Progressive renal diseases are characterized by an increased synthesis of extracellular matrix (ECM) components. The mechanisms involved in the development of these alterations are not completely known, but a crucial role for TGF-β1 has been suggested. Moreover, the ability of the ECM to modulate the phenotypic expression of different cell types has been widely described. In experiments presented here, human mesangial cells (HMC) were grown on collagen type I (COL I) or IV (COL IV). ECM protein and TGF-β1 mRNA expression were evaluated by Northern blot analysis, and TGF-β1 secretion was evaluated by ELISA. The involvement of tyrosine kinase and serine-threonine kinase pathways was studied by Western blot analysis, immunofluorescence, and in vitro kinase assays. HMC cultured on COL I showed an increased mRNA expression of COL I and COL IV, fibronectin, and TGF-β1. Both tyrosine phosphorylation and integrin-linked kinase (ILK) activity increased when HMC were cultured on COL I, and blockade of these pathways inhibited the increased secretion of TGF-β1. In conclusion, the present results support a role for extracellular COL I in the regulation of TGF-β1 synthesis during progressive renal sclerosis and fibrosis and the subsequent increase in newly synthesized ECM proteins. In addition, ILK, along with the tyrosine kinases, participates in the genesis of this effect.
- tyrosine phosphorylation
- integrin-linked kinase
glomerular sclerosis and interstitial fibrosis are characteristic findings in different forms of progressive renal disease (19) and aging (28). These pathological changes are caused by the accumulation of extracellular matrix (ECM) proteins at different levels of the renal cortex, as a consequence of an imbalance in the normal turnover of ECM components. Both normal ECM components of renal cortex, such as collagen type IV (COL IV) and fibronectin, and ECM proteins not usually found in significant amounts, such as collagens type I (COL I) and type III, may increase in these situations (1, 2, 6, 34, 38). In the healthy glomerulus, COL I is not normally present, and its appearance in some pathological conditions could modify the behavior of glomerular components, such as mesangial cells.
TGF-β1 is a multifunctional regulator of cell proliferation and differentiation that contributes to the accumulation of ECM proteins by stimulating their synthesis and/or decreasing the activity of some extracellular proteases (21, 23). In mesangial cells, TGF-β1 inhibits growth and stimulates the synthesis of collagens, fibronectin, laminin, and proteoglycans (22, 33, 37, 39). TGF-β1 may play a crucial role in the pathogenesis of glomerular sclerosis and interstitial fibrosis (27), although the mechanisms by which TGF-β1 secretion is stimulated are not completely known.
The importance of cell adhesion to the ECM in the regulation of gene expression, proliferation, apoptosis, and migration has been described previously (3, 12, 30, 35, 36). Transmembrane receptors such as integrins (29) could be involved in this phenotypic modulation. Two main pathways are activated by integrin stimulation. First, several tyrosine kinases are recruited into focal adhesions, leading to a sequential protein tyrosine phosphorylation (8). Second, a newly identified integrin-linked kinase (ILK), with serine-threonine kinase activity, interacts with the cytoplasmic tail of β1-, β2-, and β3-integrin subunits, playing a central role in the cross-modulation between growth factor response and integrin signaling (7, 15, 26).
The present experiments were designed to test the hypothesis that the interaction of mesangial cells with abnormal ECM proteins could initiate phenotypic changes that are characteristic of, or that could promote, renal disease. In particular, we hypothesized that the contact of mesangial cells with COL I could induce an increase in ECM protein synthesis compared with cells in contact with COL IV, one of the main components of normal mesangial cell ECM. We also hypothesized that TGF-β1 synthesis by mesangial cells could mediate the altered production of ECM proteins elicited by contact with COL I. Finally, we tried to define the intracellular mechanisms responsible for the COL I-dependent changes in TGF-β1 synthesis by examining both the tyrosine kinase cascade and the ILK system.
MATERIALS AND METHODS
Human mesangial cell culture.
