An oestrogen-receptor-alpha-bound human chromatin interactome.

Genomes are organized into high-level three-dimensional structures, and DNA elements separated by long genomic distances can in principle interact functionally. Many transcription factors bind to regulatory DNA elements distant from gene promoters. Although distal binding sites have been shown to regulate transcription by long-range chromatin interactions at a few loci, chromatin interactions and their impact on transcription regulation have not been investigated in a genome-wide manner. Here we describe the development of a new strategy, chromatin interaction analysis by paired-end tag sequencing (ChIA-PET) for the de novo detection of global chromatin interactions, with which we have comprehensively mapped the chromatin interaction network bound by oestrogen receptor alpha (ER-alpha) in the human genome. We found that most high-confidence remote ER-alpha-binding sites are anchored at gene promoters through long-range chromatin interactions, suggesting that ER-alpha functions by extensive chromatin looping to bring genes together for coordinated transcriptional regulation. We propose that chromatin interactions constitute a primary mechanism for regulating transcription in mammalian genomes.

Integrative model of genomic factors for determining binding site selection by estrogen receptor-α.

A major question in transcription factor (TF) biology is why a TF binds to only a small fraction of motif eligible binding sites in the genome. Using the estrogen receptor-α as a model system, we sought to explicitly define parameters that determine TF-binding site selection. By examining 12 genetic and epigenetic parameters, we find that an energetically favorable estrogen response element (ERE) motif sequence, co-occupancy by the TF FOXA1, the presence of the H3K4me1 mark and an open chromatin configuration in the pre-ligand state provide specificity for ER binding. These factors can model estrogen-induced ER binding with high accuracy (ROC-AUC=0.95 and 0.88 using different genomic backgrounds). Moreover, when assessed in another estrogen-responsive cell line, this model was highly predictive for ERα binding (ROC-AUC=0.86). Variance in binding site selection between MCF-7 and T47D resides in sites with suboptimal ERE motifs, but modulated by the chromatin configuration. These results suggest a definable interplay between sequence motifs and local chromatin in selecting TF binding.

Gold-nanoparticle-based assay for instantaneous detection of nuclear hormone receptor-response elements interactions.

Gold nanoparticles (AuNPs) are widely used as colorimetric probes for biosensing, relying on their unique particle size-dependent and/or interparticle distance-dependent extinction spectrum and solution color. Herein, we describe an AuNP-based colorimetric assay to detect binding interactions between nuclear hormone receptors and their corresponding DNA-binding elements, particularly the human estrogen receptors (ERalpha and ERbeta) and their cognate estrogen response elements (EREs). We found that the protein-DNA (ER-ERE) complexes can stabilize citrate anion-capped AuNPs against salt-induced aggregation to a larger extent than the protein (ER) or the DNA (ERE) alone, due to their unique molecular size and charge properties that provide a strong electrosteric protection. Moreover, our results show that the extent of stabilization is sequence-dependent and can distinguish a single base variation in the ERE associated with minor changes in protein-DNA binding affinity. With this assay, many important parameters of protein-DNA binding events (e.g., sequence selectivity, distinct DNA binding properties of protein subtypes, binding stoichiometry, and sequence-independent transient binding) can be determined instantly without using labels, tedious sample preparations, and sophisticated instrumentation. These benefits, in particular the high-throughput potential, could enable this assay to become the assay of choice to complement conventional techniques for large scale characterization of protein-DNA interactions, a key aspect in biological research.

Whole-genome cartography of estrogen receptor alpha binding sites.

