Machine-Learning Identifies a Strategy for Differentiation Therapy in Solid Tumors.

BACKGROUND: Although differentiation therapy can cure some hematologic malignancies, its curative potential remains unrealized in solid tumors. This is because conventional computational approaches succumb to the thunderous noise of inter-/intratumoral heterogeneity. Using colorectal cancers (CRCs) as an example, here we outline a machine learning(ML)-based approach to track, differentiate, and selectively target cancer stem cells (CSCs). METHODS: A transcriptomic network was built and validated using healthy colon and CRC tissues in diverse gene expression datasets (~5,000 human and >300 mouse samples). Therapeutic targets and perturbation strategies were prioritized using ML, with the goal of reinstating the expression of a transcriptional identifier of the differentiated colonocyte, CDX2, whose loss in poorly differentiated (CSC-enriched) CRCs doubles the risk of relapse/death. The top candidate target was then engaged with a clinical-grade drug and tested on 3 models: CRC lines in vitro, xenografts in mice, and in a prospective cohort of healthy (n = 3) and CRC (n = 23) patient-derived organoids (PDOs). RESULTS: The drug shifts the network predictably, induces CDX2 and crypt differentiation, and shows cytotoxicity in all 3 models, with a high degree of selectivity towards all CDX2-negative cell lines, xenotransplants, and PDOs. The potential for effective pairing of therapeutic efficacy (IC50) and biomarker (CDX2-low state) is confirmed in PDOs using multivariate analyses. A 50-gene signature of therapeutic response is derived and tested on 9 independent cohorts (~1700 CRCs), revealing the impact of CDX2-reinstatement therapy could translate into a ~50% reduction in the risk of mortality/recurrence. CONCLUSIONS: Findings not only validate the precision of the ML approach in targeting CSCs, and objectively assess its impact on clinical outcome, but also exemplify the use of ML in yielding clinical directive information for enhancing personalized medicine.

COVID-19 lung disease shares driver AT2 cytopathic features with Idiopathic pulmonary fibrosis.

BACKGROUND: In the aftermath of Covid-19, some patients develop a fibrotic lung disease, i.e., post-COVID-19 lung disease (PCLD), for which we currently lack insights into pathogenesis, disease models, or treatment options. METHODS: Using an AI-guided approach, we analyzed > 1000 human lung transcriptomic datasets associated with various lung conditions using two viral pandemic signatures (ViP and sViP) and one covid lung-derived signature. Upon identifying similarities between COVID-19 and idiopathic pulmonary fibrosis (IPF), we subsequently dissected the basis for such similarity from molecular, cytopathic, and immunologic perspectives using a panel of IPF-specific gene signatures, alongside signatures of alveolar type II (AT2) cytopathies and of prognostic monocyte-driven processes that are known drivers of IPF. Transcriptome-derived findings were used to construct protein-protein interaction (PPI) network to identify the major triggers of AT2 dysfunction. Key findings were validated in hamster and human adult lung organoid (ALO) pre-clinical models of COVID-19 using immunohistochemistry and qPCR. FINDINGS: COVID-19 resembles IPF at a fundamental level; it recapitulates the gene expression patterns (ViP and IPF signatures), cytokine storm (IL15-centric), and the AT2 cytopathic changes, e.g., injury, DNA damage, arrest in a transient, damage-induced progenitor state, and senescence-associated secretory phenotype (SASP). These immunocytopathic features were induced in pre-clinical COVID models (ALO and hamster) and reversed with effective anti-CoV-2 therapeutics in hamsters. PPI-network analyses pinpointed ER stress as one of the shared early triggers of both diseases, and IHC studies validated the same in the lungs of deceased subjects with COVID-19 and SARS-CoV-2-challenged hamster lungs. Lungs from tg-mice, in which ER stress is induced specifically in the AT2 cells, faithfully recapitulate the host immune response and alveolar cytopathic changes that are induced by SARS-CoV-2. INTERPRETATION: Like IPF, COVID-19 may be driven by injury-induced ER stress that culminates into progenitor state arrest and SASP in AT2 cells. The ViP signatures in monocytes may be key determinants of prognosis. The insights, signatures, disease models identified here are likely to spur the development of therapies for patients with IPF and other fibrotic interstitial lung diseases. FUNDING: This work was supported by the National Institutes for Health grants R01- GM138385 and AI155696 and funding from the Tobacco-Related disease Research Program (R01RG3780).