Human mesangial cells (HMC) were cultured as previously described (17). Isolated glomeruli were treated with collagenase (Sigma Chemical, St. Louis, MO), plated in plastic culture dishes, and maintained in RPMI 1640 supplemented with 10% FCS, l-glutamine (1 mM), penicillin (0.66 μg/ml), and streptomycin sulfate (60 μg/ml) and buffered with HEPES and bicarbonate, pH 7.4, in a 5% CO2 atmosphere. Culture media were changed every 2 days. When cells reached confluence, they were subcultured at a ratio of 1:4, using the same incubation medium. Experiments were performed in passages 3–5. The identity of the cells was confirmed by morphological and functional criteria, as previously described (17). The source of cultured mesangial cells was macroscopically normal cortex tissue that was obtained from human kidney soon after nephrectomy for renal carcinoma. This procedure was approved by the Institutional Review Board from the Principe de Asturias Hospital.
Trypsinized HMC were seeded (4 × 104 cells/cm2) on petri dishes previously treated for 16 h at 4°C with a solution of COL I or COL IV (Sigma) (16). Briefly, culture dishes were coated with 12.5 μg/ml COL I or COL IV in bicarbonate buffer (15 mM Na2CO3 and 35 mM NaHCO3, pH 9) to form a thin film of collagen. The petri dishes were washed with Hanks' balanced salt solution to restore pH 7.5. In some cases, COL I and COL IV were treated with pepsin (0.2 mg/ml, 18 h, 4°C) or with collagenase IA (3,000 U/ml, 30 min, 37°C) before the petri dishes were coated.
Cells were grown for 48 h with RPMI 1640 supplemented with 10% FBS and were then serum-starved for varying periods of time. The supernatants were collected, and total RNA was extracted from the cells. In some experiments, treatments were added 24 h before the incubation periods were finished. In selected experiments, cells were grown directly on plastic to investigate TGF-β1 mRNA expression and secretion and COL IV mRNA expression.
In some experiments, cells were transfected with His-V5-tagged kinase-deficient ILK or His-V5-tagged wild-type ILK cDNA (kindly provided by S. Dedhar, British Columbia Cancer Agency, University of British Columbia, Vancouver, BC, Canada) by using Lipofectin reagent (Life Technologies, Gaithersburg, MD) according to the manufacturer's guidelines. Subconfluent cells (60% confluency) were serum-deprived and incubated with 2 μg of cDNA plasmid and 4 μl of Lipofectin reagent. Control cultures received the empty pcDNA 3 vector. Transfections were carried out overnight in Optimem medium (GIBCO BRL, Grand Island, NY), and the cells were harvested 24 h later. Transfected cells were analyzed for cell viability by using the trypan blue exclusion viability assay. Before the experiments were started, the transfection efficiency in this cell type was evaluated by transfecting a plasmid containing green fluorescent protein (pEGFP; Clontech, Palo Alto, CA) in the same experimental conditions, as previously reported (18). Expression of the transfected constructs was verified by immunoblotting with anti-V5 antibody (Invitrogen).
Northern blot analysis of mRNA expression.
Total mRNA was isolated by repeated phenol-chloroform extractions and isopropanol precipitations (4). Total RNA was denatured, electrophoresed through a 1% agarose gel with 0.66 M formaldehyde, transferred to nitrocellulose membranes (Amersham, Amersham, UK), and fixed by UV cross-linking. The integrity and equal loading of RNA samples were assessed by methylene blue staining (2). The membranes were prehybridized at 65°C for 2 h in 10% dextran sulfate, 1% SDS, 1 M NaCl, and 0.1 mg/ml denatured salmon sperm DNA. cDNA inserts were separated from their vectors and radiolabeled with 5 μCi [32P]deoxycytosine 5′-triphosphate (3.0 Ci/mmol; Amersham) using a radiolabeled system (Ready To Go; Pharmacia, Uppsala, Sweden). The probes used were mouse α1 (IV) collagen (a gift from M. Kurkinen, Center for Molecular Medicine, Genetic, Department of Pathology, Wayne State University School of Medicine, Detroit, MI), mouse fibronectin, mouse TGF-β1, and rat α2 (I) collagen. All probes used cross-react with HMC mRNA. Blots were hybridized overnight in the same buffer used for prehybridization, with 106 cpm/ml probe at 65°C. The membranes were washed twice for 5 min in 2× SSC (20× SSC: 3 M NaCl and 0.3 M sodium citrate, pH 7.0) at room temperature and then for 5 min in 2× SSC and 1% SDS at 65°C. Autoradiographs were performed at −80°C for 5–6 h. Blots were stripped in 1% SDS and 0.1× SSC for 30 min at 100°C and subsequently hybridized with the other probes used. Densitometric analysis of the exposed film was performed with an Apple scanner and appropriate software (NIH Image, National Institutes of Health, Bethesda, MD).