Using a chromatin immunoprecipitation-paired end diTag cloning and sequencing strategy, we mapped estrogen receptor alpha (ERalpha) binding sites in MCF-7 breast cancer cells. We identified 1,234 high confidence binding clusters of which 94% are projected to be bona fide ERalpha binding regions. Only 5% of the mapped estrogen receptor binding sites are located within 5 kb upstream of the transcriptional start sites of adjacent genes, regions containing the proximal promoters, whereas vast majority of the sites are mapped to intronic or distal locations (>5 kb from 5' and 3' ends of adjacent transcript), suggesting transcriptional regulatory mechanisms over significant physical distances. Of all the identified sites, 71% harbored putative full estrogen response elements (EREs), 25% bore ERE half sites, and only 4% had no recognizable ERE sequences. Genes in the vicinity of ERalpha binding sites were enriched for regulation by estradiol in MCF-7 cells, and their expression profiles in patient samples segregate ERalpha-positive from ERalpha-negative breast tumors. The expression dynamics of the genes adjacent to ERalpha binding sites suggest a direct induction of gene expression through binding to ERE-like sequences, whereas transcriptional repression by ERalpha appears to be through indirect mechanisms. Our analysis also indicates a number of candidate transcription factor binding sites adjacent to occupied EREs at frequencies much greater than by chance, including the previously reported FOXA1 sites, and demonstrate the potential involvement of one such putative adjacent factor, Sp1, in the global regulation of ERalpha target genes. Unexpectedly, we found that only 22%-24% of the bona fide human ERalpha binding sites were overlapping conserved regions in whole genome vertebrate alignments, which suggest limited conservation of functional binding sites. Taken together, this genome-scale analysis suggests complex but definable rules governing ERalpha binding and gene regulation.

Surface plasmon resonance study of cooperative interactions of estrogen receptor alpha and transcriptional factor Sp1 with composite DNA elements.

We have applied surface plasmon resonance (SPR) spectroscopy to study the cooperative interactions of estrogen receptor alpha (ERalpha) and transcription factor Sp1 with a composite DNA element, containing an estrogen response element (ERE) half-site upstream of two adjacent Sp1 sites (+571 ERE/Sp1 composite site in promoter A of the human PR gene). Using nuclear extracts of MCF-7 breast cancer cells as sample, we have shown that Sp1 is associated with Sp1-binding sites only, whereas ERalpha can be recruited to DNA both through direct binding to the ERE half-site and/or through protein-protein interactions with DNA-bound Sp1. The ERE half-site and the proximal Sp1 site are only 4 bp apart, and our data suggests that one transcription factor bound to DNA constitutes a sterical hindrance of the accessibility of the binding site for the other transcription factor. Our data confirms previous observations that ERalpha increases the amount of Sp1 recruited to the composite binding site in a dose-dependent manner. Using recombinant proteins, we have unambiguously proved the formation of a ternary complex of ERalpha/Sp1-composite DNA, for which previously published electrophoretic mobility shift assay (EMSA) results are contradictive. With this study, we have demonstrated that the solid-liquid-phase SPR assay is a powerful alternative for studying multiprotein-DNA interactions and is superior to the EMSA experiments as it is capable of real-time measurements, can quantify the amount of protein bound, and can capture transient and weak binding interactions. The comprehensive characterization of the synergistic interactions between ERalpha-DNA, Sp1-DNA, and ERalpha-Sp1 contributes to the understanding of how ERalpha and Sp1 influence and activate gene transcription.

A two-step antibody strategy for surface plasmon resonance spectroscopy detection of protein-DNA interactions in nuclear extracts.

Surface plasmon resonance (SPR) spectroscopy has emerged as a powerful alternative to conventional biochemistry methods for studying protein-DNA interactions that involve recombinant proteins of known identity. There are, however, limited demonstrations of SPR detection of protein-DNA bindings in crude samples, e.g., cell extracts, where the challenge is to detect and identify specific DNA binding protein(s) among other protein components in a physiological setting. We have developed a two-step antibody approach for an SPR study of estrogen receptor alpha (ERalpha)-DNA interactions, in which nuclear extracts prepared from MCF-7 breast cancer cells were used as the source of ERalpha protein. Following the binding of nuclear extracts to surface-immobilized estrogen response elements, rabbit anti-ERalpha antibody followed by a secondary antibody (goat anti-rabbit IgG) were applied to recognize the bound ERalpha and amplify the signals, respectively. Through a series of experiments, we have demonstrated that the magnitude of the binding signals from the secondary antibody reflects the affinity by which ERalpha binds to different DNA sequences. The detection sensitivity is determined by the amount of nuclear extracts and the concentration of primary antibody used. The sequence specificity of the nuclear ERalpha measured using the two-step antibody approach is in agreement with that measured for recombinant ERalpha protein (using receptor binding signals).

Inhibitory effects of estrogen receptor beta on specific hormone-responsive gene expression and association with disease outcome in primary breast cancer.

