcultured cells were either left untreated (−) or treated with PMA or IL-1 for 2 hours. Cells were then collected to prepare nuclear lysates. EMSA analysis was performed using NF-κB, AP-1, or E-box probes. Arrows indicate the specific protein–DNA complexes formed with the NF-κB or AP-1 probe. Controls for each probe include a binding reaction without nuclear lysate (Neg) and a binding reaction with a purchased HeLa cell lysate. (B) EMSA was performed using nuclear lysates from freshly isolated stromal cells left untreated (−) or treated with IL-1 for 2 hours, using the NF-κB probe. Competition to ascertain specificity was performed with cold probe (Spec Inhib). The arrowhead indicates the specific, IL-1–inducible NF-κB complex. Arrows indicate the position of complexes shifted with antibodies (Ab) specific for the p50 or p65 subunits of NF-κB. An antibody for the unrelated transcription factor AP-2 served as a negative control. (C) Western blot of total cell lysates from freshly isolated or subcultured cells left untreated (−) or treated with IL-1 for 2 hours. Duplicate blots were probed with antibody to the p50 or the p65 subunits of NF-κB.
 
The authors thank Marie Girard, Jeanne Carniero, and Wendy Rowland for technical assistance, and Katherine Strissel and Judith West–Mays for helpful comments on the manuscript. The generous contribution of reagents from the following scientists is gratefully acknowledged: Constance Brinckerhoff of Dartmouth Medical School (rabbit collagenase antiserum), Yasuji Furutani of Dainippon Pharmaceuticals, Japan (rabbit IL-1α cDNA), and Robert Allen of the American Red Cross, St. Louis, MO (human GAPD cDNA). 
Nishida T. Cornea. Krachmer J Mannis MJ Holland EJ eds. Cornea: Fundamentals of Cornea and External Disease. 1997; Mosby St. Louis.
Davison PF, Galbavy EJ. Connective tissue remodeling in corneal and scleral wounds. Invest Ophthalmol Vis Sci. 1986;27:1478–1484. [PubMed]
Binder PS. Radial keratotomy and excimer laser photorefractive keratectomy for the correction of myopia. J Refract Corneal Surg. 1994;10:443–664. [PubMed]
Grosvenor T. How predictable are the results of excimer laser photorefractive keratectomy? A review. Optom Vis Sci.. 1995;72:698–712. [CrossRef] [PubMed]
Wilson SE, Kim WJ. Keratocyte apoptosis: implications on corneal wound healing, tissue organization, and disease. Invest Ophthalmol Vis Sci. 1998;39:220–226. [PubMed]
Streilein JW, Ksander BR, Taylor AW. Immune deviation in relation to ocular immune privilege. J Immunol. 1997;158:3557–3560. [PubMed]
Weimar V. Healing processes in the cornea. Duke-Elder S Perkins ES eds. The Transparency of the Cornea. 1960;111. Blackwell Scientific Publications Oxford.
Fini ME. Keratocyte and fibroblast phenotypes in the repairing corneal stroma. Prog Retinal Eye Res. 1999;18:529–551. [CrossRef]
Garana RMR, Petroll M, Chen W–T, et al. Radial keratototomy: II, role of the myofibroblast in corneal wound contraction. Invest Ophthalmol Vis Sci.. 1992;33:3271–3282. [PubMed]
Jester JV, Petroll W, Barry PA, Cavanagh HD. Expression of α-smooth muscle (α-SM) actin during corneal stromal wound healing. Invest Ophthalmol Vis Sci. 1995;36:809–819. [PubMed]
Masur SK, Cheung JKH, Antohi S. Identification of integrins in cultured corneal fibroblasts and in isolated keratocytes. Invest Ophthalmol Vis Sci. 1993;34:2690–2698. [PubMed]
Matsubara M, Girard MT, Kublin CL, Cintron C, Fini ME. Differential roles for two gelatinolytic enzymes of the matrix metalloproteinase family in the remodelling cornea. Dev Biol. 1991;147:425–439. [CrossRef] [PubMed]
Girard MT, Matsubara M, Kublin C, Tessier M, Cintron C, Fini ME. Stromal fibroblasts synthesize collagenase and stromelysin during long-term remodeling of repair tissue. J Cell Sci. 1993;104:1001–1011. [PubMed]
West–Mays JA, Sadow PM, Tobin TW, Strissel KJ, Cintron C, Fini ME. Regulation of repair phenotype in corneal fibroblasts by an IL-1α feedback loop. Invest Ophthalmol Vis Sci. 1997;38:1367–1379. [PubMed]
Kuter I, Johnson–Wint B, Beaupre N, Gross J. Collagenase secretion accompanying changes in cell shape occurs only in the presence of a biologically active cytokine. J Cell Sci. 1989;92:423–485.