Measurement of immunoreactive TGF-β1 by ELISA.
Culture supernatant (250 μl) was removed from each well and incubated with 5 μl of 1 N HCl for 60 min to activate latent TGF-β1. After the supernatants were neutralized with 1 N NaOH, samples were analyzed with a commercial TGF-β1 sandwich ELISA (Promega, Madison, WI) according to the manufacturer's recommendations (17). A standard curve was constructed using serial dilutions of ultrapure human TGF-β1 (Promega). Cells were harvested by trypsinization and counted in a Neubauer camera. Results are expressed as picograms of TGF-β1 per 105 cells. Each sample was measured in duplicate.
Analysis of tyrosine phosphorylation by immunofluorescence.
For immunofluorescence analysis of tyrosine phosphorylation (11), HMC were grown in Chamber slides (Nunc, Naperville, IL) until subconfluency, chilled on ice, and washed three times with cold PBS. Subsequently, cells were fixed in 4% paraformaldehyde for 30 min and permeabilized with 0.2% Triton X-100 for 20 min. To reduce nonspecific binding, we incubated cells in PBS-3% BSA overnight and then with 10% normal goat serum for 30 min. Tyrosine-phosphorylated proteins were detected using a mouse monoclonal anti-human phosphotyrosine antibody (Sigma) diluted 1:100 in PBS-1% BSA. The secondary antibody was a FITC-conjugated goat anti-mouse antiserum diluted 1:128 in PBS-1% BSA. Nonimmune IgG was used as a negative control, and no fluorescence staining was detectable in these cultures. Samples were examined using a Zeiss fluorescence microscope (Carl Zeiss, Göttingen, Germany). Photomicrographs were taken at constant standardized exposure times using Kodak Ektachrome 400 film (Kodak, Rochester, NY).
Protein extraction, immunoprecipitations, and immunoblot analysis.
The techniques for protein extraction, immunoprecipitation, and immunoblot analysis have been described previously (11). Briefly, cell cultures were chilled on ice and washed three times with cold PBS supplemented with 0.2 mM vanadate. Subsequently, the culture flasks were incubated for 30 min with 0.7 ml of RIPA lysis buffer (150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 50 mM Tris·HCl, pH 7.2, 1 mM PMSF, and 0.2 mM sodium vanadate, with 1 μg/ml each of aprotinin, leupeptin, and pepstatin). The protein concentration in each lysate was determined spectrophotometrically (BCA protein assay reagent; Pierce, Rockford, IL). For each immunoprecipitation, 300 μg of extracted cell proteins were used. The cell lysates were preadsorbed with nonimmune IgG and protein G-Sepharose (Amersham) for 60 min at 4°C. Incubations with the appropriate antibodies were performed overnight at 4°C. Nonimmune IgG was used as negative control in all experiments. The antigen-antibody complexes were precipitated with 30 μl of protein G-Sepharose, and the beads were pelleted and washed five times with RIPA buffer. The extracted proteins from mesangial cell lysates were solubilized by boiling in SDS loading buffer and analyzed by SDS-PAGE under reducing conditions. Subsequently, the proteins were transferred to 45-μm-pore nitrocellulose membranes, and Western blot analysis was performed. Appropriate primary antibodies including anti-ILK (affinity-purified rabbit polyclonal antibody; Upstate Biotchnology, Lake Placid, NY) and anti-phosphotyrosine (mouse monoclonal antibody; Upstate Biotechnology) at 1:1,000 dilutions were used, followed by incubation with either anti-mouse IgG (Sigma) or anti-rabbit-IgG (Chemicon International, Temecula, CA) peroxidase-conjugated secondary antibodies. Immunoblots were developed with an enhanced chemiluminiscence system (Supersignal System; Pierce) and subsequent exposure to X-OMAT film (Kodak).
Measurement of ILK activity.
