JP2003502064A - Antagonists of BMP and TGFβ signaling pathway - Google Patents

Abstract

  (57) [Summary] The present invention provides specific Smads, unique members of the Hect family of ubiquitin ligases that specifically target the BMP and TGFβ / activin pathways. The new ligases were named Smurf1 and Smurf2. They interact directly with Smads 1 and 5, and Smad7, respectively, and regulate ubiquitination, turnover and activity of Smad and other proteins in these pathways. Smurf1 interferes with the biological response to BMP but does not interfere with activin signaling. In amphibian embryos, Smurf1 inhibits endogenous BMP signals, which alters pattern formation and cell fate decisions in mesoderm and ectoderm. The present invention provides a unique regulatory link between the ubiquitination pathway during embryonic development and the regulation of cell fate by the TGFβ superfamily. Thus, Smurf1 is a negative regulator of Smad1 signaling, and by targeting Smad1, Smurf1 inhibits BMP signaling. Smurf2 suppresses TGFβ signaling in mammalian cells and inhibits formation of dorsal mesoderm in Xenopus, causing anterior cleavage of the embryo. Smurf2 forms a stable complex with Smad7, causing decreased and down-regulated TGFβ / activin signaling.

Description

Detailed Description of the Invention

BMP Antagonists and TGFβ Signaling Pathways The research that led to the present invention was conducted by the National Institute of Health.
) Partially supported by grant number SR01HD3242902. Accordingly, the United States Government has certain rights in this invention.

[0002] In the background substantially the entire phylum of the invention, an important step in embryonic development, from a cell growth factor secreted is interposed into the cell, or inductive, it is to be regulated by a signal. In particular, members of the transforming growth factor beta (TGFβ) superfamily regulate a myriad of cellular and developmental processes such as mitosis, cell differentiation, embryo patterning and organogenesis. In vertebrate embryos, various TGFβ signals influence germ specification, body patterning, cell proliferation and differentiation (1-
Four). In amphibian Xenopus embryos, different TGFβ members induce different cellular behaviors, eg activin, VgI and nodal induce characteristic mesoderm on the back of embryos such as notochord and muscle. VgI and activin also induce endoderm features. In contrast, bone morphogenetic proteins (BMPs) define mesoderm and mesenchyme, such as blood, and regulate epithelial and neuronal differentiation in the ectoderm (see (5) for a review).

Cells respond to ligands of the TGFβ family and transmit signals directly from the cell surface receptor complex to nuclear DNA targets via proteins of the Smad family (reviews (4), (6)). , (7), and (52)).

[0004] Smad is a Drosophila Mad (decapen Plesic (decapen
(mother to taplegic) [dpp] and three nematode genes Sma 2, Sma 3, and
Related to the protein encoded by Sma4. The terms Sma and Mad are fused together and are referred to as Smads. The Smad family has eight members. Phosphorylated Smads 1, 5 and 8 are functional mediators of BMP family signaling in cooperation with Smad4. Smads 2 and 3 are signaling agents for the action of TGFβ and activin. Smads 6 and 7 function as antagonists that inhibit TGFβ / BMP superfamily signaling. Interestingly,
Smad7 is localized in the nucleus and accumulates in the cytoplasm in response to TGFβ signaling (73)
. Furthermore, expression of both Smad6 and Smad7 is regulated by TGFb, BMP, growth factors and cytokines, thereby conferring negative feedback regulation of the Smad signaling pathway (53-58). Upon entering the nucleus, phosphorylated Smad 1 forms a heteromeric complex with Smad 4 and activates transcription of early response genes. BMP receptors also signal via mitogen-activated protein kinases. BMP
Regulate cell cycle progression and govern the differentiation of mesenchymal stem cells.

The two major pathways, BMP and activin / TGFβ signaling, have been described in detail. Two different receptor subunits, type I and type II transmembrane serine / threonine kinases, form an activation complex upon binding with a ligand. In these complexes, the type II subunit activates the type I subunit, which directly phosphorylates and activates a specific receptor-regulated R-Smad protein: the BMP receptor and Smad1 closely. While targeting related Smads 5 and 8, activin and TGFβ receptors target Smad1 and the closely related Smads 2 and 3. Upon activation, these R-Smads form a heteromeric complex with Smad4, a "common partner". This complex transporter to the nucleus binds to the promoter of the target gene in cooperation with the DNA-binding protein and activates transcription by assembling the coactivator. A third class of inhibitory Smads (I-Smads), Smad6 and Smad7, function as inhibitors that block Smad-Smad complex formation or Smad-receptor interactions. I-Smad is either in the cytoplasmic domain of the receptor or directly on Smad1
Combine with. Component mutations at all levels of this pathway are associated with embryonic defects and various cancers, underscoring the importance of this growth factor family in developmental and disease processes (reviewed in (4), (See (7)). In particular, Smad2 and Smad4 defects are associated with colon and lung cancer, and human Smad4 defects are associated with pancreatic cancer.

Smad has no endogenous enzymatic activity. Thus, the nature of the cellular response to Smad signaling is particularly sensitive to intracellular Smad protein levels. In fact, selective cell behavioral determination of Xenopus embryos can be achieved by relatively small changes in the amount of Smad protein expressed in cells (8-11). Therefore,
Regulation of intracellular Smad protein levels can be used as a method to regulate morphogenic signaling by the TGFβ superfamily.

[0007] Covalent protein modification of ubiquitin is recognized as a general signal that directs proteins to proteasome-mediated degradation (reviewed in (12),
(See (13)). Targets for selective ubiquitination include transcription factors, cell cycle regulators, signal transduction proteins, and membrane proteins (ref. (12)). Selective ubiquitination and degradation of specific target proteins can function as cell cycle progression, programmed cell death, differentiation and embryonic development. Dysfunction of the ubiquitination pathway is associated with disease and aberrant development. Ubiquitin ligase is a part of the 12.5 kD polypeptide, a multimeric complex that catalyzes the covalent binding of ubiquitin, targeting proteins. The binding of ubiquitin to its target acts as a molecular "flag", labeling ubiquitinated proteins for proteolysis via an organelle known as the 26S proteasome. At least 3 enzymes, namely E
1, E2 and E3 are involved in ubiquitin binding to target proteins. E
One enzyme activates the ubiquitin molecule and binds it with the E2 enzyme, which then either binds ubiquitin directly to the target protein or sends it to the E3 ubiquitin ligase. E3 recognizes a specific substrate and leads to ubiquitination.

[0008] Multiple examples of developmental regulation by the ubiquitination system have been reported by Dictyosteliu.
m) (14-16) and Drosophila (17-21). Ubiquitin-receptor binding has been used in a variety of systems to regulate endocytosis and signal transduction, and to regulate steady-state levels of receptors through proteasome- and lysosome-mediated degradation (59-60). Although direct ubiquitination of membrane receptors has been characterized in multiple systems, in some cases ubiquitin-dependent regulation does not appear to involve direct ubiquitin-receptor binding (61). -62). Many cell surface receptors are regulated by ubiquitin-dependent pathways but bind membrane proteins and target them for ubiquitination
Only a small number of E3 ubiquitin ligases are defined. Nedd4, Cd-WW-HECT domain E
3 Ubiquitin ligase can regulate the turnover of amiloride-sensitive sodium channels by binding directly to the PPXY motif present at the carboxy terminus of the channel (63-66). In addition, the RING finger protein, c-cbl, has recently been shown to function as an E3 ubiquitin ligase, which binds to the EGF receptor and mediates ubiquitination and down-regulation of the receptor complex (67-68). ). In these instances, ubiquitination of membrane proteins appears to involve a direct interaction between E3 ligase and target proteins. It is not known whether the adapter protein functions to assemble E3 ligase into a specific receptor complex. The mechanism and target of ubiquitination in the control of pattern formation have not been clarified until now.

References cited by numbers throughout this specification are listed after the examples. All references cited herein are incorporated by reference.

SUMMARY OF THE INVENTION The present invention advantageously provides a class of regulatory proteins involved in BMP and TGFβ mediated activation. In particular, these proteins regulate Smad proteins and / or promote degradation of the TGFβ receptor complex in the presence of Smad proteins. By manipulating the activity of the proteins of the invention, one of skill in the art can upregulate or downregulate cell activation, eg, via BMP or TGFβ.

Therefore, in a first aspect, the invention provides an isolated Smurf protein, and in particular a human Smurf protein. In certain embodiments, it is Smurf1 protein. In an alternative embodiment, it is the Smurf2 protein. In the particular embodiment illustrated below, human Smurf1 is shown in FIG. 10 (SEQ ID NO: 2).
And the human Smurf2 has the amino acid sequence set forth in Figure 12 (SEQ ID NO: 4). The Smurf protein of the invention may contain at least 5, and preferably at least 10 contiguous amino acids from the sequences set forth in SEQ ID NOs: 2 and 4.

[0012]   The invention further provides an antibody that specifically binds to the Smurf protein.

The invention further provides a nucleic acid encoding the Smurf protein of the invention. In a particular embodiment, the nucleic acid has the nucleotide sequence set forth in SEQ ID NO: 1 or SEQ ID NO: 3.

The present invention further provides an oligonucleotide or nucleic acid that specifically hybridizes with a nucleic acid having a sequence encoding Smurf or a sequence complementary thereto under highly stringent conditions. Such hybridizable nucleic acids include probes (they are labeled), primers (eg for PCR amplification), antisense nucleic acids, ribozymes, and triple helix forming nucleic acids.

The invention further provides a vector containing a nucleic acid encoding Smurf, eg under the control of expression control sequences. Also, a host cell containing such a vector,
Also provided are methods of producing Smurf by culturing such host cells under conditions that allow expression of the Smurf protein from the vector.

Also contemplated are transgenic non-human animals expressing the human Smurf protein and non-human animals deficient in the endogenous Smurf protein.

The invention further provides a method of inhibiting a bone morphogenetic protein or transforming growth factor β activation pathway in a cell. The method involves growing the cell under conditions that allow expression of Smurf from the vector introduced into the cell. Alternatively, the present invention provides a method of promoting a bone morphogenetic protein or transforming growth factor β activation pathway in a cell, the method comprising:
There is provided the above method, which comprises suppressing the expression of f.

Furthermore, the present discovery allows screening of modulators of Smurf activity. The screening of the present invention comprises detecting modulation of host cell Smurf activity in the presence of the test compound as compared to host cell Smurf activity in the absence of the test compound. As shown in the examples, one such activity is ubiquitination of Smad proteins. Another activity is the promotion of TGFβ receptor degradation.

[0019] The present invention relates to a family of genes encoding E3 ubiquitin ligase called Smurf proteins, the present invention is derived from any animal, particularly from a mammal or amphibian, especially the origin of human And full-length, ie naturally occurring forms, and fragments thereof having the same functional activity or the same antigenicity. In particular embodiments, Smurf1 and Smurf2
Characterization of an E3 ubiquitin ligase called

The present invention is based, in part, on the surprising discovery that two new members of the Hect family of ubiquitin ligases interact with Smads. Their ligands are named Smurf1 and Smurf2. One of these, Smur
f1 specifically targets Smads specific for the BMP pathway, thereby acting as an antagonist or negative regulator of BMP signaling. Smurf1 interacts directly with Smad1 and 5 and regulates their ubiquitination, turnover and activity. In amphibian embryos, Smurf1 inhibits endogenous BMP signaling, resulting in altered patterning and cell fate in mesoderm and ectoderm. Thus, the present invention provides a unique regulatory protein link between the ubiquitination pathway and the control of cell fate, for example during embryogenesis. The invention further provides a nucleic acid encoding Smurf1 [SEQ ID NO: 1] in addition to the Smurf1 protein [SEQ ID NO: 2] and mutants thereof.

Another new member of the Hect family of ubiquitin ligases is Smurf2. Smurf2, the C2-WW-HECT domain E3 ubiquitin ligase, is associated with Smurf1. Smurf2 does not interact with Smad1, Smad2 or Smad4 and Smurf2 does not
Does not change the steady-state level of d1, Smad 2 or Smad 4. But Smurf2 is Smad
It interacts with 7 and binds directly to the PPXY motif in Smad7. Smurf2 is TGFβ
It acts synergistically with Smad7 as a signal transduction antagonist or negative regulator to participate in the degradation of the TGFβ receptor. Activation of TGFβ signaling results in Smad7-dependent assembly of Smurf2 into the TGFβ receptor complex. In the absence of an activated TGFβ receptor complex, Smurf2 does not alter Smad7 steady-state levels and turnover. Assembly of Smurf2 on the TGFβ receptor by Smad7 promotes degradation of the Smad7-TGFβ receptor complex by both the proteasome and lysosomal pathways. The studies described herein show that as an adapter protein for Smad7, Smurf2 assembles into the TGFβ receptor complex, promoting degradation of the complex and thereby down-regulating the activated TGFβ receptor complex. Shows the function of. Localization of Smad7 to the nucleus and regulation of its interaction with Smurf2 can be used to regulate the inhibitory activity of the Smad7-Smurf2 complex.

The invention further provides a nucleic acid encoding Smurf2 [SEQ ID NO: 3], in addition to the Smurf2 protein [SEQ ID NO: 4] and variants thereof.

The E3 ubiquitin ligase exhibits very selective substrate specificity, as evidenced by the findings discussed herein. For example, Smurf1 binds to the BMP pathway-specific Smad. Furthermore, Smurf1 is a unique signaling protein in the BMP pathway, which is S
murf1 binds only to Smad1 and Smad5, shows negligible affinity for Smad2 (specific for the TGFβ and activin receptor pathways), and no affinity for Smad4 (common Smad signaling partner) Because there is no. As a result of such substrate specificity, Smurf1 can effectively interfere with or regulate the biological response to BMPs without affecting the activin pathway, which means that Has limited or no effect on the TGFβ signaling pathway.

The BMP / TGFβ signaling pathway functions in controlling tissue differentiation, morphogenesis, and cell proliferation (eg (52)). As an antagonist to the signaling pathway mediated by the TGFβ family, Smurf1 induces the BMP pathway in vivo
Alternatively, it will be inhibited in vitro. As an essential component of the regulatory system for TGFβ receptor degradation, Smurf2 will inhibit the TGFβ signaling pathway. As a result, Smurf is associated with chondrogenesis, bone formation, blood cell differentiation, cartilage formation, neural tube patterning, retinal development, cardiac induction and morphogenesis, hair growth, tooth formation, gametes. It blocks formation and a wide range of tissue and organogenesis processes and interferes with the regeneration, growth, maintenance, etc. of bone and other tissues that rely on the BMP pathway.