COVID-19 lung disease shares driver AT2 cytopathic features with Idiopathic pulmonary fibrosis.

Background: In the aftermath of Covid-19, some patients develop a fibrotic lung disease, i.e., p ost- C OVID-19 l ung d isease (PCLD), for which we currently lack insights into pathogenesis, disease models, or treatment options. Method: Using an AI-guided approach, we analyzed > 1000 human lung transcriptomic datasets associated with various lung conditions using two viral pandemic signatures (ViP and sViP) and one covid lung-derived signature. Upon identifying similarities between COVID-19 and idiopathic pulmonary fibrosis (IPF), we subsequently dissected the basis for such similarity from molecular, cytopathic, and immunologic perspectives using a panel of IPF-specific gene signatures, alongside signatures of alveolar type II (AT2) cytopathies and of prognostic monocyte-driven processes that are known drivers of IPF. Transcriptome-derived findings were used to construct protein-protein interaction (PPI) network to identify the major triggers of AT2 dysfunction. Key findings were validated in hamster and human adult lung organoid (ALO) pre-clinical models of COVID-19 using immunohistochemistry and qPCR. Findings: COVID-19 resembles IPF at a fundamental level; it recapitulates the gene expression patterns (ViP and IPF signatures), cytokine storm (IL15-centric), and the AT2 cytopathic changes, e.g., injury, DNA damage, arrest in a transient, damage-induced progenitor state, and senescence-associated secretory phenotype (SASP). These immunocytopathic features were induced in pre-clinical COVID models (ALO and hamster) and reversed with effective anti-CoV-2 therapeutics in hamsters. PPI-network analyses pinpointed ER stress as one of the shared early triggers of both diseases, and IHC studies validated the same in the lungs of deceased subjects with COVID-19 and SARS-CoV-2-challenged hamster lungs. Lungs from tg - mice, in which ER stress is induced specifically in the AT2 cells, faithfully recapitulate the host immune response and alveolar cytopathic changes that are induced by SARS-CoV-2. Interpretation: Like IPF, COVID-19 may be driven by injury-induced ER stress that culminates into progenitor state arrest and SASP in AT2 cells. The ViP signatures in monocytes may be key determinants of prognosis. The insights, signatures, disease models identified here are likely to spur the development of therapies for patients with IPF and other fibrotic interstitial lung diseases. Funding: This work was supported by the National Institutes for Health grants R01-GM138385 and AI155696 and funding from the Tobacco-Related disease Research Program (R01RG3780). One Sentence Summary: Severe COVID-19 triggers cellular processes seen in fibrosing Interstitial Lung Disease. RESEARCH IN CONTEXT: Evidence before this study: In its aftermath, the COVID-19 pandemic has left many survivors, almost a third of those who recovered, with a mysterious long-haul form of the disease which culminates in a fibrotic form of interstitial lung disease (post-COVID-19 ILD). Post-COVID-19 ILD remains a largely unknown entity. Currently, we lack insights into the core cytopathic features that drive this condition.Added value of this study: Using an AI-guided approach, which involves the use of sets of gene signatures, protein-protein network analysis, and a hamster model of COVID-19, we have revealed here that COVID-19 -lung fibrosis resembles IPF, the most common form of ILD, at a fundamental levelâ€"showing similar gene expression patterns in the lungs and blood, and dysfunctional AT2 processes (ER stress, telomere instability, progenitor cell arrest, and senescence). These findings are insightful because AT2 cells are known to contain an elegant quality control network to respond to intrinsic or extrinsic stress; a failure of such quality control results in diverse cellular phenotypes, of which ER stress appears to be a point of convergence, which appears to be sufficient to drive downstream fibrotic remodeling in the lung.Implications of all the available evidence: Because unbiased computational methods identified the shared fundamental aspects of gene expression and cellular processes between COVID-19 and IPF, the impact of our findings is likely to go beyond COVID-19 or any viral pandemic. The insights, tools (disease models, gene signatures, and biomarkers), and mechanisms identified here are likely to spur the development of therapies for patients with IPF and, other fibrotic interstitial lung diseases, all of whom have limited or no treatment options. To dissect the validated prognostic biomarkers to assess and track the risk of pulmonary fibrosis and develop therapeutics to halt fibrogenic progression.

GIV/Girdin, a non-receptor modulator for Gαi/s, regulates spatiotemporal signaling during sperm capacitation and is required for male fertility.