ILK activity was determined in cell extracts by immunoprecipitation (24). Myelin basic protein was used as a substrate for ILK, and [32P]ATP was used as the phosphate donor. The phosphorylated proteins were electrophoresed on 12% SDS-PAGE gels and, after exposure to autoradiographic film, were quantified by densitometry using an Apple scanner and appropriate software (NIH Image).
Results are means ± SE of a variable number of experiments (n). Northern blot densitometry data were corrected for variations in gel loading by accounting for the relative intensities of the 28S rRNA band. When n < 10, nonparametric statistics were used for comparisons (Friedman's test and Wilcoxon test). In the other cases, ANOVA was used. P < 0.05 was considered statistically significant.
HMC cultured on COL I exhibit overexpression of ECM protein mRNA.
The first group of experiments was performed to test the hypothesis that COL I could modulate the synthesis of ECM proteins by mesangial cells. mRNA expression for COL IV, fibronectin, and COL I was measured in cells grown on both collagen types. Significantly higher amounts of these mRNAs were detected in cells grown on COL I compared with cells on COL IV (Fig. 1). No differences between cells grown on COL I and COL IV were detected with respect to the degree of cell confluence as determined by protein measurement and cell counting (1.12 × 105 ± 0.32 × 105 cells/cm2 in COL IV and 1.25 × 105 ± 0.23 × 105 cells/cm2 in COL I) or cell viability (>98% cells excluded trypan blue in both cases).
HMC cultured on COL I show increased mRNA expression and secretion of TGF-β1.
Because TGF-β1 is considered one of the main cytokines involved in the regulation of ECM protein synthesis, we also tested the ability of COL I to modulate the TGF-β1 mRNA expression and synthesis in mesangial cells. As with ECM mRNAs, COL I compared with COL IV induced a significantly higher expression of TGF-β1 mRNA (Fig. 2A). These changes in mRNA expression were followed by an increased synthesis and secretion of immunoreactive TGF-β1 (Fig. 2B). For this reason, in the rest of our experiments, we measured indistinctly the effect of COL I on TGF-β1 production at the mRNA level (see Figs. 5 and 6) or at the protein level (see Figs. 3 and 9). The increase in TGF-β1 synthesis was significantly higher 24 h after cells were serum-starved and remained so for 96 h (Fig. 3). In selected experiments, cells were grown directly on plastic to compare the TGF-β1 biosynthesis with the HMC standard culture (COL IV). Immunoreactive TGF-β1 production was lower in cells grown on plastic (5.50 ± 1.41 and 10.03 ± 1.81 pg TGF-β1/105 cells at 72 and 96 h after serum deprivation, respectively) than in cells grown on either collagen type.
Increased COL IV mRNA expression induced by contact between COL I and HMC is partially prevented by TGF-β1 blockade.
To determine whether the COL IV overexpression induced by COL I in HMC is mediated by TGF-β1, we analyzed COL IV mRNA expression in the presence of a blocking anti-TGF-β1 antibody (30 μg/ml). This antibody partially prevented the increase in COL IV mRNA expression in cells grown on COL I (Fig. 4). COL IV mRNA expression in cells grown on COL IV was not affected by anti-TGF-β1 antibody treatment. A nonimmune IgG was used as control.
Fibrillar integrity of COL I is essential for genesis of changes observed in HMC cultured on this substrate.
To determine whether COL IV and COL I must be intact to modulate COL IV and TGF-β1 mRNA expression, we disrupted the peptidic sequences of COL I and COL IV with pepsin treatment or with collagenase IA before the cells were seeded. The effect elicited by COL I culture on COL IV (Fig. 5A) and TGF-β1 (Fig. 5B) mRNA expression was prevented by pepsin treatment, whereas pepsin treatment of COL IV only minimally modified the mRNA expression for these proteins with respect to the native COL IV. HMC grown on plastic showed patterns of COL IV and TGF-β1 mRNA expression similar to those of cells grown on COL IV. In additional experiments, COL I was treated with collagenase IA before HMC were seeded. Under these conditions, the increase in TGF-β1 mRNA expression produced by COL I disappeared (Fig. 5C), confirming the importance of the complete structure of COL I for this effect.
Both tyrosine phosphorylation and ILK activation are involved in genesis of COL I-dependent effects.
