{
  "id": "transforming-growth-factor-beta",
  "title": "Transforming Growth Factor Beta",
  "lang": "en",
  "source_lang": "en",
  "summary": "Transforming growth factor beta is a family of signaling proteins that regulates development, immunity, tissue repair, and disease through context-dependent effects on cells.",
  "category": "science",
  "subcategory": "biology",
  "type": "concept",
  "aliases": [
    "TGF-β",
    "TGF-beta",
    "Transforming growth factor β"
  ],
  "generated_at": "2026-11-04",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/transforming-growth-factor-beta/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/transforming-growth-factor-beta/",
      "json": "https://aiwiki.page/zh/transforming-growth-factor-beta.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Transforming growth factor beta (TGF-β) is a group of closely related, secreted cytokines that regulate cell proliferation, differentiation, immune activity, and tissue organization. Mammals produce three principal isoforms—TGF-β1, TGF-β2, and TGF-β3—encoded by separate genes. These proteins belong to the much larger TGF-β superfamily, which also includes activins, bone morphogenetic proteins, and other developmental signals. In ordinary usage, “TGF-β” usually denotes the three closely related isoforms rather than the entire superfamily.\n\nTGF-β is distinguished by the separation between its production and its biological activation. Cells generally secrete it in a latent complex, and local mechanisms determine where and when the active signal becomes available. Once activated, TGF-β binds cell-surface receptors that transmit information through SMAD proteins and additional signaling pathways, altering gene expression and cellular behavior.\n\nIts effects depend strongly on cell identity, developmental stage, surrounding signals, and tissue condition. TGF-β can restrain epithelial growth, maintain immune tolerance, and coordinate wound repair, yet persistent or misdirected signaling can contribute to fibrosis and cancer progression. Understanding this combination of essential physiological functions and disease-promoting activities requires attention to the entire pathway, from extracellular storage to the interpretation of signals within individual cells."
    },
    {
      "heading": "Discovery, terminology, and molecular family",
      "text": "The name “transforming growth factor” originated in research on factors that made cultured fibroblasts acquire properties associated with cellular transformation, particularly growth without attachment to a solid surface. During the late 1970s and early 1980s, studies of tumor-cell extracts and conditioned culture media distinguished activities subsequently called TGF-α and TGF-β. Work by Anita Roberts, Michael Sporn, and their colleagues was central to the isolation and characterization of TGF-β. Its growth-promoting activity in early assays often depended on cooperation with other growth factors, especially epidermal growth factor.\n\nThe historical name can therefore be misleading. TGF-β does not simply transform normal cells into cancer cells, nor does it uniformly stimulate growth. Subsequent experiments showed that it inhibits proliferation in many epithelial and immune-cell populations. Its activity in soft-agar assays reflected a particular experimental setting rather than a universal physiological function. TGF-α, despite sharing the original naming scheme, is structurally unrelated to TGF-β and signals through the epidermal growth factor receptor.\n\nThe three mammalian isoforms are encoded by TGFB1, TGFB2, and TGFB3. Their mature signaling regions are highly similar, whereas their precursor regions and expression patterns differ more substantially. TGF-β1 is especially prominent in immune regulation and adult tissue responses; TGF-β2 and TGF-β3 have distinctive developmental functions as well as overlapping activities in mature tissues. These are differences in distribution and biological requirement, not exclusive assignments: several isoforms can act in the same organ, and all three use substantially overlapping receptor machinery.\n\nThe mature ligands are dimers composed of two approximately 12.5-kilodalton chains, giving a combined mass of about 25 kilodaltons. Each chain contains 112 amino acids and a characteristic arrangement of disulfide bonds that stabilizes a cystine-knot fold. Structural studies revealed an extended, finger-like architecture that provides distinct receptor-binding surfaces. The mature dimer is the receptor-binding signal; the much larger precursor-derived complexes regulate its storage, distribution, and activation.\n\nThe broader superfamily contains dozens of related ligands, but their shared ancestry does not imply identical signaling. Bone morphogenetic proteins generally activate SMAD1, SMAD5, and SMAD9, whereas TGF-β typically activates SMAD2 and SMAD3. Activins also signal through SMAD2 and SMAD3 but use different ligand-binding receptors. Distinguishing the three TGF-β isoforms from these relatives is essential when interpreting developmental experiments, receptor inhibitors, or descriptions of “TGF-β pathway” activity."
