TGF- Signaling, Tumor Suppression, and Acute Lymphoblastic Leukemia
http://www.100md.com
《新英格兰医药杂志》
The emergence of a single cancer cell is an extremely complex process that depends not only on the escape of a normal cell from growth-control mechanisms but also on the ability of the abnormal cell to bypass the numerous molecular checkpoints that control its propagation. Among these barriers against abnormal cells is a diverse family of proteins called tumor suppressors, one of which is transforming growth factor (TGF-). This polypeptide initiates a signaling pathway that suppresses the early development of cancer cells in a number of different cell lineages.
TGF- exists in three versions, or isoforms: TGF-1, 2, and 3. They are cytokines produced by many kinds of cells, and when they bind to their cell-membrane receptors, they set off a complex signaling cascade that starts in the cytoplasm and terminates in the nucleus. A recurrent feature of this cascade, from its beginning to its end, is the recruitment of auxiliary, cooperating proteins by core elements of the cascade.
The prototype of the TGF- family, TGF-1, binds with high affinity to the type II TGF- receptor (see Figure). This type II receptor–TGF-1 unit recruits the type I TGF- receptor to form a heteromeric complex. This three-component module then allows the kinase activity of the type II receptor to phosphorylate multiple serine and threonine residues in a cytoplasmic region of the type I receptor termed the GS domain. The phosphorylation of these amino acids alters the conformation of the TGF- receptor I in a way that allows its GS domain to bind members of the Smad family of proteins (named for their homology to Caenorhabditis elegans Sma and drosophila MAD proteins).
Figure. The TGF-–Smad Signal-Transduction Pathway.
TGF-1 binds cooperatively to the type I and II receptors (TRI and TRII) and results in the phosphorylation of the type I receptor in the GS domain by the constitutively activated type II receptor. Phosphorylated and activated TGF- receptor I, in turn, interacts with Smad2, Smad3, or both in complex with the Smad anchor for receptor activation (SARA), leading to the phosphorylation of Smad proteins, their disassociation from SARA, and their movement into the nucleus. In the nucleus, Smad2/3 forms a complex with Smad4, and this complex binds to the DNA sequence CAGAC in the transcriptional regulatory region of target genes. A proportion of Smad2/3 exists in a preformed complex with E2F4/5 and the corepressor p107 and, with TGF-–induced phosphorylation, forms a new complex with Smad4 and binds to a composite E2F–Smad DNA-binding sequence in the promoter of the c-Myc gene, resulting in its repression. Smad complexes also bind in association with other transcription factors (TF) to induce the transcriptional activation of genes encoding p15INK4b (CDKN2B) and p21(CIP1) (CDKN1A). These latter proteins play an important role in inhibiting cell-cycle progression. High levels of c-Myc–MIZ-1 directly antagonize the expression of CDKN2B and CDKN1A by binding to their proximal promoters. The inhibitory Smads (I-Smads) inhibit both Smad2/3 binding to the activated TGF- receptor I and the nuclear translocation of Smad4.
The eight Smad proteins fall into three classes: receptor-activated Smads (R-Smad1, 2, 3, 5, and 8), the comediator Smad (Smad4), and the inhibitory Smads (I-Smad6 and 7). In the presence of TGF-1, the TGF- receptor I binds to and phosphorylates Smad2 and Smad3, causing their release from a cytoplasmic protein and subsequent migration into the nucleus. Within the nucleus, these two proteins join up with Smad4 to form a complex that binds with relatively low affinity to the transcriptional regulatory regions of target genes. However, interactions between this tripartite Smad complex and other transcription factors allows high-affinity binding to the regulatory regions. The effect of these interacting nuclear proteins on expression of the target gene depends on whether they recruit activators or repressors of transcription.
An important effect of TGF- is its limitation of the growth of epithelial, neuronal, and hematopoietic cells. Key elements of this inhibitory mechanism are the induction of the cyclin-dependent kinase inhibitors p15INK4b and p21 (also called CIP1) and the inhibition of c-Myc expression. The nuclear Smad complex that forms in the wake of TGF-–induced signaling binds to the distal promoters of the genes encoding p15INK4b (CDKN2B) and p21 (CDKN1A) and, along with partner transcription factors, induces the expression of these genes (see Figure). For this to occur, however, the intracellular level of c-Myc must be below a certain threshold — high levels of c-Myc inhibit the expression of CDKN2B and CDKN1A, and in this way, c-Myc blunts the ability of TGF- to suppress cellular proliferation. There is, however, a way around this circuit: the activated nuclear Smad complex can join up with two other transcription factors (E2F4/5) as well as p107, a member of the family of retinoblastoma tumor suppressors, to form a transcriptional regulatory complex that inhibits the expression of c-Myc. The lowered level of c-Myc protein will then allow TGF- to activate the genes encoding p15INK4b and p21.