For a sperm to successfully fertilize an egg, it must first undergo capacitation in the female reproductive tract and later undergo acrosomal reaction (AR) upon encountering an egg surrounded by its vestment. How premature AR is avoided despite rapid surges in signaling cascades during capacitation remains unknown. Using a combination of conditional knockout (cKO) mice and cell-penetrating peptides, we show that GIV (CCDC88A), a guanine nucleotide-exchange modulator (GEM) for trimeric GTPases, is highly expressed in spermatocytes and is required for male fertility. GIV is rapidly phosphoregulated on key tyrosine and serine residues in human and murine spermatozoa. These phosphomodifications enable GIV-GEM to orchestrate two distinct compartmentalized signaling programs in the sperm tail and head; in the tail, GIV enhances PI3K→Akt signals, sperm motility and survival, whereas in the head it inhibits cAMP surge and premature AR. Furthermore, GIV transcripts are downregulated in the testis and semen of infertile men. These findings exemplify the spatiotemporally segregated signaling programs that support sperm capacitation and shed light on a hitherto unforeseen cause of infertility in men.

Loss of Thy-1 may reduce lung regeneration after pneumonectomy in mice.

BACKGROUND: Lung regeneration plays an important role in lung repair after injury. It is reliant upon proliferation of multiple cell types in the lung, including endothelium, epithelium, and fibroblasts, as well as remodeling of the extracellular matrix. METHODS: Lung regeneration following injury progresses via an initial inflammatory response during which macrophages clear the tissue of cellular debris. This process continues through cellular proliferation when existing cells and progenitors act to repopulate cells lost during injury, followed by tissue maturation in which newly formed cells achieve a differentiated phenotype. RESULTS: Signaling pathways critical for lung regeneration include FGF, EGF, WNT, and NOTCH. In addition, HDACs, miRNAs, ELASTIN, and MMP14 have been shown to regulate lung regeneration. Partial pneumonectomy (PNX) has been used as a therapeutic and investigational tool for several decades. Following PNX the remaining lung increases in size to compensate for loss of volume and respiratory capacity. CONCLUSIONS: Much has been learned about the triggers and mechanisms regulating pulmonary regeneration. However, the role of thymocyte differentiation antigen-1 (Thy-1) in post-PNX lung growth remains incompletely characterized. Thy-1 is a phosphatidylinositol glycoprotein with a relative molecular weight of 25000~37000 Da, which is expressed in almost all types of fibroblasts and regulates many biological functions. It not only supports the structure of fibroblasts, but also can balance cell proliferation, migration and regulate the synthesis of immune inflammatory mediators.

Soluble Thy-1 reverses lung fibrosis via its integrin-binding motif.

Loss of Thy-1 expression in fibroblasts correlates with lung fibrogenesis; however, the clinical relevance of therapeutic targeting of myofibroblasts via Thy-1-associated pathways remains to be explored. Using single (self-resolving) or repetitive (nonresolving) intratracheal administration of bleomycin in type 1 collagen-GFP reporter mice, we report that Thy-1 surface expression, but not mRNA, is reversibly diminished in activated fibroblasts and myofibroblasts in self-resolving fibrosis. However, Thy-1 mRNA expression is silenced in lung with nonresolving fibrosis following repetitive bleomycin administration, associated with persistent activation of αv integrin. Thy1-null mice showed progressive αv integrin activation and myofibroblast accumulation after a single dose of bleomycin. In vitro, targeting of αv integrin by soluble Thy-1-Fc (sThy-1), but not RLE-mutated Thy-1 or IgG, reversed TGF-ß1-induced myofibroblast differentiation in a dose-dependent manner, suggesting that Thy-1's integrin-binding RGD motif is required for the reversibility of myofibroblast differentiation. In vivo, treatment of established fibrosis induced either by single-dose bleomycin in WT mice or by induction of active TGF-ß1 by doxycycline in Cc10-rtTA-tTS-Tgfb1 mice with sThy-1 (1000 ng/kg, i.v.) promoted resolution of fibrosis. Collectively, these findings demonstrate that sThy-1 therapeutically inhibits the αv integrin-driven feedback loop that amplifies and sustains fibrosis.

Transforming growth factor beta 1 induces methylation changes in lung fibroblasts.