    },
    {
      "heading": "Biosynthesis, extracellular storage, and activation",
      "text": "TGF-β is synthesized as a precursor protein containing a signal peptide, a large amino-terminal propeptide, and the carboxy-terminal mature signaling region. The precursor enters the endoplasmic reticulum, where folding, dimer assembly, and disulfide-bond formation occur. During passage through the secretory pathway, furin or related processing enzymes cleave the precursor. Cleavage does not normally release an active ligand: the mature dimer remains noncovalently associated with the propeptide, known as latency-associated peptide, or LAP.\n\nLAP encloses receptor-binding surfaces and prevents the mature ligand from productively engaging its receptors. The LAP–ligand assembly is called the small latent complex. In many tissues, LAP is also disulfide-linked to a latent TGF-β-binding protein such as LTBP1, LTBP3, or LTBP4, forming a large latent complex. These binding proteins associate with components of the extracellular matrix, including fibronectin and fibrillin-containing structures. The matrix consequently serves not merely as a scaffold but as a spatially organized reservoir of latent signaling molecules.\n\nAn important activation mechanism depends on integrins, transmembrane adhesion receptors that connect extracellular ligands to the cellular cytoskeleton. The LAP regions of TGF-β1 and TGF-β3 contain an arginine–glycine–aspartate sequence recognized by several integrins. The epithelial integrin αvβ6 is a particularly well-characterized activator. When an integrin binds LAP and the latent complex is anchored to a mechanically resistant matrix, actomyosin-generated traction can distort LAP sufficiently to expose or release active TGF-β. Matrix attachment and force transmission are therefore integral parts of the activation mechanism.\n\nThe integrin αvβ8 also activates latent TGF-β, but its mechanisms depend on cellular context and do not simply reproduce the traction-based αvβ6 model. Proteolysis can contribute, and structural studies have demonstrated arrangements in which integrin-bound latent ligand becomes accessible for receptor engagement without unrestricted diffusion of the mature cytokine. On immune cells, the membrane protein GARP, encoded by LRRC32, can tether latent TGF-β1 to the surface. This arrangement is important on activated regulatory T cells and platelets because it localizes the substrate for activation near interacting cells.\n\nProteases, thrombospondin-1, and physicochemical changes can also activate latent TGF-β under particular conditions. Acidification is widely used to activate samples experimentally, but this laboratory procedure should not be equated with the dominant mechanism in intact tissues. Crucially, the concentration of total TGF-β is not the same as the concentration of receptor-accessible TGF-β. A tissue may contain a large latent pool while maintaining very little active ligand, allowing activation to provide rapid, localized control without new protein synthesis."
    },
    {
      "heading": "Receptors and the canonical SMAD pathway",
      "text": "Active TGF-β signals primarily through two transmembrane receptor kinases: the type II receptor, TGFBR2, and the type I receptor, TGFBR1, also called ALK5. Both contain intracellular serine/threonine kinase domains. A mature ligand dimer usually assembles a complex containing two type II and two type I receptor molecules. Ligand binding brings the receptors into a productive arrangement, allowing the constitutively active type II kinase to phosphorylate a glycine- and serine-rich regulatory region of the type I receptor.\n\nThe activated type I receptor then phosphorylates the carboxy-terminal regions of SMAD2 and SMAD3. These receptor-regulated SMADs form complexes with SMAD4, a shared signaling mediator. The complexes accumulate in the cell nucleus, where they regulate transcription in cooperation with DNA-binding partners, coactivators, and corepressors. A simplified representation of the core sequence is:\n\n\\[\\mathrm{TGF\\!-\\!\\beta} \\rightarrow \\mathrm{TGFBR2/TGFBR1} \\rightarrow \\mathrm{SMAD2/3} \\rightarrow \\mathrm{SMAD4\\ complexes} \\rightarrow \\mathrm{transcriptional\\ responses}.\\]\n\nThis sequence is not a one-way conveyor belt. SMAD proteins continually shuttle between cellular compartments, undergo dephosphorylation, and interact with regulatory proteins. The duration and location of these events affect the transcriptional response. SMAD3 can bind particular DNA sequences directly, whereas the predominant full-length SMAD2 isoform has an inserted region that restricts comparable direct DNA binding. Both depend heavily on other transcription factors to achieve selective genomic targeting. Consequently, the same receptor signal can regulate different genes in epithelial cells, fibroblasts, or lymphocytes.\n\nAccessory receptors influence ligand presentation. TGFBR3, commonly called betaglycan, has no intrinsic kinase activity but can bind TGF-β and facilitate its interaction with signaling receptors. It is particularly important for efficient TGF-β2 presentation because TGF-β2 binds TGFBR2 less effectively than TGF-β1 or TGF-β3. Endoglin is another accessory component with important functions in vascular cells, although its ligand interactions extend into other branches of the superfamily. Vascular signaling also involves ALK1 and BMP ligands, so not every endoglin-associated effect represents the canonical TGF-β–ALK5 pathway.\n\nSeveral mechanisms limit signaling. SMAD7 can inhibit access of receptor-regulated SMADs to the activated type I receptor and recruit ubiquitin-associated machinery that promotes receptor downregulation. Receptor internalization, recycling, and degradation further influence signal strength. Because TGF-β itself induces inhibitory regulators such as SMAD7, the pathway contains negative feedback. This architecture permits transient responses to brief signals while allowing sustained signaling when ligand activation and receptor engagement continue."