The possible involvement of tyrosine kinases in the changes observed in TGF-β1 synthesis after cell seeding on COL I was studied by comparing protein tyrosine phosphorylation in cells grown on COL I or COL IV. Cells grown on COL I had a greater fluorescence signal for tyrosine-phosphorylated proteins than cells grown on COL IV (Fig. 6A). A similar COL I-dependent increase in tyrosine phosphorylation was demonstrated by Western blot analysis; the densitometric analysis of two representative bands corresponding to 120 and 60 kDa showed an increased phosphorylation of these proteins in cells grown on COL I compared with COL IV (Fig. 6B). Two different tyrosine kinase inhibitors and their respective inactive analogs were used to demonstrate the role of tyrosine kinase in the COL I effects. Both genistein (2 μg/ml for 24 h) and herbimycin A (2 μM for 24 h), partially blocked the increased TGF-β1 mRNA expression induced by COL I cultures (Fig. 6C). Neither the inactive analog of genistein, daidzein (2 mg/ml for 24 h), nor inactive herbimycin A prepared by treatment with 5 mM reduced glutathione for 5 min (35a) before being added to cells at 2 μM for 24 h blocked the increased TGF-β1 mRNA expression induced by COL I cultures. When cultures were assayed by trypan blue exclusion, no toxic effects of genistein or herbimycin A were found at the dose and incubation time used.
Cell culture on COL I also induced an increase in ILK activity without a corresponding change in ILK protein content (Fig. 7). This stimulatory effect of COL I was time dependent (Fig. 8), reaching a maximum after 72 h of cell contact with the culture substrate, whereas ILK activity of cells cultured on COL IV remained unchanged (Fig. 8, bottom). No activity was found in negative controls in all experiments. To assess whether the COL I-dependent stimulation of TGF-β1 synthesis is mediated by the increased ILK activity, we transfected cells grown on COL I with plasmids expressing a kinase-deficient ILK or wild-type ILK. Overexpression of kinase-deficient ILK prevented the increased TGF-β1 synthesis induced by COL I (Fig. 9A). Transfection efficiency was ∼60%, and the expression of the constructs was controlled by testing the cellular content of the exogenous protein V5, also included in the transfection plasmid (Fig. 9A). In these same cultures, ILK activity was also significantly reduced (Fig. 9B).
Mesangial cells are embedded in the glomerular mesangial matrix. The turnover of this matrix is controlled through a dynamic equilibrium between the synthesis and degradation of its components (5). In normal conditions the mesangial matrix is composed of COL IV, laminin, fibronectin, nidogen, entactin, and proteoglycans, among others. In some pathological situations that lead to glomerular sclerosis, there is overexpression of the normal ECM components and appearance of COL I and other collagens that are normally limited to the interstitium. Previous studies have demonstrated that COL I and plasma-type fibronectin promote apoptosis in cultured rat mesangial cells, whereas COL IV and laminin protect them from serum deprivation-induced apoptosis (26). In this article, we propose that COL I, when in contact with mesangial cells, induces an increased synthesis of ECM components through TGF-β1 upregulation.
Multiple studies have analyzed the relationships among ECM proteins and cell morphology, proliferation, and apoptosis (3, 11, 12, 16, 24, 26, 31). However, the effect of ECM composition on cytokine synthesis has been poorly explored. The results reported here demonstrate a differential synthetic pattern by HMC depending on the ECM component on which they are grown in vitro. Thus an increase in TGF-β1 mRNA expression followed by an increase in TGF-β1 secretion and an increased expression of mRNA from different ECM proteins were detected in mesangial cells cultured on COL I compared with cells grown on COL IV. However, from the data presented here, we cannot conclude whether an increase in the accumulation of ECM proteins is occurring in the presence of COL I because the synthesis and assembly of ECM components is also posttranscriptionally regulated. Changes in mRNA expression of some of these ECM proteins are dependent on TGF-β1 production, because the presence of a TGF-β1-neutralizing antibody in the incubation medium reduced the increased COL IV mRNA expression observed in cells grown on COL I. The TGF-β1-neutralizing antibody does not completely blunt the COL IV mRNA expression in cells grown on COL IV, suggesting the existence of alternative mechanisms of regulation of ECM protein synthesis (14, 37).