Several components of the TGF-–Smad system are bona fide tumor suppressors with the ability to inhibit the development of cancer in its early stages, and if mutations disable these components, the door to cancer is opened. Such inactivating mutations have been found in the gene encoding the TGF- receptor II (TRII) in patients with colorectal cancers that develop sporadically and in some patients with hereditary nonpolyposis coli, gastric tumors, or gliomas. Inactivating mutations in the gene encoding the TGF- receptor I (TRI) occur in some ovarian, breast, and pancreatic cancers, but more striking is the inactivation of SMAD4 in up to 50 percent of pancreatic cancers. Recently, mutations of SMAD4 have also been identified in 30 percent of metastatic colon cancers, and SMAD2 mutations have been found in rare cases of colorectal and lung cancer.
In this issue of the Journal, Wolfraim et al. (pages 552–559) report direct evidence that Smad3 is an important tumor suppressor in T-lineage acute lymphoblastic leukemia (T-cell ALL). Despite normal levels of SMAD3 messenger RNA, no Smad3 protein was detected in leukemic cells from children with T-cell ALL. The authors also show that a reduction in the level of Smad3 in normal mice impairs the responses of T cells to TGF-. The loss of Smad3 alone was insufficient to induce leukemia, but when coupled with the loss of the p27Kip1 cyclin-dependent kinase inhibitor, whose gene is frequently altered in human T-cell ALL, some cases of leukemia developed in Smad3-deficient mice.
These data suggest that TGF- participates in blocking the development of T-cell ALL by suppressing T-cell proliferation. What remains to be defined are the mechanism that leads to a lack of the Smad3 protein in childhood leukemia and the range of mutations of other genes that, together with the loss of Smad3, promote the development of ALL. Equally important is the challenge of translating our detailed understanding of the TGF- signaling pathway into new therapeutic approaches. As we have seen, this pathway can suppress the early development of cancer, but there is mounting evidence of a permissive role of TGF- in the growth and metastatic behavior of established tumors. These contrasting activities will necessitate considerable finesse in exploiting this pathway for a therapeutic benefit.
Source Information
From the Department of Pathology, St. Jude Children's Research Hospital, Memphis, Tenn.(James R. Downing, M.D.)
TGF- exists in three versions, or isoforms: TGF-1, 2, and 3. They are cytokines produced by many kinds of cells, and when they bind to their cell-membrane receptors, they set off a complex signaling cascade that starts in the cytoplasm and terminates in the nucleus. A recurrent feature of this cascade, from its beginning to its end, is the recruitment of auxiliary, cooperating proteins by core elements of the cascade.
The prototype of the TGF- family, TGF-1, binds with high affinity to the type II TGF- receptor (see Figure). This type II receptor–TGF-1 unit recruits the type I TGF- receptor to form a heteromeric complex. This three-component module then allows the kinase activity of the type II receptor to phosphorylate multiple serine and threonine residues in a cytoplasmic region of the type I receptor termed the GS domain. The phosphorylation of these amino acids alters the conformation of the TGF- receptor I in a way that allows its GS domain to bind members of the Smad family of proteins (named for their homology to Caenorhabditis elegans Sma and drosophila MAD proteins).
Figure. The TGF-–Smad Signal-Transduction Pathway.
TGF-1 binds cooperatively to the type I and II receptors (TRI and TRII) and results in the phosphorylation of the type I receptor in the GS domain by the constitutively activated type II receptor. Phosphorylated and activated TGF- receptor I, in turn, interacts with Smad2, Smad3, or both in complex with the Smad anchor for receptor activation (SARA), leading to the phosphorylation of Smad proteins, their disassociation from SARA, and their movement into the nucleus. In the nucleus, Smad2/3 forms a complex with Smad4, and this complex binds to the DNA sequence CAGAC in the transcriptional regulatory region of target genes. A proportion of Smad2/3 exists in a preformed complex with E2F4/5 and the corepressor p107 and, with TGF-–induced phosphorylation, forms a new complex with Smad4 and binds to a composite E2F–Smad DNA-binding sequence in the promoter of the c-Myc gene, resulting in its repression. Smad complexes also bind in association with other transcription factors (TF) to induce the transcriptional activation of genes encoding p15INK4b (CDKN2B) and p21(CIP1) (CDKN1A). These latter proteins play an important role in inhibiting cell-cycle progression. High levels of c-Myc–MIZ-1 directly antagonize the expression of CDKN2B and CDKN1A by binding to their proximal promoters. The inhibitory Smads (I-Smads) inhibit both Smad2/3 binding to the activated TGF- receptor I and the nuclear translocation of Smad4.