    },
    {
      "heading": "Signal integration and cell-specific responses",
      "text": "TGF-β signaling is often described as pleiotropic because one ligand family produces many biological effects. This diversity is not evidence that its actions are arbitrary. Cells interpret the signal through pre-existing regulatory programs: accessible regions of chromatin, available transcription factors, receptor abundance, and the activity of other signaling pathways. The SMAD complex supplies one component of a larger decision-making system rather than a complete instruction by itself. Tissue organization and the mechanical environment also help determine the response.\n\nIn many epithelial cells, TGF-β restricts progression through the cell cycle. Its transcriptional program can increase cyclin-dependent kinase inhibitors, including p15 and p21, while repressing the growth-promoting transcription factor MYC. These changes reduce the activity of complexes needed for entry into DNA synthesis. Depending on cellular state, TGF-β may also promote apoptosis, differentiation, or long-lasting growth arrest. None of these outcomes follows automatically from receptor occupancy: oncogenic alterations, survival signals, and loss of downstream regulators can interrupt the cytostatic program.\n\nReceptors also engage pathways outside the core SMAD cascade. These include MAP kinase pathways such as ERK, JNK, and p38; PI3K–AKT signaling; and Rho-family GTPases that regulate shape and motility. Some responses arise through direct or adaptor-mediated receptor interactions, whereas others depend on secondary factors induced after transcription. Calling all of them “SMAD-independent” is useful shorthand, but it can conceal extensive cooperation between SMAD and non-SMAD components. An observed change in migration, for example, may require both rapid cytoskeletal regulation and slower transcriptional remodeling.\n\nA prominent example of contextual signaling is epithelial–mesenchymal transition, or EMT. In responsive experimental systems, TGF-β induces transcriptional regulators such as SNAIL and ZEB proteins, reduces epithelial junctional components, and increases mesenchymal-associated proteins and motility. However, EMT is not necessarily a complete or irreversible switch between two pure cell identities. Cells can occupy partial states, retain epithelial features, or reverse aspects of the program when conditions change. These distinctions are important when relating cell-culture observations to development, fibrosis, or tumors.\n\nSignal timing also matters. A short pulse and continuous exposure can produce different transcriptional programs even when their peak receptor activity is similar. Cells can adapt through receptor regulation, inhibitory proteins, or altered ligand availability. At the tissue level, TGF-β-induced matrix deposition may change stiffness and thereby alter subsequent ligand activation. Thus, intracellular signaling and extracellular mechanics form a coupled system: the response changes the environment, and the changed environment influences the next signal."