INTRODUCTION: The impact of interactions between the two estrogen receptor (ER) subtypes, ERalpha and ERbeta, on gene expression in breast cancer biology is not clear. The goal of this study was to examine transcriptomic alterations in cancer cells co-expressing both receptors and the association of gene expression signatures with disease outcome. METHODS: Transcriptional effects of ERbeta overexpression were determined in a stably transfected cell line derived from ERalpha-positive T-47D cells. Microarray analysis was carried out to identify differential gene expression in the cell line, and expression of key genes was validated by quantitative polymerase chain reaction. Microarray and clinical data from patient samples were then assessed to determine the in vivo relevance of the expression profiles observed in the cell line. RESULTS: A subset of 14 DNA replication and cell cycle-related genes was found to be specifically downregulated by ERbeta. Expression profiles of four genes, CDC2, CDC6, CKS2, and DNA2L, were significantly inversely correlated with ERbeta transcript levels in patient samples, consistent with in vitro observations. Kaplan-Meier analysis revealed better disease outcome for the patient group with an expression signature linked to higher ERbeta expression as compared to the lower ERbeta-expressing group for both disease-free survival (p = 0.00165) and disease-specific survival (p = 0.0268). These findings were further validated in an independent cohort. CONCLUSION: Our findings revealed a transcriptionally regulated mechanism for the previously described growth inhibitory effects of ERbeta in ERalpha-positive breast tumor cells and provide evidence for a functional and beneficial impact of ERbeta in primary breast tumors.

Molecular conservation of estrogen-response associated with cell cycle regulation, hormonal carcinogenesis and cancer in zebrafish and human cancer cell lines.

BACKGROUND: The zebrafish is recognized as a versatile cancer and drug screening model. However, it is not known whether the estrogen-responsive genes and signaling pathways that are involved in estrogen-dependent carcinogenesis and human cancer are operating in zebrafish. In order to determine the potential of zebrafish model for estrogen-related cancer research, we investigated the molecular conservation of estrogen responses operating in both zebrafish and human cancer cell lines. METHODS: Microarray experiment was performed on zebrafish exposed to estrogen (17ß-estradiol; a classified carcinogen) and an anti-estrogen (ICI 182,780). Zebrafish estrogen-responsive genes sensitive to both estrogen and anti-estrogen were identified and validated using real-time PCR. Human homolog mapping and knowledge-based data mining were performed on zebrafish estrogen responsive genes followed by estrogen receptor binding site analysis and comparative transcriptome analysis with estrogen-responsive human cancer cell lines (MCF7, T47D and Ishikawa). RESULTS: Our transcriptome analysis captured multiple estrogen-responsive genes and signaling pathways that increased cell proliferation, promoted DNA damage and genome instability, and decreased tumor suppressing effects, suggesting a common mechanism for estrogen-induced carcinogenesis. Comparative analysis revealed a core set of conserved estrogen-responsive genes that demonstrate enrichment of estrogen receptor binding sites and cell cycle signaling pathways. Knowledge-based and network analysis led us to propose that the mechanism involving estrogen-activated estrogen receptor mediated down-regulation of human homolog HES1 followed by up-regulation cell cycle-related genes (human homologs E2F4, CDK2, CCNA, CCNB, CCNE), is highly conserved, and this mechanism may involve novel crosstalk with basal AHR. We also identified mitotic roles of polo-like kinase as a conserved signaling pathway with multiple entry points for estrogen regulation. CONCLUSION: The findings demonstrate the use of zebrafish for characterizing estrogen-like environmental carcinogens and anti-estrogen drug screening. From an evolutionary perspective, our findings suggest that estrogen regulation of cell cycle is perhaps one of the earliest forms of steroidal-receptor controlled cellular processes. Our study provides first evidence of molecular conservation of estrogen-responsiveness between zebrafish and human cancer cell lines, hence demonstrating the potential of zebrafish for estrogen-related cancer research.

Highly sensitive and reversible silicon nanowire biosensor to study nuclear hormone receptor protein and response element DNA interactions.