In one embodiment, a variant of the Smurf protein or a small molecule antagonist of Smurf described below is used to prevent ubiquitination of a protein, eg, a Smad or TGFβ receptor, thereby mediated by a BMP-mediated signaling pathway. Can be used to maintain Furthermore, the Hect domain of the Smurf protein, which has the catalytic activity required for Smad ubiquitination, binds to either this domain or to the WW domain that interacts with the PPXY domain of Smad, This makes it possible to create a fragment of Smad, which eliminates the binding of Smurf1 to Smad and ubiquitination of Smad. These fragments can also be used as small molecule inhibitors of Smurf-Smad interactions, for example in inhibition binding assays. Variants of Smurf proteins and fragments of Smads can be used to improve defective BMP and / or TGFβ / activin signaling contributing to disease states as a result of overexpression or increased Smurf activity .

BMP regulates bone differentiation and connective tissue repair, and clinical trials have already been conducted in indications related to bone growth and connective tissue repair. The Smurf protein can influence its function, thereby affecting the cellular response to BMP, thus presenting a new target for the discovery of clinically applicable drugs.
Therefore, the Smurf protein can be used for screening various drugs and / or antibodies that can either enhance the BMP pathway or inhibit Smurf protein activity by antagonizing or mimicking. For example, since Smurf1 is highly specific for Smad1 and 5, it can be used to screen for drugs that block or activate the BMP pathway, and to the cellular response to BMPs against other TGFβs. It can be effective without affecting superfamily members.

The Smurf protein acts intracellularly at the level at which Smad signaling occurs, thus providing another way to influence BMP and TGFβ / activin signaling. However, since the Smurf protein is an intracellular protein, all manipulations that directly change Smurf activity must function in the cell. Such manipulations can include antisense and ribozyme techniques, as well as intracellular antibody techniques.

Smurf can be delivered intracellularly by gene therapy methods to block excessive signaling by specific growth factor pathways regulated by Smads, for example Smads targeted by Smurf1 or Smurf2. You can The examples herein show that merely increasing the expression level of Smurf1 in cells antagonizes the Smad signaling pathway. Overexpression of Smurf by gene therapy can be used to ameliorate clinical conditions resulting from excessive Smad signaling. Such clinical conditions include, for example, bone, tendon, or cartilage tissue hyperplasia, or other conditions that are regulated by signals from receptors that use Smad1 or Smad5 for signal transduction (such as BMP receptors). Formation of the tissue of.

Smurf nucleic acid and its subsequences (such as PCR probes) are defective in the human genome
It will be useful as a molecular probe for the identification of the Smurf gene, especially when mutations of the Smurf gene are found with particular diseases. The Smurf protein can be used as a reagent for in vitro assays to identify intracellular proteins that are targets for ubiquitination. Purified Smurf was reconstituted with purified ubiquitinase (i.e., E1 and E2 components) to introduce a new target protein into the assay (purified protein, or translated with unknown identity). As a cDNA, it can be used in a functional (ubiquitination) assay for identification purposes.

The present invention may employ conventional molecular biology, microbiology, and recombinant DNA techniques used in the art. Such techniques are explained fully in the literature. For example, Sambrook, Fritch and Maniatis, Molecular Cloning: A Laboratory Manual, 2nd Edition (1989), Cold Spr.
ing Harbor Laboratory Press, Cold Spring Harbor, New York, USA (hereafter
"Sambrook et al., 1989"); DNA cloning: a practical approach DNA Clonin
g: A Practical Approach, Volumes I and II (DNGlover, 1985); Oligonucleotide Synthesis (MJGait, 1984); Nucleic Acid Hybridization (BD Hames and SJ Higgins, 1985)
)]; Transcription And Translation [Edited by BD Hames and SJ Higgins (198
4)]; Animal Cell Culture [RIFreshney, (1986)]; Immobilized cells and enzyme Immobi
lized Cells And Enzymes [IRL Press, (1986)]; B. Perbal, A Practical Guide To Molecular Cloning (1984); FM Ausbel et al.
(Ed), Current Protocols in Molecular biolo
See gy, John Wiley & Sons, Inc. (1994).

[0031]   As used herein, the following terms have the following meanings.

The term “about” or “approximately” means within 20%, preferably within 10%, and more preferably within 5% of a given value or range.

As used herein, the term “isolated” means free of components other than that substance normally found in the natural environment in which it is normally found. In particular, the isolated biological material is free of cellular components. In the case of nucleic acid molecules, isolated nucleic acids include PCR products, isolated mRNA, cDNA, or restriction enzyme cleavage fragments. In another embodiment, the isolated nucleic acid is preferably excised from the chromosome in which the nucleic acid is found, more preferably the nucleic acid is that the isolated nucleic acid molecule is intrachromosomal. In some cases, it is a non-regulatory region, a non-coding region located upstream or downstream of the gene contained in the nucleic acid, or is not linked to another gene. In yet another embodiment, the isolated nucleic acid lacks one or more introns. The isolated nucleic acid molecule can be inserted into plasmids, cosmids, artificial chromosomes, and the like. Thus, in one particular embodiment, the recombinant nucleic acid is one of the isolated nucleic acids. The isolated protein can be associated with other proteins or nucleic acids with which the protein is associated in the cell,
Alternatively, it can be associated with the cell membrane if it is a membrane-associated protein. The isolated intracellular organelle, cell, or tissue is removed from the anatomical site in the body where it is found. The isolated material can be, but need not be, purified.

As used herein, the term “purified” refers to irrelevant material, ie, material isolated under conditions that reduce or eliminate contaminants. For example, a purified protein is preferably one that is substantially free of other proteins or nucleic acids with which it is associated in the cell, and the purified nucleic acid molecule is preferably found in the cell with that nucleic acid. It is substantially free of proteins or other unrelated nucleic acid molecules. The term "substantially free" as used herein is used in the context of analytical testing of substances. Preferably, the purified material substantially free of contaminants is at least 50% pure, more preferably at least 90% pure, and even more preferably at least 99% pure. Purity includes chromatography, gel electrophoresis,
It can be evaluated by immunoassays, compositional analysis, biological assays, and other methods known in the art.

As used herein, a “gene” refers to a portion of a DNA molecule that contains a polypeptide coding sequence operably linked to expression control sequences. In one embodiment, a gene can be a genomic sequence or part of a genomic sequence, which sequence contains one or more introns. In another embodiment, the term gene means a cDNA molecule (ie a coding sequence lacking introns).

A DNA “coding sequence” is a double-stranded DNA sequence which is transcribed and translated into a polypeptide in a cell in vitro or in vivo when under the control of appropriate regulatory sequences. The boundaries of the coding sequence are the start codon at the 5'end (amino terminus) and the 3'end (
Defined by the translation stop codon at the carboxy terminus). Coding sequences also include, but are not limited to, prokaryotic sequences, eukaryotic mRNA derived cDNA, eukaryotic (eg, mammalian) DNA, and synthetic DNA sequences. If the coding sequence is intended for expression in eukaryotic cells, a polyadenylation signal and transcription termination sequence will usually be located 3'to the coding sequence.

“Expression control sequences”, such as transcriptional and translational control sequences, are regulatory sequences that flank the coding sequence, such as promoters, enhancers, suppressors, terminators, and the like, the regulatory sequences of which are: Allows expression of the coding sequence in the host cell. In eukaryotic cells, polyadenylation signals are regulatory sequences. On mRNA, the ribosome binding site is an expression control sequence.

A “promoter sequence” is a DNA regulatory region capable of binding RNA polymerase in a cell and initiating transcription of a downstream (3 ′ direction) coding sequence. For the purposes of defining the present invention, a promoter sequence is bounded at its 3'end by a transcription start site, upstream (5 'direction) to initiate a detectable level of transcription above background levels. Extend to contain the required minimum number of bases or elements. Within the promoter sequence is a transcription initiation site (eg nuclease S1
, Which can be conveniently defined by mapping with, and the protein-binding domain (consensus sequence) that is responsible for binding RNA polymerase.
There is.

[0039] A coding sequence is transcribed into a protein encoded by the coding sequence by RNA polymerase in the cell which transcribes the coding sequence into mRNA, which is then spliced into tRNA (if the mRNA contains an intron). "Under the control" of transcriptional and translational regulatory sequences, when made.

One nucleic acid molecule is capable of being annealed to another nucleic acid molecule if one strand of the nucleic acid molecule can anneal to the other nucleic acid molecule under conditions of suitable temperature and ionic strength of the solution.
For example, it is "hybridizable" with cDNA, genomic DNA, or RNA (Sambroo
k, et al., supra). Conditions of temperature and ionic strength determine the "stringency" of hybridization. Low stringency hybridization conditions with a Tm of 55 ° C to preliminarily screen for homologous nucleic acids
Corresponding conditions are used (eg 5 x SSC, 0.1% SDS, 0.25% milk, and formamide free; or 30% formamide, 5 x SSC, 0.5% SDS.
). Moderate stringency hybridization conditions correspond to higher Tm values, eg, 5x or 6x SSC with 40% formamide. High stringency hybridization conditions correspond to the highest Tm values, eg
50% formamide, 5x or 6x SSC. Hybridization requires two nucleic acids containing complementary sequences, although mismatches between bases may occur depending on the stringency of the hybridization. The appropriate stringency for hybridizing nucleic acids depends on the length of the nucleic acids and the degree of complementation, variables known in the art. The greater the degree of similarity or homology between two nucleic acid sequences, the greater the Tm value for a hybrid of nucleic acids having those sequences. Relative stability of nucleic acid hybridization (
Corresponding to higher Tm) decreases in the following order: RNA: RNA, DNA: RNA, DNA: DNA. For hybrids greater than 100 nucleotides in length, the equation for calculating the Tm has been derived (see Sambrook et al., Supra, 9.50-9.51). Shorter nucleic acids,
That is, the position of the mismatch becomes more important for hybridization with the oligonucleotide, and the length of the oligonucleotide determines its specificity (S
ambrook et al., supra, 11.7-11.8). The minimum length of nucleic acid that can be hybridized is at least about 10 nucleotides; preferably at least about 15 nucleotides; and more preferably at least about 20 nucleotides.
It is the length of the individual.

In one particular embodiment, “standard hybridization conditions” means 55.
The Tm of ° C and the above-mentioned conditions are used. In a preferred embodiment, the Tm is 60 ° C; in a more preferred embodiment the Tm is 65 ° C. specific
In one embodiment, "high stringency" refers to hybridization and / or wash conditions of 0.2 x SSC at 68 ° C, 50% formamide at 42 ° C, 4 x.
It is a condition that gives a hybridization level equivalent to that observed in either SSC or these two conditions.

A “vector” is a recombinant nucleic acid construct such as a plasmid, phage genome, viral genome, cosmid, or artificial chromosome to which another DNA segment is attached. In one particular embodiment, the vector, in the case of a cloning vector, results in replication of the linked segments. A segment of DNA is inserted at a specific restriction enzyme cleavage site in the vector. The DNA segment encodes the polypeptide of interest, and the segment and restriction enzyme cleavage site are designed so that the segment is inserted in the proper reading frame for transcription and translation.

A cell is exogenous or heterologous DNA and is said to be “transfected” if the DNA is introduced inside the cell. A cell is exogenous or heterologous DNA and is "transformed" when the transfected DNA effects a phenotypic change. Preferably, the transforming DNA will be integrated (covalently linked) into the chromosomal DNA to form the cell's genome.

The term “heterologous” means a combination of elements that does not occur in nature.
For example, heterologous DNA means DNA that is not naturally located within a cell or within a chromosomal site of a cell. Preferably, the heterologous DNA contains a gene foreign to the cell. A heterologous expression regulatory element is an element that is operably associated with a gene that differs from the one that naturally associates with that element in its native state. In the present invention, the Smurf1 and Smurf2 genes are heterologous to the plasmid vector DNA into which the genes are inserted for cloning or expression, and to non-human host cells in which the genes are expressed, such as CHO cells. Different.

A “nucleic acid molecule” is a ribonucleoside (adenosine, guanosine, uridine,
Or a cytidine; “RNA molecule”), or a phosphoric acid ester polymer of deoxyribonucleoside (deoxyadenosine, deoxyguanosine, deoxythymidine, or deoxycytidine; “DNA molecule”), or a phosphoric acid analog thereof, such as phosphorothioate and By any such as thioester, either in single-stranded or double-stranded form, or double-stranded helix. It can take the form of double-stranded DNA-DNA, DNA-RNA, and RNA-RNA helices. The term nucleic acid molecule, especially the DNA or RNA molecule, refers only to the primary and secondary structure of the molecule and is not limited to any particular tertiary structure. Thus, this term includes double-stranded DNA, especially those found in linear or circular DNA molecules (eg, restriction enzyme fragments), plasmids, and chromosomes. In discussing the structure of a particular double-stranded DNA molecule, its sequence is in the 5'to 3'direction in the untranscribed strand of DNA (i.e., the strand having a sequence homologous to mRNA), in accordance with normal convention. Only the aligned sequences are listed. A "recombinant DNA molecule" is a DNA molecule that has undergone a molecular biological manipulation.

The present invention provides antisense nucleic acids, including ribozymes, which include
It can be used to inhibit the expression of Smurf1 in order to enhance the BMP pathway via Smad1 and Smad4. Thus, antisense nucleic acids corresponding to the Smurf1 gene or fragments thereof can be used to alter the BMP pathway. The present invention also includes TGF
Antisense nucleic acids that inhibit Smurf2 expression are also provided to enhance the β signaling pathway. An "antisense nucleic acid" is a single-stranded nucleic acid molecule that inhibits the latter role when hybridized with complementary bases in an RNA or DNA molecule. When the RNA is a messenger RNA transcript, the antisense nucleic acid is a counter transcript, or mRNA interfering complementary nucleic acid. As currently used, "antisense" broadly includes RNA-RNA interactions, RNA-DNA interactions, ribozymes, and RNase-H mediate arrest. Antisense nucleic acid molecules can be encoded by recombinant genes for expression in cells (e.g., U.S. Pat.No. 5,814,500; U.S. Pat.No. 5,811,234) or can also be prepared synthetically (e.g., U.S. Pat. (Patent No. 5,780,607).