The eight Smad proteins fall into three classes: receptor-activated Smads (R-Smad1, 2, 3, 5, and 8), the comediator Smad (Smad4), and the inhibitory Smads (I-Smad6 and 7). In the presence of TGF-1, the TGF- receptor I binds to and phosphorylates Smad2 and Smad3, causing their release from a cytoplasmic protein and subsequent migration into the nucleus. Within the nucleus, these two proteins join up with Smad4 to form a complex that binds with relatively low affinity to the transcriptional regulatory regions of target genes. However, interactions between this tripartite Smad complex and other transcription factors allows high-affinity binding to the regulatory regions. The effect of these interacting nuclear proteins on expression of the target gene depends on whether they recruit activators or repressors of transcription.
An important effect of TGF- is its limitation of the growth of epithelial, neuronal, and hematopoietic cells. Key elements of this inhibitory mechanism are the induction of the cyclin-dependent kinase inhibitors p15INK4b and p21 (also called CIP1) and the inhibition of c-Myc expression. The nuclear Smad complex that forms in the wake of TGF-–induced signaling binds to the distal promoters of the genes encoding p15INK4b (CDKN2B) and p21 (CDKN1A) and, along with partner transcription factors, induces the expression of these genes (see Figure). For this to occur, however, the intracellular level of c-Myc must be below a certain threshold — high levels of c-Myc inhibit the expression of CDKN2B and CDKN1A, and in this way, c-Myc blunts the ability of TGF- to suppress cellular proliferation. There is, however, a way around this circuit: the activated nuclear Smad complex can join up with two other transcription factors (E2F4/5) as well as p107, a member of the family of retinoblastoma tumor suppressors, to form a transcriptional regulatory complex that inhibits the expression of c-Myc. The lowered level of c-Myc protein will then allow TGF- to activate the genes encoding p15INK4b and p21.
Several components of the TGF-–Smad system are bona fide tumor suppressors with the ability to inhibit the development of cancer in its early stages, and if mutations disable these components, the door to cancer is opened. Such inactivating mutations have been found in the gene encoding the TGF- receptor II (TRII) in patients with colorectal cancers that develop sporadically and in some patients with hereditary nonpolyposis coli, gastric tumors, or gliomas. Inactivating mutations in the gene encoding the TGF- receptor I (TRI) occur in some ovarian, breast, and pancreatic cancers, but more striking is the inactivation of SMAD4 in up to 50 percent of pancreatic cancers. Recently, mutations of SMAD4 have also been identified in 30 percent of metastatic colon cancers, and SMAD2 mutations have been found in rare cases of colorectal and lung cancer.
In this issue of the Journal, Wolfraim et al. (pages 552–559) report direct evidence that Smad3 is an important tumor suppressor in T-lineage acute lymphoblastic leukemia (T-cell ALL). Despite normal levels of SMAD3 messenger RNA, no Smad3 protein was detected in leukemic cells from children with T-cell ALL. The authors also show that a reduction in the level of Smad3 in normal mice impairs the responses of T cells to TGF-. The loss of Smad3 alone was insufficient to induce leukemia, but when coupled with the loss of the p27Kip1 cyclin-dependent kinase inhibitor, whose gene is frequently altered in human T-cell ALL, some cases of leukemia developed in Smad3-deficient mice.
These data suggest that TGF- participates in blocking the development of T-cell ALL by suppressing T-cell proliferation. What remains to be defined are the mechanism that leads to a lack of the Smad3 protein in childhood leukemia and the range of mutations of other genes that, together with the loss of Smad3, promote the development of ALL. Equally important is the challenge of translating our detailed understanding of the TGF- signaling pathway into new therapeutic approaches. As we have seen, this pathway can suppress the early development of cancer, but there is mounting evidence of a permissive role of TGF- in the growth and metastatic behavior of established tumors. These contrasting activities will necessitate considerable finesse in exploiting this pathway for a therapeutic benefit.
Source Information
From the Department of Pathology, St. Jude Children's Research Hospital, Memphis, Tenn.(James R. Downing, M.D.)