To thoroughly understand the role that estrogen receptors partake in regulation of gene expression, characterization of estrogen receptors (ERs) and estrogen-response elements (EREs) interactions is essential. In the work, we present a highly sensitive and reusable silicon nanowire (SiNW) biosensor to study the interactions between human ER proteins (ER, α and ß subtypes) and EREs (dsDNA). The proteins were covalently immobilized on the SiNW surface. Various EREs including wild-type, mutant and scrambled DNA sequences were then applied to the protein-functionalized SiNW surface. Due to negatively charged dsDNA, binding of the EREs to the ERs on the n-type SiNW biosensor leads to the accumulation of negative charges on the surface, thereby inducing increase in resistance. The results show that the specificity of the ERE-ERα binding is higher than that of the ERE-ERß binding, what is more, the mutant ERE reduces the binding affinity for both ERα and ERß. By applying various concentrations of wild-type ERE to the bound ERα, a very low concentration of 10 fM wild-type ERE was found to be able to bind to the ERα. The reversible association and dissociation between ERα and wt-ERE was achieved, pointing to a reusable biosensor for protein-DNA binding. Through the study, we have established the SiNW biosensor as a promising method in providing comprehensive study for hormone receptor-response element interactions.

De-novo identification of PPARgamma/RXR binding sites and direct targets during adipogenesis.

BACKGROUND: The pathophysiology of obesity and type 2 diabetes mellitus is associated with abnormalities in endocrine signaling in adipose tissue and one of the key signaling affectors operative in these disorders is the nuclear hormone transcription factor peroxisome proliferator-activated receptor-gamma (PPARgamma). PPARgamma has pleiotropic functions affecting a wide range of fundamental biological processes including the regulation of genes that modulate insulin sensitivity, adipocyte differentiation, inflammation and atherosclerosis. To date, only a limited number of direct targets for PPARgamma have been identified through research using the well established pre-adipogenic cell line, 3T3-L1. In order to obtain a genome-wide view of PPARgamma binding sites, we applied the pair end-tagging technology (ChIP-PET) to map PPARgamma binding sites in 3T3-L1 preadipocyte cells. METHODOLOGY/PRINCIPAL FINDINGS: Coupling gene expression profile analysis with ChIP-PET, we identified in a genome-wide manner over 7700 DNA binding sites of the transcription factor PPARgamma and its heterodimeric partner RXR during the course of adipocyte differentiation. Our validation studies prove that the identified sites are bona fide binding sites for both PPARgamma and RXR and that they are functionally capable of driving PPARgamma specific transcription. Our results strongly indicate that PPARgamma is the predominant heterodimerization partner for RXR during late stages of adipocyte differentiation. Additionally, we find that PPARgamma/RXR association is enriched within the proximity of the 5' region of the transcription start site and this association is significantly associated with transcriptional up-regulation of genes involved in fatty acid and lipid metabolism confirming the role of PPARgamma as the master transcriptional regulator of adipogenesis. Evolutionary conservation analysis of these binding sites is greater when adjacent to up-regulated genes than down-regulated genes, suggesting the primordial function of PPARgamma/RXR is in the induction of genes. Our functional validations resulted in identifying novel PPARgamma direct targets that have not been previously reported to promote adipogenic differentiation. CONCLUSIONS/SIGNIFICANCE: We have identified in a genome-wide manner the binding sites of PPARgamma and RXR during the course of adipogenic differentiation in 3T3L1 cells, and provide an important resource for the study of PPARgamma function in the context of adipocyte differentiation.

Characterization of protein--DNA interactions using surface plasmon resonance spectroscopy with various assay schemes.

Specific protein-DNA interactions play a central role in transcription and other biological processes. A comprehensive characterization of protein-DNA interactions should include information about binding affinity, kinetics, sequence specificity, and binding stoichiometry. In this study, we have used surface plasmon resonance spectroscopy (SPR) to study the interactions between human estrogen receptors (ER, alpha and beta subtypes) and estrogen response elements (ERE), with four assay schemes. First, we determined the sequence-dependent receptors' binding capacity by monitoring the binding of ER to various ERE sequences immobilized on a sensor surface (assay format denoted as the direct assay). Second, we screened the relative affinity of ER for various ERE sequences using a competition assay, in which the receptors bind to an ERE-immobilized surface in the presence of competitor ERE sequences. Third, we monitored the assembly of ER-ERE complexes on a SPR surface and thereafter the removal and/or dissociation of the ER (assay scheme denoted as the dissociation assay) to determine the binding stoichiometry. Last, a sandwich assay (ER binding to ERE followed by anti-ER recognition of a specific ER subtype) was performed in an effort to understand how ERalpha and ERbeta may associate and compete when binding to the DNA. With these assay schemes, we reaffirmed that (1) ERalpha is more sensitive than ERbeta to base pair change(s) in the consensus ERE, (2) ERalpha and ERbeta form a heterodimer when they bind to the consensus ERE, and (3) the binding stoichiometry of both ERalpha- and ERbeta-ERE complexes is dependent on salt concentration. With this study, we demonstrate the versatility of the SPR analysis. With the involvement of various assay arrangements, the SPR analysis can be further extended to more than kinetics and affinity study.