The term “oligonucleotide” as used herein refers to a nucleic acid that is usually at least 10 nucleotides, preferably at least 15, more preferably at least 20 and is a genomic DNA molecule, a cDNA. It means a molecule, or an mRNA molecule encoding a gene, mRNA, cDNA, or any other hybridizable nucleic acid of interest. Oligonucleotides include, for example, ³² P-nucleotides,
Alternatively, it can be labeled using a nucleotide to which a label such as biotin is covalently attached. In one embodiment, labeled oligonucleotides can be used as probes to detect nucleic acids. In another embodiment, 2
One oligonucleotide (which can be labeled with one or both) is used as a primer for PCR to clone either full-length Smurf1 or Smurf2 or a fragment, or to detect the presence of a nucleic acid encoding Smurf1 or Smurf2. Can be used. In yet another embodiment, the oligonucleotides of the invention are capable of forming triple helices with Smurf1 or Smurf2 DNA molecules. In yet another embodiment, a library of oligonucleotides prepared on a solid support, such as a silicon wafer or chip, can be used to detect various polymorphisms of interest. Usually, oligonucleotides are made synthetically, preferably on a nucleic acid synthesizer. Accordingly, oligonucleotides can be prepared using non-naturally occurring phosphate ester analog linkages, such as thioester linkages, and the like.

Specific examples of synthetic oligonucleotides contemplated by the invention include phosphorothioates, phosphotriesters, methylphosphonates, short chain alkyl or cycloalkyl intersugar linkages, or short chain heteroatoms or heterocyclics. Inter-sugar linkage is mentioned. Most preferred are CH2-NH-O-CH2, CH2-N (CH3) -O-CH2, CH2-ON (CH3) -CH2, CH2.
-N (CH3) -N (CH3) -CH2 and ON (CH3) -CH2-CH2 skeleton (where the phosphodiester is O-PO2-O-CH2). US Pat. No. 5,677,437 describes heteroatomic oligonucleotide linkages. Nitrogen linkers or groups containing nitrogen can also be used to prepare oligonucleotide mimetics (US Pat. Nos. 5,792,844 and 5,783,682). U.S. Pat.No. 5,637,684 is p
Hosphoramidate and phosphorothioamidate oligomeric compounds are described. Oligonucleotides having a morpholino backbone structure are also contemplated (US Pat. No. 5,034,506). In another embodiment, the phosphodiester backbone of an oligonucleotide, such as a peptide-nucleic acid (PNA), can be replaced with a polyamide backbone in which a base is directly or indirectly attached to the aza nitrogen atom of the polyamide backbone. Yes (Nielsen et al., Science 254: 1497, 1991). Other synthetic oligonucleotides can contain substituted sugar moieties at their 2'positions including one of the following: OH, SH, SCH3, F, OCN, O (CH2) nNH2, O ( CH2) nCH3, where n is 1 to about 1
0; lower alkyl of C1 to C10, substituted lower alkyl, alkaryl, or aralkyl; Cl; Br; CN; CF3; 0-, S-, or N-alkenyl; SOCH3
SO2CH3; ONO2; NO2; N3; NH2; heterocycloalkyl; heterocycloalkaryl; aminoalkylamino; polyalkylamino; substituted silyl; fluorescent moiety; RNA cleavage group; reporter group; intercalator; oligonucleotide drug Groups that improve the kinetic properties; or groups that improve the pharmacodynamic properties of the oligonucleotide, and substitutes with other similar properties. Oligonucleotides can also have sugar mimetics such as cyclobutyl or carbocyclic compounds in place of the pentofuranosyl group. Nucleotide units having nucleosides other than adenosine, cytidine, guanosine, thymidine, and uridine, such as inosine, can be used.

The term “homologous” as used herein, including all of its grammatical forms and spelling differences, refers to relationships between proteins that have a “common evolutionary ancestor”, It includes proteins from the superfamily (e.g. TGFβ superfamily) and homologous proteins from different species (e.g. Smad (human), Mad (Drosophila), etc.) (Reeck et al., Cell 50: 667, 1987). I'm out. Such proteins, and the genes encoding them, have sequence homology,
This is reflected in the high degree of similarity in their sequences.

Accordingly, the term “sequence similarity”, including all of its grammatical forms, encodes between or among the amino acids of proteins that have or do not have a common evolutionary ancestor. Indicates the degree of identity or correspondence between nucleic acids (Reeck
Et al., Cited above). However, in general usage and in this application, the term “homologous” means sequence similarity when modified by welfare such as “highly” and is not associated with a common evolutionary ancestor. It may or may not be.

In one particular embodiment, the two DNA sequences have at least about 70-75% of them.
Most preferably, at least about 80-85% of the nucleotides are “substantially homologous” or “substantially similar” if they match in a defined length DNA sequence. One example of such a sequence is the Smurf of the invention, allelic or by species.
There is a mutation in the gene. Substantially homologous sequences may be compared by comparing those sequences in a sequence databank using available standard software, or by performing Southern hybridization experiments such as the stringent string defined in that particular system. It can be identified by performing under conditions. Defining appropriate hybridization conditions is within the skill of the art. For example, Maniatis
Et al., Supra; DNA Cloning, Volumes I and II, supra .; Nucleic Acid H
ybridization, see references above.

Similarly, in certain embodiments, two amino acid sequences, if more than 70% of the amino acids are identical or about 90% are similar (functionally similar), Two amino acid sequences are "substantially homologous" or "substantially similar." Preferably, similar or homologous sequences have, for example, GCG (Genetics Comput
er Group, Program Manual for the GCG Package, Version 7, Madison, Wisconsin
) Pileup program, BLAST, and Clustal W analysis (MacVector
) Is used for the alignment. The sequence comparison algorithm is also <
http://www.nwfsc.noaa.gov/bioinformatics.html>.

The term “corresponds to” as used herein refers to sequences that are similar or homologous, whether the exact position is the same or different as the molecule whose similarity or homology is being measured. About. Nucleic acid or amino acid alignments may include voids. Thus, the term "corresponding to" refers to sequence similarity and not to amino acid residue or nucleobase position numbers.

“Homologous recombination” refers to the insertion of the foreign DNA sequence of one vector into the chromosome.
. Preferably, the vector targets a specific chromosomal site for homologous recombination. For specific homologous recombination, the vector shall contain regions of homology to the sequences of the chromosome long enough to allow complementary binding and integration of the vector into the chromosome. The longer the region of homology and the greater the degree of sequence similarity, the greater the effectiveness of homologous recombination.

A DNA “coding sequence” is an in vitro sequence when placed under the control of appropriate regulatory sequences.
A double-stranded DNA sequence that is transcribed and translated into a polypeptide in a cell, o or in vivo. The boundaries of the coding sequence are defined by the start codon at the 5 '(amino) terminus and the 3'
Determined by the translation stop codon at the '(carboxyl) terminus. A coding sequence includes, but is not limited to, prokaryotic sequences, cDNA from eukaryotic mRNA, genomic DNA sequences from eukaryotic (eg mammalian) DNA, and even synthetic DNA sequences. When assuming a coding sequence for expression in eukaryotic cells, a polyadenylation signal and transcription termination sequence will usually be located 3'to the coding sequence.

Transcriptional and translational control sequences are DNA regulatory sequences, such as promoters, enhancers, terminators, which are arranged for expression of coding sequences in host cells. In eukaryotic cells, polyadenylation signals are regulatory sequences.

The term “promoter sequence” refers to the downstream (3 ′
Direction) is a DNA regulatory region capable of initiating transcription of the coding sequence. For the purpose of defining the present invention, the promoter sequence is bounded at its 3'end by a transcription start site and is the minimum number required to initiate transcription at a detectable level above background values. Extending upstream (5 'direction) so as to include the base or element. Within the promoter sequence is a transcription start site (conveniently established by mapping with, for example, nuclease S1), as well as a protein binding domain (consensus sequence) involved in RNA polymerase binding.

When RNA polymerase transcribes a coding sequence into mRNA, the coding sequence is “under the control” of transcriptional and translational control sequences in the cell, which is then transRNA spliced and translated into the coding sequence. Becomes a protein encoded by.

Nucleic acids encoding the Smurf family of proteins, eg Smurf1 or Smurf2, can be isolated from any source, especially human or Xenopus cDNA or genomic libraries, either genomic DNA or cDNA. it can. Methods for obtaining genes are well known in the art as described above (
See, eg, Sambrook et al., 1989, supra). In a particular embodiment, the invention provides the human Smurf1 (hSmurf1) and Smurf2 (hSmurf2) genes [SEQ ID NO: 1 and SEQ ID NO: 3].

Thus, any animal cell may serve as a nucleic acid source for the molecular cloning of the Smurf gene. The DNA can be obtained from cloned DNA (eg, DNA "libraries") by standard techniques known in the art. These include EST libraries and cDNA libraries prepared from tissues expressing high levels of this protein (eg, Xenopus stage 9 (blastocyst) cDNA library-this is the cell that shows the highest level of Smurf1 expression). Is included. Smu
Other cell lines that appear to express rf1 or Smurf2 are ectoderm, mesoderm and endoderm of frog and cyst embryos; mouse embryonic stem cells; and various mammalian cells such as NIH3T3, PC12, 293T, Hela and Cos Is. DNA encoding the Smurf protein can also be obtained by chemical synthesis, cDNA cloning, or cloning genomic DNA or fragments thereof purified from the desired cells (eg, Sambrook et al., 1989, supra; Glover, DM (ed .), 1985, DNA Cloning: A Practi
cal Approach, MRL Press, Ltd., Oxford, UK Vol. I, II). Clones derived from genomic DNA may contain regulatory and intron DNA regions in addition to the coding region, whereas clones derived from cDNA do not contain intron sequences. Regardless of origin, for gene propagation the gene must be molecularly cloned into a suitable vector.

Identification of a particular DNA fragment containing the desired Smurf gene can be accomplished by a variety of methods known in the art. For example, a portion of the Smurf gene exemplified below can be purified and labeled to prepare a labeled probe, and the prepared DNA can be screened by nucleic acid hybridization to this labeled probe (Benton and Davis, Science 196: 180,1977; Grunstein and Hogness, Proc.
Natl.Acad.Sci.USA72: 3961,1975). A DNA fragment substantially homologous to the probe,
An allelic variant or the like from another individual will hybridize. In one particular embodiment, high stringency hybridization conditions are used to identify homologous Smurf1 and Smurf2 genes.

The present invention also provides a cloning vector containing a gene encoding the Smurf gene of the present invention, for example, the analogs and derivatives of Smurf1 or Smurf2, which have similar or homologous functional activities. Concerned. Derivatives and analogs related to Smurf1 and Smurf2 are within the scope of the invention. For example, a truncated version of Smurf1 or Smurf2 can be provided. Such truncated forms include Smurf1 or Smurf2 with deletions. In a particular embodiment, the derivative or analog is functionally active, ie capable of exhibiting one or more functional activities associated with the full length wild type Smurf1 or Smurf2 of the invention.
These functions include binding to Smad proteins such as Smad1, Smad5 or Smad7, as well as catalysis of bound Smads, activated TGFβ receptor / Smad complexes, or ubiquitination of other proteins in the TGFβ / activin pathway. The action is included. In another embodiment, the fragment has binding affinity but lacks or has reduced catalytic capacity.

Smurf derivatives can be made by altering the encoding nucleic acid sequence by substitutions, additions or deletions that provide a functionally equivalent molecule. Alternatively,
For use in treating diseases associated with weakened BMP signaling pathways such as those described above, which compete with wild-type Smuef protein in the BMP pathway,
Non-functional mutant forms of the Smurf protein can be prepared that are less effective in class ubiquitination. In one particular embodiment below, the mutant is C710A, which is further described in the Examples. In another specific embodiment for Smurf2, the mutant is C716A.

Based on the properties of the gene, for example, the gene may have an isoelectric point, electrophoretic property, amino acid composition, partial or complete amino acid sequence, antibody binding activity, or ligand binding profile of the 1Smurf protein disclosed herein. Further screening can be performed as to whether it encodes a protein product it has. Thus, the presence of that gene can be detected by assays based on the physical, chemical, immunological, or functional properties of the expressed product. For example, a cDNA clone, or a DNA clone that hybridizes to the appropriate mRNA, such as Smurf
An electrophoretic mobility similar or identical to that measured for 1 or Smurf2,
Those that produce proteins with isoelectric focusing or non-equilibrium pH gel electrophoresis behavior, proteolytic digestion maps, or antigenic properties can be selected.

The present invention also relates to analogs and derivatives of Smurf proteins, homologues from another species, and mutants having the same or homologous functional activity. The production and use of derivatives, analogs and mutants related to Smurf protein is within the scope of the present invention. In a particular embodiment, the derivative or analog is functionally active, ie full length wild type Smurf1 or
It is capable of presenting one or more functional activities related to Smurf2.

Smurf derivatives can be made by altering the encoding nucleic acid sequence by substitutions, additions or deletions that provide a functionally equivalent molecule. Preferably, it has an enhanced functional activity as compared to native Smurf, for example Smurf1 or Smurf2, or lacks a functional activity such as the catalytic activity in the Hect ubiquitin ligase domain of the C-terminal portion of the Smurf protein. To produce the derivative. In a particular embodiment, hSmurf1, a mutant with a point mutation in C710A, which eliminated ubiquitination of Smad1 and 4 by disrupting the catalytic activity of the Hect domain.
I will provide a. In another particular embodiment, disrupting the catalytic activity of the Hect domain eliminated proteolytic disruption of the TGFβ receptor-Smad7 complex, C
We provide hSmurf2, a mutant with a point mutation in 716A. Alternatively, the Smurf protein derivative encodes a soluble fragment of one of the Smurf protein domains, eg the WW domain, and has the same or higher affinity for the natural ligand, eg Smad1, 5 or 7. Such soluble derivatives could be potent inhibitors of the binding of ligand (ie Smad1, 5 or 7) to Smurf protein.

In another particular embodiment, the E3 ligase binds to Smurf1 and the E3 ligase binds to cellular Sm.
Derivatives or fragments of Smad1 and 4 can be generated that eliminate further binding to ad and suppress its ubiquitination. In another embodiment, S
Derivatives or fragments of Smad7 can be generated that reduce binding of murf2 to both Smad7 and TGFβ receptors. For example, in one embodiment, the present invention provides
Conserved motifs recognized by WW domains, such as those found in Smurf1 or Smurf2 (Rotin, Curr.topics Microbiol.Immunol., 228: 115-133, 1998, and (
33))) and the peptide containing the linker region of R-Smad with the PPXY sequence and the corresponding nucleic acid sequence are envisioned.