Fini ME, Girard MT. The pattern of metalloproteinase expression by corneal fibroblasts is altered with passage in cell culture. J Cell Sci. 1990;97:373–383. [PubMed]
Girard ME, Matsubara M, Fini ME. Transforming growth factor-β and Il-1 modulate expression of metalloproteinases by corneal stromal cells. Invest Ophthalmol Vis Sci. 1991;32:2441–2454. [PubMed]
Jester JV, Barry PA, Lind GJ, Petroll WM, Garana R, Cavanagh HD. Corneal keratocytes: in situ and in vitro organization of cytoskeletal contractile proteins. Invest Ophthalmol Vis Sci. 1994;35:730–743. [PubMed]
Fini ME, Strissel KJ, Girard MT, West–Mays JA, Rinehart WB. Interleukin-1α mediates collagenase synthesis stimulated by phorbol myristate acetate. J Biol Chem. 1994;269:11291–11298. [PubMed]
West–Mays JA, Strissel KJ, Sadow PM, Fini ME. Competence for collagenase gene expression by tissue fibroblasts requires activation of an IL-1α autocrine loop. Proc Natl Acad Sci USA. 1995;92:6768–6772. [CrossRef] [PubMed]
Funderburgh JL, Funderburgh ML, Mann MM, Prakash S, Conrad GW. Synthesis of corneal keratan sulfate proteoglycans by bovine keratocytes in vitro. J Biol Chem. 1996;371:31431–31436.
Strissel KJ, Girard MT, West–Mays JA, et al. Role of serum amyloid A as a signalling intermediate in the IL-1 and PMA-stimulated signalling pathways regulating expression of rabbit fibroblast collagenase. Exp Cell Res. 1997;237:275–287. [CrossRef] [PubMed]
Dinarello CA, Savage N. Interleukin-1 and its receptor. Crit Rev Immunol. 1989;9:1–20. [PubMed]
Johnson-Muller B, Gross J. Regulation of corneal collagenase production: epithelial-stromal cell interactions. Proc Natl Acad Sci USA. 1978;75:4417–4422. [CrossRef] [PubMed]
Brinckerhoff CE, Fini ME, Ruby L, Plucinska IM, Borges KM, Karmilowicz MJ. Coordinate regulation of metalloproteinase gene expression in synovial cells. Sen A Thornhill,/SNM> T eds. Development and Diseases of Cartilage and Bone Matrix. 1987;299–317. Alan R. Liss New York.
Fini ME, Plucinska IM, Mayer AS, Gross RH, Brinckerhoff CE. A gene for rabbit synovial cell collagenase: member of a family of metalloproteinases which degrades the connective tissue matrix. Biochemistry. 1987;26:6156–6164. [CrossRef] [PubMed]
Furutani Y, Notake M, Yamayoshi M, et al. Cloning and characterization of the cDNAs for human and rabbit interleukin-1 precursor. Nucleic Acids Res. 1985;13:5869–5882. [CrossRef] [PubMed]
Feinberg AP, Vogelstein B. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem. 1983;132:6–13. [CrossRef] [PubMed]
Allen RW, Trach KA, Hoch JA. Identification of the 37-kDa protein displaying a variable interaction with the erythroid cell membrane as glyceraldehyde-3-phosphate dehydrogenase. J Biol Chem. 1987;262:649–653. [PubMed]
Hembry RM, Murphy G, Cawston TE, Dingle JT, Reynolds JJ. Characterization of a specific antiserum for mammalian collagenase from several species: immunolocalization of collagenase in rabbit chondrocytes and uterus. J Cell Sci. 1986;81:105–123. [PubMed]
Harlow E, Lane D. Antibodies: A Laboratory Manual. 1988; Cold Spring Harbor Laboratory Press Cold Spring Harbor, NY.