Understanding ligand binding effects on the conformation of estrogen receptor alpha-DNA complexes: a combinational quartz crystal microbalance with dissipation and surface plasmon resonance study.

Estrogen receptors are ligand-activated transcription factors that regulate gene expression by binding to specific DNA sequences. To date, the effect of ligands on the conformation of estrogen receptor alpha (ERalpha)-DNA complex remains a poorly understood issue. In our study, we are introducing the quartz crystal microbalance with dissipation monitoring (QCM-D) as a new alternative to study the conformational differences in protein-DNA complexes. Specifically, we have used QCM-D, in combination with surface plasmon resonance (SPR) spectroscopy, to monitor the binding of ERalpha to a specific DNA (estrogen response element, ERE) and a nonspecific DNA in the presence of either the agonist ligand, 17b-estradiol, the partial antagonist ligand, 4-hydroxytamoxifen, or vehicle alone. Both with presence and absence of ligand, the specific ERalpha-ERE complexes are observed to adopt a more compact conformation compared to nonspecific complexes. This observation is well correlated to the biophysical changes occurring during protein-DNA interaction shown by past structural and mechanism studies. Notably, pretreatment of ERalpha with E2 and 4OHT affects not only the viscoelasticity and conformation of the protein-DNA complex but also ERalpha binding capacity to immobilized ERE. These results affirm that ligands have remarkable effects on ERalpha-DNA complexes. Understanding these effects will provide insight into how ligand binding promotes subsequent events required for gene transcription.

Combinational application of surface plasmon resonance spectroscopy and quartz crystal microbalance for studying nuclear hormone receptor-response element interactions.

Conventional methodologies for studying protein-DNA complexes, such as electrophoretic mobility shift assays (EMSAs), lack the real-time sensitivity and precision to accurately characterize the complex dynamics of interactions between transcription factors and their binding sites. To better understand the interactions between estrogen receptor (ER) subtypes and the estrogen response elements (EREs), we employed surface plasmon resonance (SPR) spectroscopy and quartz crystal microbalance with dissipation measurement (QCM-D) and made the following observations: (1) base substitutions in ERE half-sites reduced binding affinity for both ERalpha and ERbeta, (2) ERalpha has a higher sequence specificity than ERbeta or there were more nonspecific interactions between ERbeta and control DNA, and (3) ERalpha bound ERE as dimers and ERbeta bound as tetramers. These findings highlight intrinsic differences in DNA-binding properties between receptor subtypes, which are not apparent based on the high degree of conservation (96% identity) in their DNA-binding domains and results from EMSA studies. With this study, we demonstrate the potential of utilizing SPR and QCM in combination for a comprehensive characterization of ER-DNA interactions, including sequence-dependent binding mechanisms and structural differences in ERalpha-DNA and ERbeta-DNA complexes.

Multiplatform genome-wide identification and modeling of functional human estrogen receptor binding sites.

BACKGROUND: Transcription factor binding sites (TFBS) impart specificity to cellular transcriptional responses and have largely been defined by consensus motifs derived from a handful of validated sites. The low specificity of the computational predictions of TFBSs has been attributed to ubiquity of the motifs and the relaxed sequence requirements for binding. We posited that the inadequacy is due to limited input of empirically verified sites, and demonstrated a multiplatform approach to constructing a robust model. RESULTS: Using the TFBS for the estrogen receptor (ER)alpha (estrogen response element [ERE]) as a model system, we extracted EREs from multiple molecular and genomic platforms whose binding to ERalpha has been experimentally confirmed or rejected. In silico analyses revealed significant sequence information flanking the standard binding consensus, discriminating ERE-like sequences that bind ERalpha from those that are nonbinders. We extended the ERE consensus by three bases, bearing a terminal G at the third position 3' and an initiator C at the third position 5', which were further validated using surface plasmon resonance spectroscopy. Our functional human ERE prediction algorithm (h-ERE) outperformed existing predictive algorithms and produced fewer than 5% false negatives upon experimental validation. CONCLUSION: Building upon a larger experimentally validated ERE set, the h-ERE algorithm is able to demarcate better the universe of ERE-like sequences that are potential ER binders. Only 14% of the predicted optimal binding sites were utilized under the experimental conditions employed, pointing to other selective criteria not related to EREs. Other factors, in addition to primary nucleotide sequence, will ultimately determine binding site selection.