Due to the degeneracy of the nucleotide coding sequence, other DNA sequences that encode an amino acid sequence substantially identical to one of the Smurf genes may also be used in the practice of the invention. These include, but are not limited to, alleles, homologous genes from another species, as well as nucleotide sequences containing all or part of the Smurf gene, within another sequence that encodes the same amino acid residue. Altered to produce a silent conversion by codon substitution. Similarly, the invention
The Smurf gene includes, but is not limited to, all or part of the amino acid sequence of the Smurf protein as a primary amino acid sequence, and residues within the sequence are replaced with functionally equivalent amino acid residues. And the resulting modified sequence resulting in a conservative amino acid substitution. For example, one or more amino acid residues in the sequence
Substitution with another amino acid of similar polarity, which acts as a functional equivalent, can result in a silent alteration. Substitutions of amino acids within the sequence can be selected from other members of the class to which the amino acid belongs. For example,
Non-polar (hydrophobic) amino acids include alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan and methionine.
Amino acids containing aromatic ring structures are phenylalanine, tryptophan and tyrosine. Polar neutral amino acids include glycine, serine, threonine, cysteine, tyrosine, asparagine and glutamine. Positively charged (basic) amino acids include arginine, lysine and histidine. Negatively charged (acidic) amino acids include aspartic acid and glutamic acid. Such changes are not expected to affect the apparent molecular weight determined by polyacrylamide gel electrophoresis or isoelectric point. Particularly preferred substitutions include:-A positive charge is maintained, Arg is Lys, or vice versa;-Negative charge is maintained, Asp is Glu, or vice versa;-Free OH is maintained Thr to Ser; and-free CONH ₂ is maintained, Asn to Gln.

Amino acid substitutions can also be introduced to replace amino acids with particularly favorable properties. For example, Cys can be introduced at a site that may disulfide bridge with another Cys. In particular, His can be introduced as a “catalytic” site (ie His acts as an acid or base and is the most common amino acid in biochemical catalysis). Due to its particularly flat structure, Pro can be introduced to induce β-turns in the protein structure.

Genes encoding the Smurf derivatives and analogs of the invention can be produced by various methods known in the art. The operations leading to its production occur at the gene or protein level. For example, the cloned Smurf1 or Smurf2 gene sequences can be modified by any of the numerous strategies known in the art (Sambrook et al., 1989, supra). Cleavage with restriction endonucleases at appropriate sites in the sequence, followed by further enzymatic modification if desired, isolation, and in vitro
Can be connected with. In the production of genes encoding Smurf derivatives or analogs, the modified gene remains in the same translational reading frame as the gene and the desired activity is blocked by a translational stop signal within the encoded gene region. Care should be taken to ensure that it does not.

Furthermore, in order to disrupt and / or create translation, initiation, and / or termination sequences, or to create variants within the coding region and / or to create new restriction endonuclease sites or in advance. Smurf to destroy further restriction endonuclease sites and facilitate further in vitro modifications.
The nucleic acid sequence encoding the can be mutated in vitro or in vivo. Preferably, such mutants enhance the functional activity of the mutant Smurf gene product. Any known technique for mutagenesis can be used. This includes in vitro site-directed mutagenesis (Hutchinson, C., et al., 1978, J. Biol. Che.
m. 253: 6551; Zoller and Smith, 1984, DNA 3: 4790488; Oliphant et al., 1986,
Gene 44: 177; Hutchinson et al., 1986, Proc. Natl. Acad. Sci. USA 83: 710), TA.
Includes, but is not limited to, the use of B linkers (Parmacia) and the like. It is preferable to use PCR technology for site-directed mutagenesis (Higuchi, 1989, "Using PCR to En
gineer DNA "PCR Technology Medium: Principles and Applications for DNA Amplif
ication, edited by H. Erlich, Stockkton Press, Chapter 6, pp. 61-70).

The identified and isolated gene can then be inserted into a suitable cloning vector. Many vector-host systems known in the art can be used. Available vectors include plasmids or modified viruses,
There is no limitation. However, the vector system must be compatible with the host cell used. Examples of vectors include, but are not limited to, plasmids such as E. coli bacteriophage, such as lambda derivatives, or pBR322 derivatives or pUC plasmid derivatives (e.g., pGEX vector, pMal-c, pFLAG, pGBT9, etc.). It is not done. Insertion into a cloning vector is carried out, for example, by ligating a DNA fragment into a cloning vector having complementary overhanging ends. However, if the complementary restriction enzyme sites used to fragment the DNA are not present in the cloning vector, the ends of the DNA molecule may be enzymatically modified. Alternatively, a linking nucleotide sequence (linker) can be ligated to the end of DNA to create any desired site, these linked linkers being specially chemically synthesized to encode a restriction endonuclease recognition sequence. It may include an oligonucleotide. Recombinant molecules can be introduced into host cells by transformation, transfection, infection, electroporation, etc., which can produce multiple copies of the gene sequence. Preferably cells for cloning
A gene cloned into a shuttle vector plasmid is included that allows expression in (eg, E. coli) and, if desired, can be easily purified for subsequent introduction into a suitable expression cell line. For example, the shuttle vector is a vector capable of replicating in one or more kinds of organisms, and a sequence derived from E. coli plasmid and a sequence derived from yeast 2μ plasmid are linked to each other.
ces cerevisiae).

Expression of Smurf Polypeptides Nucleotide sequences encoding Smurf proteins, or antigenic fragments, derivatives or analogs thereof, or functionally active derivatives thereof, including chimeric proteins, can be expressed in any suitable expression vector, ie inserted protein code. It can be inserted into a vector containing the necessary elements for the transcription and translation of the sequence. Such elements are referred to herein as "promoters." Therefore, the Smur of the present invention
The nucleic acid encoding the f protein is operably linked to the promoter in the expression vector of the invention. Under the control of such regulatory sequences, both cDNA and genomic sequences can be cloned and expressed. Further, the expression vector preferably contains an origin of replication.

The necessary transcriptional and translational signals can be provided on the recombinant expression vector. Alternatively, these signals may be provided by the native gene encoding the Smurf protein and / or its flanking regions.

Possible host-vector systems include, but are not limited to the following examples, virus (eg, vaccinia virus, adenovirus, etc.) infected mammalian cell lines; virus (eg, baculovirus) infected insect cells. System; microorganisms such as yeast containing yeast vector; or bacteria transformed with bacteriophage, DNA, plasmid DNA or cosmid DNA. The expression elements of a vector vary in their strength and specificity. Any one of a number of suitable transcription and translation elements may be used, depending on the host-vector system utilized. Host cells containing the recombinant vector carrying the nucleic acid encoding the Smurf protein of the present invention are cultured in a suitable cell culture medium under conditions conferring expression by the cell. Host cells useful for the expression of Smurf include C ₂ C ₁₂ , 293T, CHO, COS,
There are HEK, Hela, HepG2, NIH3T3, PC12, P19 and other cell lines, kidney, brain, and osteocytes.

A recombinant Smurf protein of the invention, or a functional fragment, derivative thereof,
The chimeric construct or analog can be chromosomally expressed after recombinantly incorporating the coding sequence. In this regard, high levels of stable gene expression can be achieved using any of a number of amplification systems (see Sambrook et al., Supra, 1989).

Any of the methods described above for inserting a DNA fragment into a cloning vector can be used to construct an expression vector having a gene and protein coding sequence consisting of appropriate transcription / translation control signals. Examples of these methods include in vitro recombinant DNA and synthetic methods, and in vivo recombination (gene recombination).

Expression of the Smurf protein can be controlled by promoter / enhancer elements known in the art, but these regulatory elements must be functional in the host chosen for expression. Promoters that can be used to control Smurf gene expression include, but are not limited to, the SV40 early promoter region (Benoist and Chambon, 1981, Nature 290: 304-310), 3'of Rous sarcoma virus. Promoters contained in long terminal repeats (Yamamoto, et al., 198
0, Cell 22: 787-797), herpes thymidine kinase promoter (Wagner et al.,
1981, Proc. Natl. Acad. Sci. USA 78: 1441-1445), regulatory sequences of the metallothioinene gene (Brinster et al., 1982, Nature 296: 39-42); β-lactamase promoter (Villa-Kamaroff, et al., 1978, Proc. Natl. Acad. Sci. USA 75: 3727-3
731) and the lac promoter (DeBoer, et al., 1983, Proc. Natl. Acad. Sci. USA 8
0: 21-25) and other prokaryotic expression vectors; Scientific American, 1980, 242: 74-97.
See also "Useful Proteins from Recombinant Bacteria"; promoters from yeast or other fungi such as the Gal4 promoter, ADC (alcohol dehydrogenase) promoter, PGK (phosphoglycerol kinase) promoter, alkaline phosphatase promoter. Element; and animal transcriptional regulatory region showing tissue specificity and utilized in transgenic animals: elastase I gene regulatory region active in pancreatic gland cells (Swift et al., 1984, Cell 38: 639-646; O
rnitz et al., 1986, Cold Spring Harbor Symp. Quant. Biol. 50: 399-409; MacDon.
ald, 1987, Hepatology 7: 425-515); insulin gene regulatory region active in β cells of the pancreas (Hanahan, 1985, Nature 315: 115-122), immunoglobulin gene regulatory region active in lymphoid cells (Grosschedl et al. , 1984, Cell 38: 647-658
Adames et al., 1985, Nature 318: 533-538; Alexander et al., 1987, Mol. Cell Bi.
7: 1436-1444), mouse mammary tumor virus regulatory region active in testis, breast, lymphatic system and mast cells (Leder et al., 1986, Cell 45: 485-495), albumin gene regulation active in liver. Domain (Pinkert et al., 1987, Genes and Devel. 1
: 268-276), an α-fetoprotein gene regulatory region active in the liver (Krumlauf
Et al., 1985, Mol. Cell. Biol. 5: 1639-1648; Hammer et al., 1987, Science 235: 53-.
58), the α-antitrypsin gene regulatory region active in the liver (Kelsey et al., 1987,
Genes and Devel. 1: 161-171), β-globin gene regulatory region active in myeloid cells (Mogram et al., 1985, Nature 315: 338-340; Kollias et al., 1986, Cell 4).
6: 89-94), myelin basic protein gene regulatory region active in oligodendrocyte cells of the brain (Readheard et al., 1987, Cell 48: 703-712), myosin light chain-2 gene active in skeletal muscle. The regulatory region (Sani, 1985, Nature 314: 283-286) and the gonadotropin-releasing hormone gene regulatory region active in the hypothalamus (Mason et al., 1986, Science 234: 1).
372-1378).

A variety of host / expression vector combinations may be employed in expressing the DNA sequences of this invention. For example, useful expression vectors may be composed of segments (parts) of chromosomal, non-chromosomal and synthetic DNA sequences. Suitable vectors include derivatives of SV40 and known bacterial plasmids such as E. coli plasmids col E1, pCR1, pBR322, pMal-C2, pET, pGEX (Smith et al., 1988,
Gene 67: 31-40), pMB9 and its derivatives, plasmids such as RP4; phage DNA
, Many phage 1 derivatives, such as NM989, and other phage DNA,
For example, M13 and filamentous single-stranded phage DNA; yeast plasmids such as 2m plasmids or derivatives thereof; vectors useful in eukaryotic cells such as vectors useful in insect or mammalian cells; using phage DNA or other expression control sequences Vectors derived from a combination of plasmid and phage DNA, such as plasmids that have been modified for this purpose.

For example, in the baculovirus expression system, pVL941 (BamH1 cloning site; Summers), pVL1393 (BamH1, SmaI,
XbaI, EcoR1, NotI, XmaIII, BglII and PstI cloning sites; Invitrogen)
, PVL1392 (BglII, PstI, NotI, SmaIII, EcoRI, XbaI, SmaI and BamH1 cloning sites; Summers and Invitrogen), and pBlueBacIII (BamH1, BglII, Ps
tI, NcoI and HindIII cloning sites, non-fusion transfer vectors such as recombinant blue / white screenable; Invitrogen) and pAC700 (BamH1 and KpnI cloning sites, but not limited to the following examples The BamH1 recognition site begins at the start codon; Summers), pAc701 and pAc702 (similar to pAc700, but with different reading frames), pAc360 (36 base pairs BamH1 cloning site downstream of the polyhedrin start codon; Invitrogen (195). ) And pBlueBac
HisA, B, C (BamH1, BglII, PstI, NcoI and HindIII cloning sites, ProBo
Three different reading frames with N-terminal peptides for nd purification and blue / white recombinant screening of plaques; fusion transfer vectors such as Invitrogen (220) can be used.

Mammalian expression vectors used in the present invention include dihydrofolate reductase (DHFR) promoter, for example, an expression vector having a DHFR expression vector, or pED (PstI, SalI, SbaI, SmaI and EcoRI cloning sites). Having a vector expressing both the cloned gene and DHFR; Kaufman,
There are vectors with inducible promoters, such as the DHFR / methotrexate co-amplification vector (see Current Protocols in Molecular Biology, 16.12 (1991)). Alternatively, a glutamine synthetase / methionine sulfoxyimine co-amplification vector, such as pEE14 (which vector expresses glutamine synthase and the cloned gene, HindIII, XbaI, SmaI, SbaI and BclI.
There is also a cloning site). According to another aspect, Epstein-Barr virus (EBV)
A vector that directs episomal expression under the control of, for example, pREP4 (BamH1, SflI,
XhoI, NotI, NheI, HindIII, NheI, PvuII and KpnI cloning sites, constitutive
RSV-LTR promoter, hygromycin selectable marker; Invitrogen), pCEP4
(BamH1, SfiI, XhoI, NotI, NheI, HindIII, NheI, PvuII and KpnI cloning sites, constitutive bCMV immediate early gene, hygromycin selective marker; Invitrog
en), pMEP4 (KpnI, PvuI, NheI, HindIII, NotI, XhoI, SfiI, BamH1 cloning site, inducible metallothioinene Ila gene promoter, hygromycin selective marker; Invitrogen), pREP8 (BamH1, XhoI, NotI, HindIII, NheI And KpnI cloning site, RSV-LTR promoter, histidinol selectivity marker; Invitrogen), pREP9 (KpnI, NheI, HindIII, NotI, XhoI, SfiI and BamH1
Cloning site, RSV-LTR promoter, G418 selectable marker; Invitrogen) and pEBVHis (RSV-LTR promoter, hygromycin selectable marker, ProBon
d-resin-purified enterokinase cleaved N-terminal peptide; Invitrogen)
Can be used. As the selective mammalian expression vector used in the present invention,
There are pRc / CMV (HindIII, BstXI, NotI, SbaI and ApaI cloning sites, G418 selection; Invitrogen), pRc / RSV (HindIII, SpeI, BstXI, NotI, XbaI cloning sites, G418 selection; Invitrogen) and others. The vaccinia virus mammalian expression vector used according to the present invention (see Kaufman, 1991, above) is not limited to the following examples, but pSC11 (SmaI cloning site, TK- and b-
gal selection), pMJ601 (SalI, SmaI, AflI, NarI, BspMII, BamHI, ApaI, NheI, Sac
II, KpnI and HindIII cloning sites, TK- and b-gal selection) and pTKgptF1
S (EcoRI, PstI, SalI, AccI, HindII, SbaI, BamHI and Hpa cloning sites, TK or XPRT selection).

