Genes of Smad proteins were discovered in Drosophila and C.elegans through genetic screening. The name Smad is a fusion of two gene names, Drosophila mothers against dpp (Mad) and C.elegans Sma. Mad and Sma proteins (Smads), are of 42-60 KD, were discovered as molecules which act as essential factors in downstream of Ser/Thr kinase receptors in TGF-β pathway. In humans, eight Smad proteins have been identified and classified into three groups on the basis of structure and function.
Receptor regulated Smads (R-Smads) directly interact with TGF-β receptor kinases. These include Smad 1, 2, 3, 5 and 8. Smad 1, 5 and 8 share close homology and mediate BMP signaling whereas Smad 2 and 3 mediate TGF-β and Activin signaling. Activated type I receptor kinases phosphorylate particular R-Smads at two serine residues on a conserved C-terminal SSXS motif. Loss of signaling has been observed upon mutation of these serine residues. Smad 1, 5 and 8 are phosphorylated by ALK1, ALK2, ALK3 and ALK6 whereas Smad 2 and 3 are phosphorylated by ALK4, ALK5 and ALK7. ALKs are Human activin receptor like kinases and are type I receptors in TGF-β signal transduction pathway.
Common Smads (Co-Smads) associate with R-Smads forming heteromeric complexes and carry the signal further to the nucleus. This group includes only one protein called Smad4 which is similar in structure as R-Smads but is not phosphorylated. Smad4 takes part in TGF-β, activin and BMP signaling pathway along with corresponding R-Smads.
Inhibitory Smads (I-Smads), inhibit the TGF-β signaling done by R-Smads and Co-Smads. These include Smad 6 and 7. The Smad6 inhibit BMP signaling whereas the Smad7 inhibit both, the TGF-β and BMP signaling.
Smad Protein Structure
The Smad proteins are around 500 amino acids in length and contain two conserved structural domains, the amino-terminal MH1 domain and the carboxy-terminal MH2 domain. The R-Smads are the only ones that contain a characteristic SXS motif at their Carboxy termini (refer to this domain as "Tail"). The MH1 (MAD-homology 1) domain of most R-Smads and Smad4 display DNA binding activity, it may play a role in nuclear import, and negatively regulates the function of the MH2 domain.
The amino-terminal region of I-Smads displays weak sequence homology to the MH1 domain of R-Smads and does not bind to DNA. The MH2 domain, on the other hand, is highly conserved among all Smad proteins and participates in protein-protein interactions, namely in receptor interaction, in the formation of homomeric as well as heteromeric Smad complexes, and in the interaction with the nuclear pore complex for nucleocytoplasmic shuttling. The phosphorylation of the C-terminal two serine residues in the SXS motif of the MH2 domain (pTail) by the receptor kinases drives the activation of the R-Smad. Once Smads get activated and translocated to the nucleus, the interactions of both, MH1 and MH2 domains with nuclear proteins determine the final transcriptional output.
The MH1 and MH2 domains are connected by a linker region that is less conserved among Smads and its function is yet poorly understood. In Smad1 the linker contains four MAPK phosphorylation sites (PxSP motifs) and two putative GSK-3P sites (S/TxxxS motifs) all conserved in Smads 5 and 8. Downstream of these sites lays a PPAY (or "FY") motif that mediates specific interaction with the HECT-domain-containing E3 Smurf ubiquitin ligases. In Smad2 and 3 the linker region contains four SP sites for proline-directed kinases, and a PY motif after the first SP site. Those phosphorylations sites allow specific crosstalks with other signaling pathways.
Phosphorylation of the tail of the TGF-β R-Smads destabilizes Smad interaction with SARA (Smad anchor for receptor activation). SARA contains a phospholipid binding FYVE domain, which targets the molecule to the membrane of early endosomes where it allows more efficient recruitment of Smad2 or Smad3 to the receptors for phosphorylation.
The destabilization of the interaction between Smad2 and 3 and SARA mediated by the phosphorylation of the tail allows the dissociation of Smad from the complex and the subsequent exposure of a nuclear import region on the Smad MH2 domain and also augments the affinity for Smad4. The interface between the C-terminal pSer-X-pSer motif of the R-Smad and the basic surface pocket of the Smad4 MH2 domain play an important role in the formation of a heteromeric complex. The R-Smads-Smad4 complex accumulates in the nucleus and in association with other cofactors regulates the gene ex
DNA recognition by Smads
All R-Smads with the exception of Smad2 bind to DNA in a sequence specific manner. The Smad binding element (SBE), identified as the best DNA binding sequence for Smad3 contains only four base pairs, 5'-AGAC-3', although most naturally occurring DNA sequences contains an extra base C at the 5' end. Smad4 also recognize the same sequence. The crystal structure of the Smad3 MH1 domain bound to SBE revealed that a highly conserved β hairpin makes specific contacts to three bases of the SBE.
Surprisingly, the most common spliced form of Smad2 contains an insertion of 30 residues between the DNA binding β hairpin and the helix H2 which results in the poor DNA binding ability of Smad2. These extra amino acids correspond to the exon 3 of Smad2 and may play a crucial role in modulating the function of Smad2 by interfering with the direct DNA binding to target genes. The less commonly expressed isoform of Smad2, lacking this insertion, has the ability to bind to DNA. The very short DNA sequence recognized by Smads has made very difficult the identification of biologically relevant Smad-responsive DNA elements and at the same time emphasizes the cooperation with other DNA binding partners.