    },
    {
      "heading": "Embryonic development and tissue homeostasis",
      "text": "TGF-β family signaling contributes to the coordination of proliferation, differentiation, and tissue architecture during development. The three mammalian TGF-β isoforms have overlapping biochemical activities but are not interchangeable in vivo. Differences in the timing and location of expression, latent-complex organization, and receptor access generate distinct developmental requirements. Genetic disruption in mice has been especially informative because it reveals functions that are difficult to infer from experiments in isolated cells.\n\nLoss of Tgfb2 in mice causes multiple developmental abnormalities affecting structures that can include the heart, lungs, and skeleton. Tgfb3-deficient mice show failure of normal secondary-palate fusion, demonstrating a requirement for this isoform in the coordinated behavior of epithelial tissues during craniofacial development. Tgfb1-deficient mice that survive early development develop severe inflammatory disease after birth, revealing a particularly important role in immune homeostasis. Phenotypic severity and survival can vary with genetic background, so these models identify biological requirements rather than a single invariant sequence of abnormalities.\n\nIn adult tissues, TGF-β helps regulate the balance between renewal and differentiation. It can restrain epithelial proliferation and influence the behavior of hematopoietic stem cells and other stem or progenitor populations. These effects are strongly dependent on niche conditions and cell state. A signal that maintains quiescence in one population may promote differentiation or a stress response in another. TGF-β also influences interactions among epithelial cells, stromal cells, vascular cells, and resident immune cells, making its homeostatic role broader than a direct action on stem cells alone.\n\nThe vascular system illustrates the consequences of disturbed pathway regulation. Pathogenic variants in TGFBR1, TGFBR2, SMAD3, TGFB2, and TGFB3 are associated with forms of Loeys–Dietz syndrome, a group of inherited connective-tissue disorders that can involve arterial aneurysms and other cardiovascular, skeletal, and craniofacial features. These conditions cannot be understood simply as “too little” TGF-β signaling. Diseased tissues may show increased downstream signaling markers despite variants that impair aspects of receptor or ligand function, reflecting compensatory responses and interactions among different cell populations.\n\nFibrillin biology provides another connection between structural integrity and signaling. In Marfan syndrome, pathogenic variants in FBN1 alter an extracellular structural protein whose microfibrils interact with latent TGF-β-binding proteins. Abnormal matrix organization can affect both tissue mechanics and growth-factor regulation. The relative contribution of these processes varies among organs and experimental models. These examples demonstrate why measurements of a single signaling component cannot, by themselves, establish the direction or cause of a complex tissue phenotype."
    },
    {
      "heading": "Immune regulation and mucosal function",
      "text": "TGF-β is a major regulator of immune tolerance, but it is not a universal inhibitor of every immune response. Its effects depend on the responding cell, antigenic stimulation, and accompanying cytokines. The severe inflammatory phenotype of TGF-β1-deficient mice established that endogenous TGF-β1 is essential for limiting inappropriate immune activation. Subsequent cell-specific receptor deletions showed that direct TGF-β responses in T cells are an important part of this regulation.\n\nIn suitable conditions, TGF-β promotes the differentiation of naïve CD4 T cells into peripheral regulatory T cells expressing the transcription factor FOXP3. This process requires additional signals, notably interleukin-2, and is influenced by antigen presentation and tissue-derived factors. It should be distinguished from the development of regulatory T cells in the thymus, which follows a different developmental program. Activated regulatory T cells can themselves present latent TGF-β1 through GARP, creating an additional means of controlling its local availability.\n\nTGF-β also participates in the development of inflammatory and tissue-associated T-cell populations. In commonly used mouse culture systems, TGF-β together with interleukin-6 supports differentiation toward IL-17-producing helper T cells, whereas different cytokine combinations can favor other outcomes. Human differentiation pathways are not identical to those reconstructed in mice. This context dependence explains why describing TGF-β exclusively as “anti-inflammatory” is incomplete: it can restrain certain immune functions while helping establish cell populations involved in barrier defense and inflammation.\n\nAt mucosal surfaces, TGF-β contributes to B-cell class-switch recombination toward immunoglobulin A, an antibody class central to intestinal and other mucosal secretions. This response requires cooperation with activation signals and additional cytokines rather than TGF-β alone. In the intestine, dendritic cells expressing αvβ8 can activate latent TGF-β and help shape local T-cell responses. The restriction of activation to particular cell contacts or tissue sites helps reconcile a broadly distributed latent cytokine with highly localized immune effects.\n\nTGF-β can suppress cytotoxic programs in natural killer cells and CD8 T cells, including the expression of selected killing-associated molecules. It also influences macrophage behavior, dendritic-cell function, and tissue-resident lymphocyte programs. These activities may protect healthy tissues from excessive immune injury, but similar mechanisms can limit immune attack against tumors. Immune regulation by TGF-β therefore depends not only on the amount of cytokine present but also on which cells activate it, which cells receive the signal, and whether the surrounding tissue is healthy, infected, damaged, or malignant."