The yeast expression system can also be used in accordance with the present invention to express Smurf1. Only two are exemplified, for example, non-fused pYES2 vector (XbaI, SphI, ShoI, NotI, GstXI,
EcoRI, BstXI, BamH1, SacI, Kpn1 and HindIII cloning sites; Invitroge
n) or fusion pYESHisA, B, C (XbaI, SphI, ShoI, NotI, BstXI, EcoRI, BamH
1, SacI, Kpn1 and HindIII cloning sites, N-terminal peptides purified by ProBond resin and cleaved with enterokinase; Invitrogen) can be used according to the invention.

Once the individual recombinant Smurf DNA molecule has been identified and isolated, it can be propagated using several methods known in the art. Once a suitable host system and growth conditions are established, recombinant expression vectors can be propagated and produced in quantity. As described above, the vectors that can be used include, but are not limited to, the following vectors and derivatives thereof. That is, to cite a few examples, human or animal viruses such as vaccinia virus and adenovirus; insect viruses such as baculovirus; yeast vectors; bacteriophage vectors (eg, λ) and plasmids and cosmid DNA vectors. is there.

In addition, a host cell strain may be chosen which controls the expression of the inserted sequences, or modifies and processes the gene product in the specific fashion desired. Different host cells have characteristic and specific mechanisms for the translational and post-translational processing and modification of proteins. By choosing an appropriate cell line or host system, the desired modification and processing of the expressed foreign protein can be ensured. For example, expression in a bacterial system can be used to produce a non-glycosylated core protein product, while expression in yeast can produce a glycosylated product. Expression in eukaryotic cells can increase the likelihood of "native" folding of heterologous mammalian proteins. Furthermore, expression in mammalian cells can provide a reconstitution or constitutive tool for Smurf activity. Furthermore, it is known in the art that different vector / host expression systems affect processing reactions (eg, proteolytic cleavage) to different extents.

The vector may be a vector known in the art, for example, transfection, electroporation, microinjection, transduction, cell fusion,
Introduced into the desired host cell by DEAE dextran, calcium phosphate precipitation, lipofection (lysosomal fusion), using a gene gun, or a DNA vector transporter (e.g., Wu et al., 1992, J. Biol. Chem. 267: 963- 967; Wu and Wu, 1988,
J. Biol. Chem. 263: 14621-14624; Hartmut et al., Canadian Patent Application No. 2,012,311 filed March 15, 1990).

Analysis of Gene Expression Mediated by Smurf Functional Activity According to one embodiment, oligonucleotide array methods can be used to assess gene expression, for example, and after binding Smurf1 to Smads1 or 5, Or Smu
After binding rf2 to Smad7, in tissues with TGFβ and BMP-treated cells, cells with injured and / or healed tissue, tumor cells, and also mitosis and cell differentiation, embryo pattern formation, development and organ development. Gene expression can be identified that correlates with, or is different from, gene expression during cellular and developmental processes such as. For example, gene expression in cells treated with TGFβ or BMP can be compared in the presence or absence of Smurf expression. Gene Chip
p) Expression analysis (Affymetrix, Santa Clara, CA) produces data for conducting gene expression profiles and other biological assays. Oligonucleotide expression arrays simultaneously and quantitatively examine thousands of mRNA transcripts (genes or ESTs), simplifying large-scale genomic studies. Each transcript can be displayed on the probe array by multiple probe pairs, allowing discrimination between closely related gene family members. Each probe cell has several million copies of a specific oligonucleotide probe, and can detect a low intensity mRNA hybridization pattern accurately and with high sensitivity. The differential expression data allows a clear understanding of the cellular pathways.

After the hybridization, for example, Gene Array (Gene Gene) of Hewlett Packard
Intensity data is obtained using an Array ^™ scanner. The software can be used to automatically calculate the intensities for each probe cell. The intensity of probe cells can be used to calculate the average intensity of each gene that directly correlates with the expression level of mRNA. Expression data can be quickly sorted on all analytical parameters and displayed in various graphic formats for all selected gene subsets. Standard and custom GeneChip expression probe arrays are currently available for humans, mice, yeast and other organisms. The GeneChip product line will continue to expand into yet other organism analysis and expression arrays for applications such as toxicology and pharmaceutical genomes.

The transgenic vector Smurf can be introduced into cells to treat diseases associated with excessive BMP or TGFβ activation, such as cancer. Smurf activity can be assessed by introducing the Smurf vector into cells and monitoring the cells for Smurf expression. This can be done by in vitro or in vivo, or in vitro, also referred to as ex vivo, followed by transplantation. Alternatively, as described above, the Smurf or Smurf inhibitor (antisense, ribozyme, intracellular antibody) may be delivered by a vector of regulatory Smads, such as Smad's Smurf where BMP or TGFb activity is desirable, such as in bone regeneration. It is possible to interfere with regulation and to study the Smurf regulation process in vivo.

Smurf activity can be inhibited by various means. That is, delivery of a virus encoding a dominant negative Smurf derivative (e.g., Ala variant to Cys710) to cells, antisense nucleic acids (including ribozymes and triplex forming oligonucleotides; these are detailed above), And expression of anti-Smurf intrabodies, such as single chain Fv antibodies (generally Chen, Mol. Med. Today, 3: 160-167, 1997; Spitz et al.
al., Anticancer Res., 16: 3415-3422, 1996; Indolfi et al., Nat. Med., 2: 634-6.
35, 1996; Kikima et al., Pharmacol. Ther., 68: 247-267, 1995).

As mentioned above, a vector is any means for delivering a nucleic acid into a host cell according to the present invention. These include viral vectors such as lentivirus, retrovirus, herpes virus, adenovirus and adeno-associated virus. Therefore, a gene encoding a functional or mutant Smurf protein or a polypeptide domain fragment thereof can be isolated using a viral vector,
Alternatively, it can be introduced in vivo, ex vivo or in vitro by direct introduction of DNA. Expression in targeted tissues can be achieved by targeting the transgenic vector to specific cells that carry viral vectors or receptor ligands, or by using tissue-specific promoters, or both. You can Targeted gene delivery is described in International Patent Publication WO 95/28494 published in October 1995.

Viral vectors commonly used for in vivo or ex vivo targeting and therapeutics are DNA-based viral and retroviral vectors. Viral Methods for constructing and using viruses are known in the art [see, eg, Miller and Rosman, BioTechniques 7: 980-990 (1992)]. Viral vectors are preferably replication defective. That is, it is desirable that the target cell cannot autonomously replicate. Generally, the replication defective viral vectors used within the scope of the present invention lack at least one region required for viral replication in infected cells. These areas can be defined by any method known to those skilled in the art.
It can be removed (all or part) or disabled. These methods include total removal, substitution (due to other sequences, especially due to the inserted nucleic acid), or partial deletion or addition of one or more bases in the (replication) essential region. Such methods can be performed in vitro (on isolated DNA) or in situ using genetic engineering techniques or by treatment with mutagens. The replication defective virus preferably retains its genomic sequence necessary to wrap the viral particle.

The DNA virus vector includes a DNA virus having a transcriptional attenuation or deletion, and is not limited to the following examples. For example, herpes simplex virus (HSV)
), Papillomavirus Epstein-Barr virus (EBV), adenovirus, adeno-associated virus (AAV) and the like. Defective viruses with no or few viral genes are preferred. Defective viruses are not infectious after introduction into cells. By using a defective viral vector, cells can be administered to a specific local area without fear that the vector may infect other cells. Therefore, it is possible to specifically target a specific tissue. Specific vectors include, but are not limited to the following examples, defective herpesvirus 1 (HSV1) vector [Kaplitt et al., Molec. Cell. Neurosci., 2: 320-330 (
1991)], a defective herpesvirus vector lacking the glycoprotein L gene [Patent
Publication RD 371005 A] or other defective herpesvirus vector [1994
International Publication WO 94/21807, published September 29, 1996; International Publication WO 92/05263, published April 2, 1994; Stratford-Perricaudet et al., [J. Clin. Invest., 90: 626]. -630,
1992; La Salle et al., Science, 259: 988-990, 1993], and transcription-attenuated adenovirus vectors; and defective adeno-associated virus vectors [Samulski et al., J. Virol., 61: 3096-3101, 1987; Samulski et al.,
J. Virol., 63: 3822-3828, 1989; Lebkowski et al., Mol. Cell. Biol., 8: 398.
8-3996, 1988)] can be exemplified.

For in vivo administration, it is desirable to use a suitable immunosuppressive therapy in combination with, for example, a viral vector such as an adenovirus vector to avoid immune inactivation of the viral vector and transfected cells. For example, administration of immunosuppressive cytokines such as interleukin 12 (IL-12) and interferon-g (IFN-g), or anti-CD4 antibody to block humoral or cellular immune response against the viral vector Can be [eg Wilson, Nature Me
See dicine (1995)]. Furthermore, it is advantageous to use viral vectors that have been genetically engineered to express a minimal number of antigens.

Adenovirus Vector Adenovirus is a eukaryotic DNA and can be modified and modified so that the nucleic acid of the present invention can be efficiently delivered to various cell types. There are various serotypes of adenovirus. Among these serotypes, within the scope of the present invention, human adenovirus type 2 or 5 (Ad2 or Ad5), animal-derived adenovirus (WO 94/2
6914) is preferably used. These animal-derived adenoviruses that can be used within the scope of the present invention include dogs, cows, mice (eg Mav1, Beard et al., V
irology 75 (1990) 81), adenoviruses from sheep, pigs, birds and monkeys (eg SAV). The animal-derived adenovirus is preferably a canine adenovirus, and more preferably CAV2 adenovirus (for example, Manhattan or
A26 / 61 shares (ATCC VR-800) etc.).

The replication defective adenovirus vector of the present invention is preferably an ITR, ie,
It comprises a capsid forming sequence and a nucleic acid of interest. More preferably, at least the E1 region of the adenovirus vector is non-functional. The deletion of the E1 region is due to nucleotides 455-3329 (PvuII-BglII fragment) in the Ad5 adenovirus sequence, or 382
Up to 3446 (HinfII-Sau3A fragment). Other regions, in particular, E3 region (WO 95/02697), E2 region (WO 94/28938), E4 region (WO 94/28152, WO 94/12649 and WO 95/02697), or late gene L1-L5 Either may be modified and modified.

According to a specific embodiment, the adenovirus vector has a deletion in its E1 region (Ad 1.0). An example of an E1-deleted adenovirus is described in EP 185,573, the content of which is incorporated herein by reference. According to another embodiment, the adenovirus vector has a deletion in its E1 and E4 regions (Ad 3.0). Examples of E1 / E4 deleted adenoviruses are disclosed in WO 95/02697 and WO 96/22378, the contents of which are hereby incorporated by reference. According to yet another embodiment, the adenovirus vector has a deletion in the E1 region with the E4 region and the nucleic acid sequence inserted (FR9
See 4 13355. The contents of which are hereby incorporated by reference).

Replication-defective recombinant adenoviruses according to the invention can be made by methods known to those skilled in the art (Levrero et al., Gene 101: 195 1991; EP 185 573; Graham, EMBO J.
. 3: 2817, 1984). In particular, it can be produced by homologous recombination between an adenovirus and a plasmid having the DNA sequence of interest. Homologous recombination can be performed by co-transfecting the adenovirus and the plasmid into an appropriate cell line. The cell line used preferably comprises (i) a transformable by the above element and (ii) a sequence capable of complementing a part of the genome of the replication defective adenovirus, preferably in integrated form. It is necessary to be out. This is to avoid the risk of recombination. An example of a cell line that can be used is the human embryonic kidney cell line 293 (Graham et al., J. Gen. Virol. 36:59 1977, which contains the left-hand side of the Ad5 adenovirus (12%) genome integrated into its genome. ), And cell lines capable of complementing E1 and E4 functions, as described in WO 94/26914 and WO 95/02697. Recombinant adenovirus is recovered and purified using standard molecular biology techniques well known to those of skill in the art.

Adeno-Associated Viruses Adeno-associated virus (AAV) is a relatively small size DNA virus that can be integrated into the genome of cells infected by these viruses in a stable and site-specific manner. Adeno-associated virus is capable of infecting a wide range without affecting cell proliferation, morphology or differentiation and is not involved in human disease. The AAV genome has been cloned, sequenced and characterized. The AAV genome has about 4700 bases and contains an inverted terminal repeat (ITR) region of about 145 bases at each end, which serves as a viral origin of replication. The latter genome is divided into two essential regions and is responsible for encapsidation functions. The left-hand side of the genome contains the rep genes involved in viral replication and expression of viral genes, and the right-hand side of the genome contains the cap gene, which encodes the viral capsid protein.

The use of AAV-derived vectors to carry genes in vitro and in vivo has already been reported (WO 91/18088; WO 93/09239; US Pat. No. 4,797,368).
No. 5,139,941 and European Patent No. 488,528). These publications include r
Various AAV-derived constructs in which the ep and / or cap genes have been deleted and replaced by the gene of interest, and the above-mentioned genes of interest, are
It has been described for use in vitro (to cells in culture) or in vivo (directly to organisms). The replication defective recombinant AAV according to the present invention comprises two AAV inverted terminal repeats (IT
(R) co-transfecting a plasmid containing the nucleic acid sequence of interest flanked by a region and a plasmid carrying the AAV encapsidation genes (rep and cap genes) into a cell line infected with a human herpesvirus (e.g., adenovirus). You can make it by doing. The AAV recombinants produced are then purified by standard techniques.

Retroviral Vectors In other embodiments, retroviral vectors (eg Anderson et al., US Pat. No. 5,399,346; Temin et al., US Pat. No. 4,650,764; Temin et al., US Pat. No. 4,980,
289; Markawitz et al., 1988, J. Virol. 62: 1120; Temin et al., US Pat. No. 5,124,2.
63; EP453242, EP178220; Bernstein et al., Genet. Eng. 7 (1985) 235; McCor.
mik, BioTechnology 3 (1985) 689; Dougherty et al., Patent International Publication No. WO95 / 07358 (1
(Published March 16, 995); and Kuo et al., 1993, Blood 82: 845). Retroviruses infect dividing cells and the virus becomes integrated. The retroviral genome contains two LTRs, an encapsidation sequence and three coding regions (gag, pol and env). In recombinant retroviral vectors, the gag, pol and env genes are generally wholly or partially deleted to replace the heterologous nucleic acid sequence of interest. These vectors are HIV, MoMuL
V (Mouse Moloney leukemia virus "MSV (Mouse Moloney sarcoma virus)"), HaSV
("Harvey sarcoma virus"); SNV ("spleen necrosis virus"); RSV ("Rous sarcoma virus"); and various species of retroviruses such as the Friend virus. Protective retroviral vectors are disclosed in WO95 / 02697.