Schreiber E, Mathias P, Muller MM, Schaffner W. Rapid detection of octamer binding proteins prepared from a small number of cells. Nucleic Acids Res. 1989;17:6419. [CrossRef] [PubMed]
Furutani Y, Notake M, Fukui T, et al. Complete nucleotide sequence of the gene for human interleukin-1α. Nucleic Acids Res. 1986;14:3167–3179. [CrossRef] [PubMed]
Lin Y-Z, Yao S, Veach RA, Torgerson TR, Hawiger J. Inhibition of nuclear translocation of transcription factor NF-κB by a synthetic peptide containing a cell membrane-permeable motif and nuclear localization sequence. J Biol Chem. 1995;270:14255–14258. [CrossRef] [PubMed]
Barnes PJ, Karin M. Nuclear factor-kappaB: a pivotal transcription factor in chronic inflammatory diseases. New Engl J Med. 1997;336:1066–1071. [CrossRef] [PubMed]
Karin M, Liu Zg, Zandi E. AP-1 function and regulation. Curr Opin Cell Biol.. 1997;9:240–246. [CrossRef] [PubMed]
Fini ME, Cook JR, Mohan R, Brinckerhoff CE. Regulation of matrix metalloproteinase gene expression. Parks W Meacham R eds. Matrix Metalloproteinases. 1998;299–355. Academic Press San Diego.
Rosette C, Karin M. Cytoskeletal control of gene expression: depolymerization of microtubules activates NF-κB. J Cell Biol. 1995;128:1111–1119. [CrossRef] [PubMed]
Bauerle PA, Baltimore D. IκB: a specific inhibitor of the NF-κB transcription factor. Science. 1988;242:540–546. [CrossRef] [PubMed]
Bauerle PA, Henkel T. Function and activation of NF-κB in the immune system. Annu Rev Immunol. 1994;12:141–179. [CrossRef] [PubMed]
Schreck R, Rieber P, Baeuerle PA. Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF-κB transcription factor and HIV-1. EMBO J. 1991;10:2247–2258. [PubMed]
Aggeler J, Frisch SM, Werb Z. Changes in cell shape correlate with collagenase gene expression in rabbit synovial fibroblasts. J Cell Biol. 1983;98:1662–1671.
Baldwin AS, Jr, Azizhkan JC, Jensen DE, Beg AA, Codley LR. Induction of NF-kappaB binding activity during the G0-to-G1 transition in mouse fibroblasts. Mol Cell Biol. 1991;11:4943–4951. [PubMed]
Cressman DE, Greenbaum LE, Haber BA, Taub R. Rapid activation of post-hepatectomy factor/Nuclear Factor κB in hepatocytes, a primary response in the regenerating liver. J Biol Chem. 1994;269:30429–30435. [PubMed]
Perkins ND, Felzien LK, Betts JC, Leung K, Beach DH, Nabel DJ. Regulation of NFκB by cyclin-dependent kinases associated with the p300 coactivator. Science. 1997;275:523–527. [CrossRef] [PubMed]
Mohan R, Rinehart WB, Bargagna–Mohan P, Fini ME. Gelatinase B-LacZ transgenic mice: a model for mapping gelatinase B gene expression during developmental and injury-related tissue remodelling. J Biol Chem. 1998;273:25903–25914. [CrossRef] [PubMed]
Seitz CS, Lin Q, Deng H, Khavari PA. Alterations in NF-κB function in transgenic epithelial tissue demonstrate a growth inhibitory role for NF-κB. Proc Natl Acad Sci USA. 1998;95:2307–2312. [CrossRef] [PubMed]
Juliano RL, Haskill S. Signal transduction from the extracellular matrix. J Cell Biol. 1993;120:577–585. [CrossRef] [PubMed]
Beg AA, Baltimore D. An essential role for NF-kB in preventing TNF-α-induced cell death. Science. 1996;274:782–789. [CrossRef] [PubMed]
Wang C-Y, Mayo MW, Baldwin AS, Jr. TNF- and cancer therapy-induced apoptosis: potentiation by inhibition of NF-kB. Science. 1996;274:784–787. [CrossRef] [PubMed]
Van Antwerp DJ, Martin SJ, Kafri T, Green DR, Verma IM. Suppression of TNF-α-induced apoptosis by NF-κB. Science. 1996;274:787–789. [CrossRef] [PubMed]
Wilcox BD, Dumin JA, Jeffrey JJ. Serotonin regulation of interleukin-1 messenger RNA in rat uterine smooth muscle cells. Relationship to the production of interstitial collagenase. J Biol Chem.. 1994;269:29658–29664. [PubMed]
Fini ME, Cook JR, Mohan R. Proteolytic Mechanisms in Corneal Ulceration and Repair. Arch Dermatol Res.. 1998;290:S12–23. [CrossRef] [PubMed]
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×