Idiopathic pulmonary fibrosis is a complex disease of unknown etiology. Environmental factors can affect disease susceptibility via epigenetic effects. Few studies explore global DNA methylation in lung fibroblasts, but none have focused on transforming growth factor beta-1 (TGF-ß1) as a potential modifier of the DNA methylome. Here we analyzed changes in methylation and gene transcription in normal and IPF fibroblasts following TGF-ß1 treatment. We analyzed the effects of TGF-ß1 on primary fibroblasts derived from normal or IPF lungs treated for 24 hours and 5 days using the Illumina 450k Human Methylation array and the Prime View Human Gene Expression Array. TGF-ß1 induced an increased number of gene expression changes after short term treatment in normal fibroblasts, whereas greater methylation changes were observed following long term stimulation mainly in IPF fibroblasts. DNA methyltransferase 3 alpha (DMNT3a) and tet methylcytosine dioxygenase 3 (TET3) were upregulated after 5-days TGF-ß1 treatment in both cell types, whereas DNMT3a was upregulated after 24h only in IPF fibroblasts. Our findings demonstrate that TGF-ß1 induced the upregulation of DNMT3a and TET3 expression and profound changes in the DNA methylation pattern of fibroblasts, mainly in those derived from IPF lungs.

Integrating multiomics longitudinal data to reconstruct networks underlying lung development.

A comprehensive understanding of the dynamic regulatory networks that govern postnatal alveolar lung development is still lacking. To construct such a model, we profiled mRNA, microRNA, DNA methylation, and proteomics of developing murine alveoli isolated by laser capture microdissection at 14 predetermined time points. We developed a detailed comprehensive and interactive model that provides information about the major expression trajectories, the regulators of specific key events, and the impact of epigenetic changes. Intersecting the model with single-cell RNA-Seq data led to the identification of active pathways in multiple or individual cell types. We then constructed a similar model for human lung development by profiling time-series human omics data sets. Several key pathways and regulators are shared between the reconstructed models. We experimentally validated the activity of a number of predicted regulators, leading to new insights about the regulation of innate immunity during lung development.

Thy-1 dependent uptake of mesenchymal stem cell-derived extracellular vesicles blocks myofibroblastic differentiation.

Bone marrow-derived mesenchymal stem cells (MSC) have been promoted for multiple therapeutic applications. Many beneficial effects of MSCs are paracrine, dependent on extracellular vesicles (EVs). Although MSC-derived EVs (mEVs) are beneficial for acute lung injury and pulmonary fibrosis, mechanisms of mEV uptake by lung fibroblasts and their effects on myofibroblastic differentiation have not been established. We demonstrate that mEVs, but not fibroblast EVs (fEVs), suppress TGFß1-induced myofibroblastic differentiation of normal and idiopathic pulmonary fibrosis (IPF) lung fibroblasts. MEVs display increased time- and dose-dependent cellular uptake compared to fEVs. Removal or blocking of Thy-1, or blocking Thy-1-beta integrin interactions, decreased mEV uptake and prevented suppression of myofibroblastic differentiation. MicroRNAs (miRs) 199a/b-3p, 21-5p, 630, 22-3p, 196a-5p, 199b-5p, 34a-5p and 148a-3p are selectively packaged in mEVs. In silico analyses indicated that IPF lung fibroblasts have increased expression of genes that are targets of mEV-enriched miRs. MiR-630 mimics blocked TGFß1 induction of CDH2 in normal and IPF fibroblasts, and antagomiR-630 abrogated the effect of mEV on CDH2 expression. These data suggest that the interaction of Thy-1 with beta integrins mediates mEV uptake by lung fibroblasts, which blocks myofibroblastic differentiation, and that mEVs are enriched for miRs that target profibrotic genes up-regulated in IPF fibroblasts.

Thy-1 interaction with Fas in lipid rafts regulates fibroblast apoptosis and lung injury resolution.