Generally, to construct a recombinant retrovirus containing a nucleic acid sequence, a plasmid containing the LTR, encapsidation sequence and coding sequence is constructed. This construct is used to transfect a packaging cell line, which is capable of providing retroviral function in situ that the plasmid lacks. In general, packaging cell lines can express the gag, pol and env genes in this way. Such packaging cell lines are conventional, particularly the cell line PA317 (US Pat. No. 4,861,
719); PsiCRIP cell line (WO90 / 02806) and GP + envAm-12 cell line (WO89 / 07150). In addition, recombinant retroviral vectors can contain modifications within the LTR to repress transcriptional activity and can contain a wide range of encapsidation sequences that can include a portion of the gag gene (Bender et al., J. Virol. 61. (1987) 1639). Recombinant retroviral vectors are purified by standard techniques known to those of ordinary skill in the art.

Retroviral vectors can be constructed to function as infectious particles or to undergo transfection manipulation. In the former case, the virus is modified to carry all its genes except those responsible for oncogenic transformation properties and to express heterologous structural genes. Non-infectious viral vectors disrupt the viral packaging signal but retain the heterologous structural gene and the structural gene required for packaging of the cotransfected virus genetically engineered to contain the packaging signal. It is made to do. Therefore, the virus particles produced cannot produce additional virus.

Lentiviral Vectors In another embodiment, lentiviral vectors are used as agents for direct release and sustained expression of transgenes in several tissues including brain, retina, muscle, liver and blood. Can be done. These vectors are able to efficiently transduce dividing and non-dividing cells into these tissues and maintain long-term expression of the gene of interest.

Lentiviruses contain at least two regulatory genes tat and rev that are essential for replication, and four accessory genes that encode critically important virulence factors [for review, Naldini, L., Curr. Opin. Biotechnol., 9: 457-63, 1998]. Viral sequences that are not essential for transduction are removed, thereby improving the biological safety of this particular vector. HIV-1 self-inactivating
A vector is known, which contains a 3'long terminal repeat containing the TATA box (LT
R) has a deletion and reduces the likelihood that replication-competent retroviruses will be generated in vector producers and target cells, significantly improving the biological safety of HIV-derived vectors. (Zufferey et al., J. Virol., 72: 9873-80,
1998). In addition, the deletion caused an LTR that was previously associated with transcriptional inhibition and in vivo repression.
By removing sequences and allowing the construction of more stringent tissue-specific or regulatable vectors, possible vector performance is improved.

Lentivirus packaging cell lines are available and known in the art. These facilitate the production of high titer lentiviral vectors for gene therapy. One example is the tetracycline inducible VSV-G pseudotyped lentivirus packaging cell line, which has a titer of> 10 (6) IU / ml of at least 3 to 4
Viral particles can be produced daily (Kafri et al., J. Virol., 73: 576-584, 1999). Vectors produced by inducible cell lines are capable of inducing non-dividing cells in vitro and in v
It can be concentrated as needed for efficient transduction in ivo.

Non-viral vectors or, alternatively, the vector can be introduced in vivo by lipofection. Over the past decade, the use of liposomes for nucleic acid encapsulation and in vivo transfection has been increasing. Synthetic cationic lipids designed to limit the difficulties and risks encountered in liposome-mediated transfection can be used to make liposomes for in vivo transfection of a gene encoding a marker [Felgner et al., Proc. Natl. Acad. Sci. USA, 84:74
13-7417, 1987; Mackey et al., Proc. Natl. Acad. Sci. USA, 85: 8027-8031, 1
988; Ulmer et al., Science, 259: 1745-1748, 1993]. The use of cationic lipids can facilitate encapsulation of negatively charged nucleic acids and also fusion with negatively charged cell membranes [Felgner and Ringold, Science, 337: 387-388, 198.
9]. Particularly useful lipid compounds and compositions for transfer of nucleic acids are described in International Patent Publication WO 95/18863.
And WO 96/17823, and US Pat. No. 5,459,127. The use of lipofection to introduce foreign genes into specific organs in vivo has certain practical advantages. The aim of the molecular targeting of liposomes to specific cells is one area of advantage. Transfection directed against specific cell types may be particularly advantageous in tissues with cell heterogeneity such as pancreas, liver, kidney and brain. Lipids can be chemically linked to other molecules for the purpose of targeting [Mackey et al., Supra]. Targeted peptides, such as hormones or neurotransmitters, and proteins such as antibodies, or non-peptide molecules could be chemically attached to liposomes.

Other molecules are also useful for facilitating transfection of nucleic acids in vivo, eg cationic oligopeptides (eg International Patent Publication WO95 / 21931), DN
A peptide derived from an A-binding protein (for example, International Patent Publication WO96 / 25508), or a cationic polymer (for example, International Patent Publication WO95 / 21931).

It is also possible to introduce the vector in vivo as a naked DNA plasmid. Raw DNA vectors for gene therapy are known in the art, such as transfection, electroporation, microinjection, transduction, cell fusion, DEAE dextran, calcium phosphate precipitation, use of gene gun, or DNA vectors. Use of Transporters [eg Wu et al., J. Biol. Chem., 267:
963-967, 1992; Wu and Wu, J. Biol. Chem., 263: 14621-14624, 1988; Hartm.
ut et al., Canadian Patent Application No. 2,012,311, filed March 15, 1990; Williams et al., Proc.
Natl. Acad. Sci. USA, 88: 2726-2730, 1991]] into a desired host cell. Receptor-mediated DNA delivery means can also be used [Curi
el et al., Hum, Gene Ther., 3: 147-154, 1992; Wu and Wu, J. Biol. Chem., 262.
: 4429-4432, 1987]. US Pat. No. 5,580,859 discloses delivery of exogenous DNA sequences free of transfection facilitating agents to mammals.

Antibodies Against Smurf Proteins According to the present invention, Smurf polypeptides comprising fusion proteins and fragments, variants, or other derivatives or analogs thereof produced recombinantly or by chemical synthesis are provided with Smurf polypeptides. It can be used as an immunogen to generate recognizing antibodies. Such antibodies include, but are not limited to, polyclonal, monoclonal, chimeric, single chain, Fab fragments, and Fab expression libraries. Such antibodies are specific for the Smurf protein, which may recognize mutants of Smurf, or wild type Smurf. These antibodies (eg anti-Smurf intracellular domain antibodies) can be used to alter the BMP pathway by inhibiting the Smurf protein or can be used for diagnostic purposes.

Various procedures known in the art can be used to produce polyclonal antibodies to Smurf polypeptides or derivatives or analogs thereof. To produce antibodies, various host animals including, but not limited to, rabbits, mice, rats, sheep, goats, etc. are immunized by injecting Smurf polypeptide or a derivative thereof (eg, fragment or fusion protein). can do. In certain embodiments, the Smurf polypeptide or fragment thereof can be conjugated to an immunogenic carrier such as bovine serum albumin (BSA) or keyhole limpet hemocyanin (KLH). Freund's adjuvant (complete and incomplete) depending on the host species, mineral gels such as aluminum hydroxide, surface-active substances such as lysolecithin, pluronic polyols, polyanions, peptides, oil emulsions, keyhole limpet hemocyanin, dinitrophenol, and Possible useful human adjuvants such as BCG (Bacille Calmette-Gu
erin) and various adjuvants, including but not limited to Corynebacterium parvum, can be used to enhance the immunological response.

To produce monoclonal antibodies to Smurf polypeptides or fragments, analogs, or derivatives thereof, any technique that results in the production of antibody molecules by continuous cell lines in culture can be used. These include Kohler and Milstein (Nature
256: 495-497, 1975), as well as the hybridoma technology first developed by Trioma, the human B-cell hybridoma technology (Kozbor et al., Immunology Today 4: 7).
2, 1983; Cote et al., Proc. Natl. Acad. Sci. USA 80: 2026-2030, 1983) and EBV hybridoma technology for producing human monoclonal antibodies (Cole et al.,
in Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc., pp. 77-
96, 1985), but is not limited thereto. In another embodiment of the invention, the monoclonal antibody can be produced in an embryo-free animal (International Patent Publication WO 89/1
2690, published December 28, 1989). In fact, according to the invention, production of a "chimeric antibody" by splicing a gene from a mouse antibody molecule specific for a Smurf polypeptide with a gene from a human antibody molecule of suitable biological activity. Technology developed for (Morrison et al., J. Bacteriol. 159: 870, 1984; Neuberg.
er et al., Nature 312: 604-608, 1984; Takeda et al., Nature 314: 452-454, 1985). Such antibodies are within the scope of the invention. Such human or humanized chimeric antibodies are preferred for use in the treatment of human diseases or disorders (below) because human or humanized antibodies are themselves heterogeneous in that they may elicit an immune response, especially an allergic response. It is much less than antibodies.

According to the invention, the techniques described for the production of single chain antibodies (US Pat. Nos. 5,476,786 and 5,132,405 to Huston; US Pat. No. 4,946,778) describe Smurf polypeptide-specific single chain antibodies. Can be adapted to produce. Another embodiment of the invention relates to the construction of Fab expression libraries to allow rapid and easy identification of monoclonal Fab fragments with the desired specificity for Smurf polypeptides, or derivatives or analogs thereof. Described technology (Huse et al., Sc
ience 246: 1275-1281, 1989).

Antibody fragments containing the idiotype of the antibody molecule can be generated by known techniques. For example, such fragments include, but are not limited to, F (ab ') 2 fragments that can be produced by pepsin digestion of antibody molecules; generated by reducing the disulfid bridge of F (ab') 2 fragments. Fab 'fragments that can be generated, and Fab fragments that can be generated by treating antibody molecules with papain and a reducing agent are included. Smurf activity is also a technique known in the art (generally Chen, Mol. Med. Today 3: 160-167, 1997; Spi
tz et al., Anticancer Res. 16: 3415-3422, 1996; Indolfi et al., Nat. Med. 2: 634-63.
5, 1996; Kijima et al., Pharmacol. Ther. 68: 247-267, 1995), and can be inhibited by the expression of anti-Smurf intracellular antibodies, such as single chain Fv antibodies.

In the production of antibodies, screening for the desired antibody can be accomplished by techniques known in the art, eg radioimmunoassay, ELISA (enzyme linked immunosorbent assay), “sandwich” immunoassay, immunoradiometric assay, Gel diffusion precipitation reactions, immunodiffusion assays, in situ immunoassays (eg using colloidal gold, enzymes or radioisotope labels),
Western blots, precipitation reactions, agglutination assays (eg gel agglutination assays, hemagglutination assays), complement fixation assays, immunofluorescence assays, protein A assays, and immunoelectrophoresis assays, etc. In one embodiment, antibody binding is detected by detecting a label on the primary antibody. In another embodiment, the primary antibody is detected by detecting binding of a secondary antibody or reagent to the primary antibody. In yet another embodiment, the secondary antibody is labeled. Many means for detection of binding in immunoassays are known in the art and are within the scope of the present invention. For example, to select antibodies that recognize specific epitopes of Smurf polypeptides, one can assay the hybridomas generated for products that bind to Smurf polypeptide fragments containing such epitopes. Antibodies specific for Smurf polypeptides from animals of a particular species can be selected based on positive binding to Smurf polypeptides expressed by or isolated from cells of that animal species.

The antibody may be Smur using any of the detection techniques described above or known in the art.
It can be used in any method known in the art for the localization and activity of f polypeptides, such as Western blotting, in situ Smurf polypeptide imaging, measuring its level in a suitable physiological sample, and the like. Such antibodies can also be used in the assays for ligand binding described, for example, in US Pat. No. 5,679,582.

In a detailed embodiment, Smurf polypeptide activator or antagonistic antibodies can be generated. Such antibodies can be tested using the assays described below for ligand identification.

Screening In accordance with the present invention, nucleotide sequences derived from the gene encoding Smurf and peptide sequences derived therefrom are useful for identifying agents effective in the treatment of disorders mediated by the BMP signaling pathway. It is a target because it can be used for Smurf protein agonism or antagonism as discussed above (Schmitt et al., J. Orthopedic Res., 17: 269, 1999). Drug targets include, but are not limited to, (i) nucleic acid isolates derived from the gene encoding Smurf, and (ii) peptides and polypeptide isolates derived from the Smurf polypeptide.

In particular, the identification and isolation of the Smurf protein is the development of screening assays,
For example, the development of high throughput screens for molecules that result in up-regulation or down-regulation of Smurf activity, especially by allowing expression of Smurf protein in an amount greater than that which can be isolated from natural sources,
Or expressed from a gene after transfection or transformation of cells
It provides for development in indicator cells specifically engineered to show the activity of the Smurf protein. Accordingly, the present invention contemplates methods for identifying specific ligands for the Smurf protein using various screening assays known in the art.

Any screening technique known in the art can be used to screen for agonists or antagonists of Smurf protein. The present invention provides screening of small molecule ligands or ligand analogs and mimetics, as well as Sm in vivo.
It is intended to screen for natural ligands that bind to, antagonize or act on the urf protein. For example, use the natural product library with Smurf
It can be screened using the assays of the invention for active agonist or antagonist molecules.

Information on the primary sequences of Smurf 1 and Smurf 2 and the similarity of those sequences to proteins of known function allow one of ordinary skill in the art to determine inhibitors or antagonists of Smurf protein. Useful information is provided. The identification and screening of antagonists is further facilitated by measuring the structural properties of the protein using, for example, X-ray crystallography, neutron diffraction, nuclear magnetic resonance spectroscopy, and other techniques for structural determination. . These techniques provide rational design or identification of agonists and antagonists.

Another approach is to use recombinant bacteriophage to generate large libraries. "Phage method" (Scott and Smith, Science 249, 386-
390, 1990; Cwirla et al., Proc. Natl. Acad. Sci., 87: 6378-6382, 1990; Devlin.
Et al., Science 49: 404-406, 1990), a very large (chemical presence 10 ⁶ -10 ⁸ ) library can be constructed. The second means mainly uses chemical methods, examples of which are Geysen method (Geysen et al., Molecular Immunology 23: 709-71).
5, 1986; Geysen et al., J. Immunologic Method 102: 259-274, 1987) and Fodor.
Et al. (Science 251: 767-773, 1991). Furka et al. (14th International
Congress of Biochemistry, Volume # 5, Abstract FR: 013, 1988; Furka, Int.
J. Peptide Protein Res. 37: 487-493, 1991), Houghton (U.S. Pat.No. 4,631,21).
No. 1, licensed December 1986) and Rutter et al. (US Pat. No. 5,010,175, April 1991 2
3 day permit) describes a method for producing a peptide mixture which can be tested as an agonist or antagonist.