    },
    {
      "heading": "Wound repair and fibrotic disease",
      "text": "Normal wound repair requires coordinated inflammation, cell migration, new matrix deposition, and remodeling. TGF-β participates in each of these processes. Platelets contain substantial stores of TGF-β1 and release it during activation, while recruited immune cells and local stromal cells provide additional sources. The biological effects of these sources depend on subsequent latent-ligand activation. TGF-β can influence cell recruitment and migration, promote fibroblast activity, and help replace damaged tissue with a provisional matrix that later undergoes reorganization.\n\nIn fibroblasts, TGF-β commonly increases the production of collagens, fibronectin, and other extracellular-matrix components. It can also favor matrix accumulation by changing the expression of enzymes and inhibitors involved in matrix turnover. Under appropriate mechanical and biochemical conditions, it promotes the differentiation of fibroblasts into contractile myofibroblasts, often identified by increased α-smooth-muscle actin and stress-fiber formation. These cells contribute to wound contraction, but their persistence is associated with excessive scarring.\n\nFibrosis develops when matrix production and remodeling remain dysregulated, leading to the accumulation of connective tissue and disruption of normal organ architecture. TGF-β is a central participant in fibrotic responses affecting the lung, liver, kidney, skin, and other organs. The cells producing the matrix differ among tissues: activated hepatic stellate cells are major contributors in liver fibrosis, whereas resident fibroblast populations are important in many lung and kidney settings. Shared signaling features do not imply that all fibrotic diseases have the same initiating cause or cellular organization.\n\nMechanical feedback can sustain fibrosis. As collagen accumulates and matrix stiffness increases, cells can generate greater traction against their surroundings. Where suitable integrins and matrix-anchored latent complexes are present, this promotes further TGF-β activation. The resulting signal supports additional matrix deposition and contractility. Persistent injury, inflammatory mediators, and altered matrix turnover interact with this loop; TGF-β alone does not account for every component of disease progression.\n\nThe relationship between EMT and fibrosis requires particular care. Injured epithelial cells may acquire partial mesenchymal features, change their secretory behavior, and contribute to a profibrotic environment. However, this does not establish that they become the major collagen-producing fibroblast population in vivo. Genetic lineage-tracing studies have restricted earlier interpretations based largely on overlapping marker expression in cultured cells or tissue sections.\n\nTGF-β isoforms can also differ in their effects on repair and scarring, but a simple division into “profibrotic TGF-β1” and “antifibrotic TGF-β3” is inadequate. Isoform effects depend on timing, dose, tissue, and mode of delivery. Because TGF-β also supports normal repair and immune regulation, blocking the pathway affects physiological processes as well as pathological matrix accumulation."
    },
    {
      "heading": "Tumor suppression and cancer progression",
      "text": "TGF-β has a dual relationship with cancer. In many normal and early neoplastic epithelial cells, it suppresses proliferation, promotes differentiation, or induces cell death. These actions can oppose tumor development. During progression, however, cancer cells may lose sensitivity to the growth-inhibitory program while retaining other TGF-β responses. The surrounding stromal and immune cells can remain responsive even when the malignant cells themselves have defects in receptor signaling.\n\nGenetic alterations provide direct evidence for tumor-suppressive functions. TGFBR2 is frequently disrupted in colorectal cancers with defective DNA mismatch repair, in part because a repetitive sequence within the gene is vulnerable to frameshift mutation. SMAD4 is a tumor-suppressor gene commonly inactivated in pancreatic ductal adenocarcinoma and in subsets of other cancers. Such changes can disable important components of the canonical pathway, although their consequences depend on the other alterations present in the tumor.\n\nOnce cytostatic restraints have been bypassed, TGF-β can support invasive behavior. It can induce partial or more extensive EMT-associated programs, alter cell adhesion, and increase motility in responsive cancer cells. It also acts on fibroblasts, vascular cells, and immune cells within the tumor microenvironment. Cancer-associated fibroblasts may increase matrix deposition and remodeling, changing both the physical properties of the tissue and the availability of signaling molecules. These effects are not uniform across tumor types or even across different regions of one tumor.\n\nImmune suppression is another major mechanism. TGF-β can reduce cytotoxic lymphocyte function and influence the organization of fibroblast-rich stroma associated with restricted T-cell access to malignant cells. This provides a mechanistic basis for studying combinations of TGF-β pathway inhibition with immune-checkpoint blockade. Nevertheless, the presence of a TGF-β-associated expression signature does not establish that every tumor will respond to pathway inhibition. The relevant source of ligand, the activating cell population, and the ability of immune cells to reach the tumor all matter.\n\nThe phrase “TGF-β paradox” is sometimes used for its apparently opposing roles in cancer, but these roles are better understood as stage- and compartment-dependent. A ligand can restrain proliferation in one cell while promoting matrix production or immune suppression in another. Tumor-cell receptor mutations, stromal activation, and spatially localized ligand activation can coexist.\n\nThis complexity also limits simple biomarker interpretations. High total TGF-β in blood or tissue does not directly measure active signaling in malignant cells. Likewise, phosphorylated SMAD2 or SMAD3 indicates activity in a shared signaling branch rather than proving a particular tumor-promoting mechanism. Cell identity, genomic alterations, and spatial context are necessary for interpreting such measurements."