Thy-1-negative lung fibroblasts are resistant to apoptosis. The mechanisms governing this process and its relevance to fibrotic remodeling remain poorly understood. By using either sorted or transfected lung fibroblasts, we found that Thy-1 expression is associated with downregulation of anti-apoptotic molecules Bcl-2 and Bcl-xL, as well as increased levels of cleaved caspase-9. Addition of rhFasL and staurosporine, well-known apoptosis inducers, caused significantly increased cleaved caspase-3, -8, and PARP in Thy-1-transfected cells. Furthermore, rhFasL induced Fas translocation into lipid rafts and its colocalization with Thy-1. These in vitro results indicate that Thy-1, in a manner dependent upon its glycophosphatidylinositol anchor and lipid raft localization, regulates apoptosis in lung fibroblasts via Fas-, Bcl-, and caspase-dependent pathways. In vivo, Thy-1 deficient (Thy1-/-) mice displayed persistence of myofibroblasts in the resolution phase of bleomycin-induced fibrosis, associated with accumulation of collagen and failure of lung fibrosis resolution. Apoptosis of myofibroblasts is decreased in Thy1-/- mice in the resolution phase. Collectively, these findings provide new evidence regarding the role and mechanisms of Thy-1 in initiating myofibroblast apoptosis that heralds the termination of the reparative response to bleomycin-induced lung injury. Understanding the mechanisms regulating fibroblast survival/apoptosis should lead to novel therapeutic interventions for lung fibrosis.

Selecting the most appropriate time points to profile in high-throughput studies.

Biological systems are increasingly being studied by high throughput profiling of molecular data over time. Determining the set of time points to sample in studies that profile several different types of molecular data is still challenging. Here we present the Time Point Selection (TPS) method that solves this combinatorial problem in a principled and practical way. TPS utilizes expression data from a small set of genes sampled at a high rate. As we show by applying TPS to study mouse lung development, the points selected by TPS can be used to reconstruct an accurate representation for the expression values of the non selected points. Further, even though the selection is only based on gene expression, these points are also appropriate for representing a much larger set of protein, miRNA and DNA methylation changes over time. TPS can thus serve as a key design strategy for high throughput time series experiments. Supporting Website: www.sb.cs.cmu.edu/TPS.

Chromatin accessibility and epigenetic modifications differ between frequently and infrequently rearranging VH genes.

The molecular mechanisms that control the temporal and lineage-specific accessibility, as well as the rearrangement frequency of V(H) genes for V(H)-to-DJ(H) recombination, are not fully understood. We previously found a positive correlation between the extent of histone acetylation and the differential rearrangement frequency of individual V(H) genes. Here, we demonstrated that poorly rearranging V(H) genes are more highly associated with histone H3 dimethylated at lysine 9, a marker of repressive chromatin, than frequently rearranging V(H) genes. We also observed a positive relationship between the differential binding of Pax5 to individual V(H)S107 genes and rearrangement frequency. Furthermore, we showed that accessibility of the regions flanking the Pax5 binding site and the recombination signal sequence (RSS) to restriction enzyme cleavage correspond with the differential rearrangement frequency of the V(H)S107 family members. In addition, we found that the CpG sites located in the coding regions of V(H) genes are methylated in general, while the extent of DNA methylation drops dramatically near the RSS. For the V(H)S107 family, one CpG site located 101bp upstream of the RSS showed variable methylation that correlates with rearrangement frequency, and the methylation status of a CpG site located 34bp downstream of the RSS could also favor the rearrangement of V1 over V11. These findings suggest that the extent of histone modifications, chromatin accessibility, DNA methylation, as well as the differential binding of Pax5 to V(H) coding regions, could all influence the rearrangement frequency of individual V(H) genes, although some of these mechanisms are not strictly B cell specific.

Transcription factor Pax5 (BSAP) transactivates the RAG-mediated V(H)-to-DJ(H) rearrangement of immunoglobulin genes.

Immunoglobulin rearrangement from variable heavy chain (V(H)) to diversity (D)-joining heavy chain (J(H)), which occurs exclusively in B lineage cells, is impaired in mice deficient for the B lineage-specific transcription factor Pax5. Conversely, ectopic Pax5 expression in thymocytes promotes the rearrangement of D(H)-proximal V(H)7183 genes. In exploring the mechanism for Pax5 regulation of V(H)-to-DJ(H) recombination, we have identified multiple Pax5 binding sites in the coding regions of human and mouse V(H) gene segments. Pax5 bound to those sites in vitro and occupied V(H) genes in early human and mouse B lineage cells. Moreover, Pax5 interacted with the recombination-activating gene 1 (RAG1)-RAG2 complex to enhance RAG-mediated V(H) recombination signal sequence cleavage and recombination of a V(H) gene substrate. These findings indicate a direct activating function for Pax5 in RAG-mediated immunoglobulin V(H)-to-DJ(H) recombination.