In another embodiment, a synthetic library (Needels et al., Proc. Natl. Acad. Sci.
USA 90: 10700-4, 1993; Ohlmeyer et al., Proc. Natl. Acad. Sci. USA 90: 10922-109.
26, 1993; Lam et al., International Patent Publication WO92 / 00252; Kocis et al., International Patent Publication WO94 / 280.
28) etc. can be used for the screening of Smurf ligands according to the present invention.

Test compounds are screened from large libraries of synthetic or natural compounds. A large number of tools are currently used for random and directed synthesis of saccharide, peptide, and nucleic acid based compounds. Synthetic Compound Library Mayb
ridge Chemical Co. (Trevillet, Cornwall, UK), Comgenex (Princeton,
NJ), Brandon Associates (Merrimack, NH), and Microsource (New Milford,
It can be purchased from CT). Rare chemical library
Is available from Aldrich (Milwaukee, WI). Or bacteria, fungi,
Libraries of natural compounds in the form of plant and animal extracts are available, for example, from Pan Labora
Available from tories (Bothell, WA) or MycoSearch (NC) or can be easily made. Moreover, natural and synthetically produced libraries and compounds are readily modified by conventional chemical, physical, and biochemical means (Blondelle et al., Tib Tech, 14:60, 1996). .

In Vitro Screening Methods In one set of embodiments, the isolated nucleic acid containing the Smurf gene is labeled in vitro.
In vitro, its ability to bind a test compound in a sequence specific manner is tested. This method provides (i) a nucleic acid containing a specific sequence corresponding to all or part of the Smurf protein; (ii) contacting said nucleic acid with a number of test compounds under conditions suitable for binding; and (iii) identifying a compound that selectively binds to the nucleic acid sequence.

Selective binding, as used herein, refers to any measurable difference in any parameter of binding, such as binding affinity, binding capacity, and the like.

In one set of embodiments, an isolated peptide or polypeptide comprising a Smurf protein, or fragment thereof, is tested in vitro for its ability to sequence specifically bind a test compound. . This screening method provides (i) a peptide or polypeptide corresponding to the Smurf protein or fragment thereof, or a fragment thereof; (ii) contacting the peptides with a number of test compounds under conditions suitable for binding. And (iii) identifying a compound that selectively binds to the peptides.

In a preferred embodiment, high throughput screening protocols are used to investigate the ability of a large number of test compounds to bind in a sequence-specific manner to the above-mentioned genes or peptides.

In Vivo Screening Methods Intact cells or whole animals expressing variants of the gene encoding the Smurf protein can be used in screening methods to identify candidate drugs. The following method is applicable to normal or wild type Smurf.

[0129] In one set of embodiments, from an individual exhibiting expression of the mutant Smurf gene,
A permanent cell line is established. Alternatively, by introducing an appropriate vector into cells (including not only mammals, but also mammalian cells, insect cells, yeast cells, and bacterial cells), a gene containing one or more mutant Smurf sequences is expressed. , Program the cells. Identification of candidate compounds includes, but is not limited to (i) test compound Smur
an assay that measures selective binding of f to a particular variant; (ii) an assay that measures the ability of a test compound to alter (ie, inhibit or increase) a measurable activity or function of Smurf; and (iii ) Using any suitable assay, including assays that measure the ability of a compound to modify (ie, inhibit or increase) the transcriptional activity of a sequence derived from the promoter (ie, regulatory) region of the Smurf gene. Can be achieved.

In another set of embodiments, (i) human Smurf having one or more mutations
Is stably inserted into the genome of the transgenic animal; and / or (ii) the endogenous Smurf gene is inactivated to produce a transgenic animal in which the mutant human Smurf gene is replaced. For example, Coffman, Semin. Nephrol. 17: 404, 1
997; Esther et al., Lab. Invest. 74: 953, 1996; Murakami et al., Blood Press. Suppl.
. 2:36, 1996. A preferred method of producing said transgenic animal is to insert a human gene to replace the endogenous gene under the expression control of the regulatory sequences of the endogenous gene, in what is termed the "knock-in" method. (See Rodrigues et al., Cell, 97: 199, 1999). The animal can be treated with the candidate compound and monitored for any changes in the BMP and TGFβ / activin pathways.

In addition, populations that are not sensitive to established treatments or accelerated regeneration for tissue and bone degeneration (eg, osteoporosis) can be selected for testing alternative treatments. Moreover, treatments that are not effective in the general population, but are highly effective in selected populations (otherwise overlooked) can be identified. This is a particularly powerful advantage of the present invention as it eliminates some of the chaos associated with clinical trials.

High Throughput Screening Agents of the invention can be identified by high throughput assays, including, but not limited to, cell-based assays or cell-free assays. One of ordinary skill in the art will appreciate that different types of assays can be used to detect different types of agents. Several automated assay methods have been developed in recent years that have made it possible to screen tens of thousands of compounds in a short time. Such a high throughput screening method is particularly preferable. The use of high throughput screening assays to test drugs is greatly facilitated by the availability of large amounts of purified polypeptide as provided by the present invention.

The reporter gene Smad protein is a mediator of intracellular signaling by members of the transforming growth factor-β (TGF-β) superfamily that affects cell proliferation, differentiation, and pattern formation during early spine development. Was identified as. A heteromeric complex consisting of a pathway-restricted Smad and a common Smad4 that is thought to play an important role in gene transcription following receptor activation and is subsequently translocated into the nucleus. Smad is incorporated into the body (heteromeric complexes). A Smad binding element (Smad B containing a CAGACA sequence in the JunB gene promoter)
inding Elements; SBE) are TGF-β, activin, and bone morphogenetic protein (BMP).
) 2 is a strongly induced immediate early gene (Jonk et al., J. Biol. Chem.,
273: 21145, 1998). The two JunB SBEs are arranged as inverted repeats,
It is transactivated in response to co-overexpression of Smad3 and Smad4, and TGF-β
7 shows the inducible binding of Smad3 and Smad4 containing complexes in nuclear extracts from cells treated with. The Smad protein binds directly to SBE. SBE multiplexing results in a TGF-β inducible enhancer that also has strong reactivity with activin and BMP (
See, eg, Kim et al., Nature, 388: 304, 1997).

Expression of the reporter gene can be linked to expression or activation of any component of the BMP signaling pathway. That is, a nucleotide sequence having an SBE sequence can be isolated using methods known in the art and operably linked to a reporter gene, Smurf, its derivatives and its mutants exhibiting activity of the promoter. It can be decided whether or not to modify. In addition, other known genes in the BMP signaling pathway, such as Tlx2 or Smad7, which have promoters regulated by Smad1 and 5, can be operably linked to a reporter gene. In these assays, candidate compounds were either Smad1, Smad5, Smad7, Smurf1 or Smuf1.
The ability to modify BMP in TGFβ signaling in cells expressing rf2, or derivatives or variants thereof, can be tested.

Various reporter gene assays are known and can be used to assess the effect of Smurf on the BMP or TGFβ / activin signaling pathways. As a reporter, luciferase, β-galactosidase (β-gal or lac-Z)
, Chloramphenicol acetyl transferase (CAT), horseradish peroxidase, and alkaline phosphatase. further,
Expression of most any protein can also be detected using specific antibodies. For example, the green fluorescent protein (GFP) expression assay allows evaluation of Smad-induced SBE activity. GFP retains its function but has been modified to produce proteins with different fluorescent properties. Several US patents teach the use of GFP to visualize signaling pathways linked to reporter genes by inducible promoters. For example, US Pat. No. 5,625,048 (modified GFP that produces amino acid changes provides a visually distinct color and increased emission intensity), WO 96/23898 (enzyme recognition). A modified GFP-encoding construct that also contains a site), WO 97/11094 (fluorescent protein with increased intensity), WO 97/266333 (higher expression levels in mammals) Optimized humanized GFP protein to provide)
International Patent Publication No. WO97 / 42320 (modified GFP having fluorescence with increased intensity)
, International Patent Publication No. WO98 / 06737 (modified GFP that is easily distinguished from green and blue fluorescent proteins), International Patent Publication No. WO98 / 21355 (GFP mutant that can be excited using blue and white light) Please refer to.

Screening Kit The components necessary to carry out the screening method described above can be manufactured in the form of a kit for the convenience of the user. The kit is preferably adapted for use in an automated screening device.

Identification of Smurf1 Binding Partners In yet another embodiment, the invention can then be analyzed for mutations leading to diseases mediated by the BMP or TGFβ / activin pathways.
, Smad7 and Smad5, as well as the identification of Smurf binding partners. One method of identifying the binding partner is preferably recombination with a hematopoietic stem cell library.
A yeast two-hybrid assay system using a yeast transformed with Smurf. Alternatively, the Smurf protein can be used to affinity purify the protein from cell preparations, eg, using cells that endogenously produce Smurf. Partially purified preparations can be probed with labeled Smurf protein to identify specific binding partners, eg, in Western systems or other antibody assay systems (antibody for example assay). See, above ,; generally, any protein can be labeled as an antibody and its binding to a binding partner assessed.

[0138] EXAMPLES The present invention may be better understood by reference to the following examples. These examples are not intended to limit the invention.

Example 1 Smurf1 Targets the BMP Pathway and Affects Embryonic Patterning The TGFβ superfamily regulates various biological processes including cell proliferation, differentiation and patterning. Any dysregulation of TGFβ signals (misregula
tion) can cause illness. The signal from the activated TGFβ receptor is transferred directly from the receptor to the nucleus by the Smad protein. Currently, several relationships have been established between ubiquitin-mediated degradation and regulation of morphogenetic signaling during development. This example describes Smurf1, a novel E3 ubiquitin ligase that selectively interacts with BMP pathway-specific Smads to trigger their ubiquitination, degradation and loss of activity. In amphibian embryos, Smurf1 BMP
It specifically blocked the signal and affected pattern formation. Thus, targeted ubiquitination of Smads may function to control both embryonic development and various cellular responses to TGFβ signaling.

Method Yeast Two-Hybrid Screening Xenopus Smad1 cDNA (36) was cloned into pGBT9 vector and Xenopus oocyte cDNA live by yeast two-hybrid method (46) using Xenopus Smad1 as bait protein. Rally (Clonte
ch manufactured). Partial cDNA was isolated, and a cDNA library of Xenopus stage 9 (blastula) was used for screening.
A full-length Smurf1 cDNA [SEQ ID NO: 1] was obtained. A human Smurf1 cDNA encoding all amino acids other than the first 8 amino acids was identified in the EST database (AA292123),
It was used to construct human Smurf1. Corresponding Xenopus Smurf1 cDN
The A sequence was used to reconstruct the first 8 amino acids. FLA at the N-terminus of hSmurf cDNA
G tag is attached. A ubiquitin ligase mutant of hSmurf1 was generated by substituting cysteine at position 710 with alanine using a PCR-based approach and this mutation was confirmed by screening.

For co-immunoprecipitation immunoprecipitation assays, Xenopus Smad1 (36), mouse Smad4,
Also, a FLAG tag was added to the C-terminal of human Smad2 (47), and in vitro translation (rabbit reticulocyte extract; Promega) was performed in the presence of ³⁵ S-Met. The FLAG-tagged Smad was bound to anti-FLAG antibody-bound beads (Kodak) and co-IP buffer (10 mM Tris, pH 7.5, 90 mM NaCl, 1 mM EDTA, 1% Triton-X 100, 10 was added. % Glycerol, 1 mM phenylmethylsulfonylfluoride) and then incubated with ³⁵ S-Met labeled Smurf1 in the same buffer. After washing in co-IP buffer and elution in gel loading buffer, proteins were separated by SDS-PAGE and visualized by autoradiography.

Experimental methods for embryos and RT-PCR embryo production, injection of prepared and synthesized mRNA into embryos, animal pole cap and ventral marginal zone assay, embryo RNA isolation, embryo fixation and whole mount in situ hybridization,
And developmental RT-PCR was performed as described (36, 48). Smurf1 RT-P
The primers for CR were 5'-GTCCTGTGACTGGAACCC-3 '(sense) [SEQ ID NO: 5] and 5'-GAGGACTGCTAGACAAT-3' (antisense) [SEQ ID NO: 6], and these 5
The'ends are located at positions 482 and 726 of the Smurf1 cDNA, respectively.

Northern blots were performed for mSmurf1 expression in mSmurf1 embryonic and adult mouse tissues. Equal amount of poly A + m from post-coitum 7th, 11th and 15th day embryo tissue
RNA and tissue samples from testis, kidney, bone, muscle, embryo, spleen, brain and heart were analyzed. Cytoskeletal actin expression was assayed on the same blot to confirm that the amount of mRNA applied was equal in all lanes. Whole mount on mouse embryo
In situ hybridization was also performed.

Immunoprecipitation and immunoblotting Lipofectamine (manufactured by GibroBRL) and calcium phosphate precipitation method (respectively)
49) was used to transiently transfect COS-1 and 293T cells. Anti-F
lag M2 monoclonal antibody (Sigma), anti-Smad1 / 5 poloclonal antibody (50)
Alternatively, immunoprecipitation and immunoblotting were performed as described above (50) using anti-WW2 Nedd4 (51) poloclonal antibody. Detection was performed using an appropriate HRP (manufactured by Amersham) that binds to goat anti-mouse secondary antibody or goat anti-rabbit secondary antibody and has enhanced chemiluminescence.

Pulse Chase Analysis COS-1 cells were transfected as described above. Two days after transfection, 50 μCi [ ³⁵ S] -methionine / ml in EMEM without methionine was used to induce 37
The cells were labeled at 10 ° C. for 10 minutes (pulse labeling). Next, the cell layer was washed twice, and DMEM + 1
Incubated as instructed in 0% FCS (chase). At each time point in the chase, TNTE cell lysis buffer containing protease and phosphatase inhibitors (50 mM Tris / HCl, pH 7.4, 150 mM NaCl, 0.5% Triton X-100 and 1%).
Cell lysates prepared in mM EDTA) were immunoprecipitated with anti-Smad1 poroclonal antibody. Immune complexes were resolved by SDS-PAGE and visualized by autoradiography. Phosphorimager; Mole
The amount of metabolically labeled Smad1 present at each time point was measured using cular Dynamics).