    },
    {
      "heading": "Experimental measurement and therapeutic targeting",
      "text": "TGF-β research requires a distinction among ligand abundance, ligand activation, receptor engagement, and biological response. Immunoassays often measure total TGF-β after acid treatment releases the mature cytokine from latent complexes. Measurements performed without activation are intended to assess active ligand, but sample handling and assay design can strongly affect the result. Blood measurements are particularly sensitive to platelet release: serum formation involves clotting, which can substantially increase measured TGF-β1 relative to carefully prepared platelet-poor plasma.\n\nCell-based reporter assays provide a functional measure of signaling activity. A commonly used design employs TGF-β-responsive regulatory sequences from the SERPINE1 gene, which encodes plasminogen activator inhibitor-1, coupled to a reporter such as luciferase. Other assays measure SMAD-dependent transcription, receptor-proximal phosphorylation, or changes in endogenous target genes. No single assay captures all aspects of the pathway. Reporter responses may depend on the receptor repertoire of the assay cells, and target-gene changes may integrate signals from several pathways.\n\nPhosphorylated SMAD2 and SMAD3 are widely examined by immunoblotting, immunohistochemistry, and related methods. They provide useful evidence of pathway activity but are not uniquely specific for TGF-β because activins and some related ligands can activate the same SMAD branch. Conversely, the absence of a strong phosphorylation signal at one sampling time does not exclude a previous transient response. Genetic perturbation, ligand-specific blockade, and receptor-selective experiments help distinguish among these possibilities. Single-cell and spatial approaches further reveal which cells produce ligands, express activation machinery, and respond within complex tissues.\n\nTherapeutic strategies investigated in human disease include ligand-neutralizing antibodies, soluble receptor-based ligand traps, small-molecule receptor kinase inhibitors, and agents directed at latent-ligand activation. Fresolimumab is an example of an antibody developed to neutralize the three mammalian TGF-β isoforms. Galunisertib is an example of an investigational ALK5 kinase inhibitor. These approaches act at different levels: neutralizing an active ligand is not equivalent to blocking its activation at a selected tissue site or inhibiting a receptor used by multiple local sources.\n\nSelectivity is a central issue. Systemic pathway inhibition can interfere with immune restraint, epithelial homeostasis, vascular biology, and wound repair, whereas local or activation-specific approaches aim to restrict which signaling pools are affected. Experimental compounds that inhibit ALK5 may also inhibit related kinases at relevant concentrations, so pharmacological findings require careful attribution. Likewise, deleting a receptor throughout development can produce consequences different from inhibiting it temporarily in an adult.\n\nClinical interpretation therefore depends on disease setting, molecular target, exposure, and evidence from the specific intervention. A reduction in a circulating marker or a tissue signaling readout demonstrates a different endpoint from improved organ function, delayed tumor progression, or prolonged survival. Studies of TGF-β-targeted agents distinguish these pharmacodynamic effects from clinical outcomes and examine whether pathway inhibition acts on the intended cell populations."
    }
  ],
  "links": [
    "cytokine",
    "gene-expression",
    "amino-acid",
    "disulfide-bond",
    "protein",
    "endoplasmic-reticulum",
    "cytoskeleton",
    "cell-nucleus",
    "transcription-biology",
    "transcription-factor",
    "feedback",
    "chromatin",
    "cell-cycle",
    "apoptosis",
    "loeys-dietz-syndrome",
    "immune-tolerance",
    "t-cell",
    "regulatory-t-cell",
    "foxp3",
    "interleukin-two",
    "b-cell",
    "class-switch-recombination",
    "natural-killer-cell",
    "macrophage",
    "platelet",
    "collagen",
    "fibrosis",
    "cancer",
    "dna-mismatch-repair",
    "immune-checkpoint",
    "biomarker",
    "immunohistochemistry",
    "antibody"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}