The Ah receptor inhibits estrogen-induced estrogen receptor beta in breast cancer cells.

We have studied the effect of the aryl hydrocarbon receptor ligand 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on estrogen receptor (ER) beta gene expression in the human breast cancer cell line, T47D. TCDD inhibited 17beta-estradiol (E2)-induced up-regulation of both ER beta wild type and ER beta cx mRNA. Cycloheximide pre-treatment had no inhibitory effect, and the estimated half-life of ER beta mRNA of about 33 min was not changed by any hormone administration. Chromatin immunoprecipitation experiments showed recruitment of ER alpha to the ER beta promoter. Gel mobility shift experiments revealed an E2-induced protein binding to a half site estrogen response element in the ER beta promoter, and TCDD reduced that binding. These results show that ER alpha regulates the expression of its own heterodimerization partner, ER beta, in T47D cells. TCDD, an anti-estrogenic compound, inhibits ER alpha-mediated induction of ER beta mRNA. These findings add to our understanding of cross talk between dioxin and estrogen signaling in human cells.

The focal adhesion protein vinexin alpha regulates the phosphorylation and activity of estrogen receptor alpha.

Steroid receptors are transcription factors that regulate hormone-responsive genes and whose activity is controlled by their interaction with numerous other proteins. Observations reported here reveal that estrogen receptors alpha and beta (ERalpha and ERbeta), androgen receptor, and glucocorticoid receptor bind in vitro to vinexin alpha, a multiple SH3 motif-containing protein associated with the cytoskeleton. The SH3 domains are not involved in this interaction. Furthermore, we demonstrate that vinexin alpha stimulates the ligand-induced transactivation function of these receptors, although it is devoid of intrinsic transcriptional activity when tethered to DNA. In addition, the ectopic coexpression of vinexin alpha and ERalpha results in a loss of ERalpha phosphorylation on serines and the partial redistribution of vinexin alpha into the nucleus, where it colocalizes with ERalpha. These results establish a new model of transcriptional regulation where components of the cell-cell and cell-substrate adhesion complexes can regulate the phosphorylation and activity of steroid receptors.

Discovery of estrogen receptor alpha target genes and response elements in breast tumor cells.

BACKGROUND: Estrogens and their receptors are important in human development, physiology and disease. In this study, we utilized an integrated genome-wide molecular and computational approach to characterize the interaction between the activated estrogen receptor (ER) and the regulatory elements of candidate target genes. RESULTS: Of around 19,000 genes surveyed in this study, we observed 137 ER-regulated genes in T-47D cells, of which only 89 were direct target genes. Meta-analysis of heterogeneous in vitro and in vivo datasets showed that the expression profiles in T-47D and MCF-7 cells are remarkably similar and overlap with genes differentially expressed between ER-positive and ER-negative tumors. Computational analysis revealed a significant enrichment of putative estrogen response elements (EREs) in the cis-regulatory regions of direct target genes. Chromatin immunoprecipitation confirmed ligand-dependent ER binding at the computationally predicted EREs in our highest ranked ER direct target genes, NRIP1, GREB1 and ABCA3. Wider examination of the cis-regulatory regions flanking the transcriptional start sites showed species conservation in mouse-human comparisons in only 6% of predicted EREs. CONCLUSIONS: Only a small core set of human genes, validated across experimental systems and closely associated with ER status in breast tumors, appear to be sufficient to induce ER effects in breast cancer cells. That cis-regulatory regions of these core ER target genes are poorly conserved suggests that different evolutionary mechanisms are operative at transcriptional control elements than at coding regions. These results predict that certain biological effects of estrogen signaling will differ between mouse and human to a larger extent than previously thought.