The extent of histone acetylation correlates with the differential rearrangement frequency of individual VH genes in pro-B cells.

During B lymphocyte development, Ig heavy and L chain genes are assembled by V(D)J recombination. Individual V, D, and J genes rearrange at very different frequencies in vivo, and the natural variation in recombination signal sequence does not account for all of these differences. Because a permissive chromatin structure is necessary for the accessibility of VH genes for VH to DJH recombination, we hypothesized that gene rearrangement frequency might be influenced by the extent of histone modifications. Indeed, we found in freshly isolated pro-B cells from muMT mice a positive correlation between the level of enrichment of VHS107 genes in the acetylated histone fractions as assayed by chromatin immunoprecipitation, and their relative rearrangement frequency in vivo. In the VH7183 family, the very frequently rearranging VH81X gene showed the highest association with acetylated histones, especially in the newborn. Together, our data show that the extent of histone modifications in pro-B cells should be considered as a mechanism by which accessibility and the rearrangement level of individual VH genes is regulated.

Many levels of control of V gene rearrangement frequency.

V, D, and J gene segments rearrange at very different frequencies. As with most biological systems, there are multiple levels of control of V gene recombination frequency, and here we review some of the work from our laboratory that addresses these various control mechanisms. One of the important factors that affect non-random V gene rearrangement frequency is the natural heterogeneity in recombination signal sequences (RSSs). Not only does variation in the heptamer and nonamer affect rearrangement, but variation in the spacer can also dramatically affect recombination. However, there are clearly other factors which control V gene rearrangement, as revealed by the fact that genes with identical RSSs can rearrange at different frequencies in vivo. Some of these other influences most likely affect the earliest stages of control--the change from an inaccessible state to an accessible state. Transcription factors can play a role in inducing these changes. Rearrangement of many VkappaI genes can be induced in a non-lymphoid cell line after ectopic expression of E2A, while neighboring VkappaII and VkappaIII genes do not rearrange, demonstrating that at least one level of control of induction of accessibility occurs at the level of the individual gene. Also, changes in chromatin structure can affect accessibility and might influence individual V gene rearrangement frequency.

Development of a microarray assay that measures hybridization stoichiometry in moles.

Microarray data is most useful when it can be compared with other genetic detection technologies. In this report, we designed a microarray assay format that transforms raw data into a defined scientific unit (i.e., moles) by measuring the amount of array feature present and the cDNA sequence hybridized. This study profiles a mouse reference universal RNA sample on a microarray consisting of PCR products. In measuring array features, a labeled DNA sequence was designed that hybridizes to a conserved sequence that is present in every array feature. To measure the amount of cDNA sample hybridized, the RNA sample was processed to ensure consistent dye to DNA ratio for every labeled target cDNA molecule, using labeled branched dendrimers rather than by incorporation. A dye printing assay was then performed in order to correlate molecules of cyanine dye to signal intensity. We demonstrate that by using this microarray assay design, raw data can be transformed into defined scientific units, which will facilitate interpretation of other experiments, such as data deposited at the Gene Expression Omnibus and ArrayExpress.

Characterization of a thyroid-specific and cyclic adenosine monophosphate-responsive enhancer far upstream from the human sodium iodide symporter gene.

We describe the cloning and characterization of a human sodium iodide (NIS) upstream enhancer (NUE). This putative enhancer was cloned based on its sequence homology (69% identity) to the rat NUE. A 296 base pair (bp) genomic DNA fragment, which is located 9000 bp upstream from the human hNIS gene, was amplified by polymerase chain reaction (PCR) and inserted into a luciferase reporter gene in front of both the homologous NIS promoter and the heterologous SV40 promoter. No enhancer activity could be found after transfection into HeLa cells, but in FRTL-5 cells representing the thyroid model, a threefold stimulation of the NIS promoter was found. This enhancer activity was present in both directions and was stimulated threefold by thyrotropin (TSH) and 14-fold by the cyclic adenosine monophosphate (cAMP) agonist forskolin. A small element (TGACGCA) in this enhancer was found to be of central importance, because its site-directed mutagenesis abolished the enhancer activity. This element bound specifically to proteins in nuclear extracts from FRTL-5 cells and to a lesser extent also from HeLa cells. In summary, we describe a thyroid-specific and cAMP-responsive enhancer far upstream from the human NIS gene, which is located in the intronic region of another gene coding for a ribosomal protein.