The mechanisms involved in COL I-induced TGF-β1 upregulation have begun to be analyzed, and of particular importance is the peptidic sequence of COL I and the role of intracellular kinases. The fibrillar acid-soluble COL I used for the experiments was treated with pepsin to remove the aminotelopeptides (10), disrupting the intermolecular cross-linking between telopeptides and the triple helix structure. The fibrils of pepsin-treated COL I are thicker than normal (32, 33). These telopeptides also are critical for the effect of COL I on TGF-β1 synthesis, because pepsin treatment completely abrogated it. Moreover, the digestion of COL I with collagenase produced the same effect, confirming that the triple helix structure is essential in COL I regulation of TGF-β1. Our results support the relevance of COL I integrity modulation of HMC TGF-β1 mRNA levels.
Both tyrosine kinase and ILK seem to be involved in COL I-dependent upregulation of TGF-β1 synthesis. Protein tyrosine phosphorylation, a marker of tyrosine kinase activity, increased in cells cultured on COL I. Both genistein, a pharmacological blocker of these kinases, and herbimycin A, a chemically dissimilar tyrosine kinase inhibitor, partially inhibited the effect of COL I on mRNA TGF-β1 expression. A similar mechanism was described by Ford et al. (9) with respect to the stimulation of TGF-β1 synthesis by angiotensin II. In those experiments, genistein blocked the increased angiotensin II-dependent TGF-β1 synthesis. In contrast to our results, Miralem et al. (24) found a decrease in tyrosine phosphorylation in mesangial cells cultured on COL I, but the time periods analyzed in these experiments were shorter than ours and were therefore not comparable. ILK activity was augmented after the cells were seeded on COL I, and the transfection of the cells with a dominant-negative form of ILK abrogated the effect of COL I on TGF-β1 synthesis, thus supporting the importance of this pathway in the regulation of TGF-β1 production. A link between ILK and proteinuria has been proposed (20), and increased ILK staining in the early phases of glomerular sclerosis in the kidneys of diabetic patients has been described (13). Because TGF-β1 expression is also increased in diabetic patients, there may be a link between ILK and TGF-β1, the main cytokine involved in the renal fibrotic process (40).
Taking all of these results together, we propose the following mechanism for ECM regulation of the HMC phenotype. Two structurally related but distinct ECM proteins, such as COL IV and COL I, activate both tyrosine kinases and ILK, but with different intensities. As a consequence, in COL I, cells overexpress TGF-β1 mRNA and secrete more TGF-β1. This TGF-β1, when released to the extracellular compartment, induces the mRNA synthesis of diverse ECM proteins that may include components different from those normally present. Our results could be very relevant to the pathophysiological consequences of glomerular sclerosis, a common marker of different chronic renal diseases (19, 29). In this context, quantitative and qualitative changes occur in ECM composition (1, 2, 6, 35, 39). Our working hypothesis is that mesangial cells from injured glomeruli could contact abnormal ECM proteins, resulting in changes in the mesangial cell synthetic phenotype. Consequently, increased local synthesis of TGF-β1 would take place, with further accumulation of normal and abnormal ECM proteins in glomerular structures. This amplification signal could contribute to the progression of chronic renal disease.
In summary, the present results support a role for ECM matrix proteins in the regulation of TGF-β1synthesis, with the subsequent increase of newly synthesized ECM proteins. In addition to the well-described system of tyrosine kinases, ILK seems to play a relevant role in the genesis of this stimulatory effect. This mechanism could be relevant in the progression of chronic renal disease.
This work was supported by Comision Interministerial de Ciencia y Tecnologia (CICYT) Grants SAF-2001/0395 (to D. Rodríguez-Puyol) and SAF 2001/1036 (to M. Rodríguez-Puyol) and by Fondo de Investigaciones Sanitaria Grant 01/0434 (to M. Rodríguez-Puyol). R. Ortega-Velazquez is a fellow from Ministerio de Educacion, Cultura y Deportes; M. Gonzalez-Rubio is a postdoctoral fellow from Comunidad Autonoma de Madrid; and M. P. Ruiz-Torres is a postdoctoral fellow from CICYT.
We are indebted to Fuad Ziyadeh for providing the neutralizing anti-TGF-β1 antibody.
↵* R. Ortega-Velazquez and M. Gonzalez-Rubio contributed equally to this work.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society