Ubiquitination Assay As described above, HA-tagged ubiquitin, untagged Smad1, and Flag-hSmurf.
293T cells were transfected with either 1 or Flag-hSmurf1 (C710A). Two days after transfection, cells were lysed and anti-Smad1 immunoprecipitated. Then TNTE-0.1% Triton, SDS-RIPA (TNTE cell lysis buffer, 0.1% sodium dodecyl sulfate and 1% deoxycholate), and 5
The immunoprecipitates were washed twice consecutively in 00 mM LiCl, 50 mM Tris / HCl, pH 7.4 and 0.1% Triton. By SDS-PAGE, HA-ubiquitinated Sma in the immune complex was detected.
The presence of d1 was visualized and then immunoblotted with monoclonal anti-HA 12CA5. Protein concentrations of untagged Smad1, Flag-hSmurf1 and Flag-hSmurf1 (C710A) were analyzed by immunoblotting aliquots from whole cells with the appropriate antibody.

Results Isolation of an E3 Ubiquitin Ligase That Interacts with the BMP Signaling Agent Smad1 To isolate a factor that interacts with Smad1 and potentially regulates Smad1, we fed Xenopus Smad1. A yeast two-hybrid screen was used as the (bait) protein. There are several overlaps that encode proteins with significant homology to the Hect subclass of E3 ubiquitin ligase (22)
The cDNA was isolated. This novel gene and each (Smad ubiquitination regulator-1
The closely related human homologues, Smurf1 and hSmurf1, are proteins consisting of 731 amino acids in length with 91% sequence identity and H at the C-terminal part of the molecule.
ect Contains the ubiquitin ligase domain. This domain is unique to a particular class of E3 ubiquitin ligases (22), which includes E6-AP and Nedd4 in animals and RSP5p and Pub1 (23-26) in yeast.

Identification of the Human Smurf1 Genomic Gene Location The human Smurf1 genomic clone is located in PAC clone DJ0808A01, from chromosomal region 7q21.1-q31.1, located from nucleotide 2669 to nucleotide 53763.
Computational prediction of intron / exon splicing products created a single conceptual mRNA with an open reading frame that directly corresponds to the actually cloned human cDNA.

The E3 ubiquitin ligase, along with the E1 and E2 enzymes, functions to link ubiquitin to specific protein substrates whose proteins are the target of subsequent degradation by the proteasome (12). The E1 enzyme then assembles and activates the ubiquitin-conjugating enzyme, ubiquitin, which is transmitted to E2. E2 then transfers ubiquitin to the E3 ubiquitin ligase, which either adds ubiquitin directly to the protein to be modified or confers substrate selectivity on the ubiquitinated complex. By functioning to assemble specific target proteins into the conjugation machinery, E3 activity confers selectivity on the ubiquitination process. E3 activity can be provided structurally and functionally by a wide variety of proteins. For example, E3 exists as a multi-binding protein complex that facilitates substrate recognition, but may or may not have direct ligase activity. An example of this is the N-terminal rule E3 class (Ne
nd rule E3s), and Skp1 / Cullin / F-box (SCF) complexes (27-29)
. In the case of the Hect family of E3 ubiquitin ligases, of which Smurf1 is a member, a single protein provides both substrate specificity and catalytic activity (22,30).

Smurf1 also exhibits some other structural features unique to RSP5, Nedd4 and Pub1 ubiquitin ligases. It is an amino-terminal phospholipid and calcium-binding C2
Domains, as well as two WW domains (31-33) that facilitate protein-protein interactions by binding to PPXY motifs on partner proteins. Overall, Smurf1 shows that cdc25 is proteosomally degraded.
It is most closely related to Pub1, which is a ubiquitin ligase from Sacromyces pombe that regulates mitosis because it is directed to egradation (23).

Expression of mSmurf1 in Embryonic and Adult Tissues Northern blots of mSmurf1 showed that a transcript of approximately 6 kb was the predominant form of Smurf1 expressed in embryos (FIG. 2A). Major transcripts of 3.0 and 6.0 kb were found in tissues (Fig. 2B). Cytoskeletal actin expression was assayed on the same blot and confirmed equal mRNA abundance in all lanes. In mice, mSmurf1
Was present in very early development, at least 7 days before embryogenesis.
In adults, mSmurf1 was expressed in most tissues assayed, although there were differences in transcript abundance. We found no detectable mSmurf1 transcript in skeletal muscle, with low levels of large transcripts expressed in kidney and spleen. The abundance of small transcripts in these studies was extremely high, very weak, or expressed in kidney, lung, spleen, brain, and heart, if not all. Equal amounts of poly A + RNA were added to each lane, but both blots were prepared so that variations in RNA loading and transfer were not taken into consideration, and blots were blotted with tubulin probe, but I admitted that there was no significant difference. Thus, the variation seen in the blots reflected the relative expression of mSmurf1 in embryos and tissues.

For whole mount in situ hybridization of mouse embryos, 10 to 1 post-mating
From day 4, Smurf1 expression was revealed in the nervous system, eye, gill arch, axial mesoderm (chordata), ganglion and limb bud.

Localization of Xenopus Smurf1 to the Egg Animal Pole and Embryonic Ectodermal In embryogenesis of Xenopus, Smurf1 mRNA was found to be expressed in eggs throughout the swimming tadpole stage, eggs, blastula and gastrates. Maximum levels were observed during the early stages of development during the embryonic period. Smurf1 levels declined sharply near the end of gastrulation stage and zygote expression was still maintained at low levels until late swimming tadpole stage
(Figure 3A). In whole mount in situ hybridization (Figure 3B), substantial Smurf
1 Transcript is localized in the animal polar hemisphere of the egg and in dividing blastulas, as confirmed by Northern blot analysis on RNA isolated from animal and plant hemispheres of mid-blastula embryos (Not shown). In the gastrulation stage,
The expression of Smurf1 became more widespread in the embryo, with a concomitant decrease in its transcript observed by RT-PCR. However, upon neural plate occlusion, Smurf1 expression was localized in the developing nervous system and was well expressed in the central nervous system, eye, pharyngeal sac and progenitor by the tadpole stage. Embryonic expression patterns of Smurf1 are expressed in the blastoderm and gastrula ectoderm and in the tadpole nervous system, eye and pharyngeal sac
It partially overlaps with the expression of Smad1 and BMP-4 (34-36).

Expression of hSmurf1 was selectively low at steady-state levels of Smad1 and Smad5 in mammalian cells.
Smurf1 bring down was candidate proteins that interact with Smad1 because they contain a putative E3 ligase and Hect domain. To investigate whether Smurf1 regulates the steady-state level of Smad1 protein, two mammalian cell lines, 293T and COS-1, were transfected with Smad1 alone or with increasing amounts of Flag-hSmurf1. The steady level of Smad1 protein was then assessed by immunoblot of whole cell lysates. Snad1 was easily detected in the absence of hSmurf1 (Fig. 4A). However, hSmurf1 expression resulted in a significant dose-dependent decrease in Smad1 protein levels, even at low levels that were not readily detectable by Western blotting. No detectable Smad1 protein was found at the highest level of hSmurf1 expression. Furthermore, co-expression of a constitutively activated form of the BMP type I receptor ALK6 that phosphorylates Smad1 (37, 38) did not alter the hSmurf1-dependent decrease in Smad1 levels (FIG. 4B). Expression of hSmurf1 therefore results in a dose-dependent reduction in the steady-state level of Smad1. This type of action is due to the type I BMP receptor
It can occur independently of the action of Smad1.

To determine if Smurf1 activity is restricted to Smad1, we present TGFβ.
Sm regulated by a receptor that acts in and activates signal transduction pathways
Its effect on ad Smad2 was investigated. Was sensitive to the lowest dose of hSmerf1
Unlike Smad1, it acted only slightly on the steady-state levels of Smad2 protein found only at the highest levels of Flag-hSmurf1 expression (FIG. 4C). Other Smads were also tested: Flag-hSmurf1 had little or no effect on Smad3 or Smad4 protein levels, but induced a strong reduction in Smad5 protein, which is closely related to Smad1 (FIG. 4D). Together, these data demonstrate that hSmurf1 preferentially regulates constant levels of Smad1 and Smad5, Smads regulated by two receptors that act in BMP signaling.

HSmurf1 regulates degradation and ubiquitination of Smad1 Inclusion of a putative E3 ligase or Hect domain supported the hypothesis that Smurf1 functions as an E3 ubiquitin ligase. To investigate whether Smurf1 regulates Smad degradation, we focused on Smad1 (91% homology with Smad5). Smad1 turnover (turnover (turn turn) by pulse chase (chase) experiment
over)) analysis revealed that Smad1 had a half-life of approximately 6 hours in the absence of hSmurf1. However, in the presence of hSmurf1, Smad1 turnover was significantly enhanced (half-life less than 2 hours) (FIG. 5A). Thus hSmurf1 increases the turnover speed of Smad1.

To investigate the mechanism of hSmurf1-mediated Smad1 turnover,
The ubiquitination of Smad1 in whole cells was evaluated. To help detect ubiquitin, 293T cells were transfected with HA-tagged ubiquitin along with Smad1 in the presence or absence of Flag-hSmurf1. In the absence of hSmurf1, Smad1 showed little or no detectable ubiquitination. However, when co-transfected with hSmurf1, we observed the appearance of a ladder of Smad1 coupled to ubiquitin (FIG. 5B). To confirm that ubiquitination of Smad1 requires the catalytic activity of the HS domain of hSmurf1, a point mutation of hSmurf1 (hSmurf1 (C710A)
) Was built. This residue is important for the catalytic activity of the Hect domain and this mutation appears to target a cysteine residue that forms a thioester bond with ubiquitin (22). In contrast to wild type hSmurf1, expression of hSmurf1 (C710A) did not result in ubiquitinated Smad1. Furthermore, hSmurf1 (C710A) did not affect Smad1 constant protein levels compared to wild-type hSmurf1 despite effective expression of this mutant protein (FIG. 5C). Together, these data suggest that hSmurf1 alters steady-state levels of Smad1 by inducing ubiquitin-mediated degradation of Smad1 through its Hect domain.

Smurf1 Selectively Interacts with Smad1 and Smad5 In Vivo To assess the basis of the selective targeting of Smad1 and 5 for degradation by Smurf1, we investigated the interaction of Smurf1 with Smad proteins. It was First, the yeast two-hybrid assay was used to test the ability of Xenopus Smurf1 to interact with Smad1, Smad4 or a non-specific control nuclear lamin. Yeast co-transfected with Smurf1 and Smad1 showed significant β-galactosidase activity, and Smurf1
And yeast co-transfected with either Smad4 or lamin did not show β-galactosidase activity (Fig. 6A right). Smurf1, but not Smad2 or Smad4, selectively bound to Smad1 (FIG. 6A right), as shown by the ability of ³⁵ S-labeled Smurf1 to bind and immunoprecipitate Smad in vitro.

To test the ability of ³⁵ S-labeled Smurf1 to bind and immunoprecipitate Smad in vitro, the specificity of the Smurf1-Smad interaction in intact cells, hSmur in 293T cells.
The association of f1 with various Smads was investigated. No interaction with Smad1 was detected in wild-type hSmurf1. However, in whole cells, the association between ubiquitin ligases and their substrates was found to be difficult in other systems, suggesting that Smurf1-Smad
Interactions are transient in nature. Smad1 and ubiquitin ligase variant F
By examining the interaction with lag-hSmurf1 (C710A), protein association could be detected (Fig. 6B, C). Furthermore, this interaction is a constitutively active BMP type I receptor
Not affected by co-expression of ALK2 (data not shown), but hSmu
Consistent with rf1 regulating Smad1 turnover-dependent BMP signaling. hSmurf1 (C710A) also bound efficiently to Smad5, but analysis of its association with Smad2 revealed only minor interactions, if any (Fig. 6).
A and B).

The linker region of R-Smad contains a PPXY sequence, which is a conserved motif recognized by WW domains like those found in Smurf1. Unlike wild-type Smad1, mutants of Smad1 lacking the PY motif only weakly associate with Smurf1 and were resistant to Smurf1-mediated degradation (data not shown).
These results indicate that hSmurf1 specifically associates with Smad1 and Smad5 and their interaction is Smad1.
Shows that it is mediated by the binding between the PY motif of Smurf1 and the WW domain of Smurf1.

The specificity of hSmurf1 as a ubiquitin kinase in targeting Smads was further tested by comparison with the structurally related ubiquitin ligase Nedd4. Smad1
Effectively co-precipitated with hSmurf1 (C710A), but did not interact with the corresponding Nedda mutant. Consistent with this, Nedd4 overexpression did not affect the steady-state level of Smad1 protein (FIG. 6C lower panel). Together, these data indicate that both human and Xenopus Smurf1 proteins selectively associate with BMP-regulated Smad1 and Smad5, an interaction that is associated with Smurf1 rather than the general characteristics of Hect ubiquitin ligases. It is shown to be specific.

Smurf1 antagonizes endogenous BMP signaling in Xenopus embryos and ventral mesoderm
In Xenopus blastulas, which dorsalize the leaves and deactivate the ectoderm, Smurf1 expression is localized in the animal polar ectoderm and partially overlaps with the zone, i.e., in the equator of the blastula that forms the mesoderm. The belt is located. This pattern of expression interacts with ubiquitination of BMP pathway-specific Smads,
Combined with the ability of Smurf1 to further degrade, it is suggested that Smurf1 functions in ectodermal and mesoderm patterning by antagonizing BMP signaling via Smad1 or Smad5.

Currently, naturally occurring TGFβ inhibitors are described at their ligand level.
These factors include cordine, follistatin and noggin, which act outside the cell by binding to specific ligands including BMP and activin. BMP in the abdominal zone during induction and patterning of Xenopus mesoderm
Signals specify tissue development that is characteristic of tadpole zones such as blood and mesenchyme. However, if BMP signaling is inhibited by ligand antagonists such as cordin, follistatin and noggin (secreted by the dorsal Spemann organizer) or by artificial inhibitors such as dominant negative BMP ligands or receptors, then it would be expected. Abdominal mesoderm differentiates into dorsal tissues such as muscle and notochord (a process called "dorsification") (5,
39).

Since Smurf1 has the potential to regulate the BMP signal in the band, Smurf1
Ectopic expression of 1 was tested for interference with ventral tissue patterning. Smurf1 mRNA
Were microinjected into the ventral zone (VMZ) of the 4-cell blastocyst stage to induce the formation of ectopic secondary axial structures in the ventral zone of 52% (n = 25) of tadpoles. These secondary axial structures induced by Smurf1 are characteristic of dorsalization caused by BMP inhibition. Their formation is rescued by co-expression of Smad1 and Smurf1, demonstrating that the action of Smurf1 is limited to inhibition by the BMP / Smad1 pathway (FIG. 7A). In addition, the dorsal zone (D
There was no effect on Smurf1 overexpression in MZ) cells (data not shown). This observation was consistent with previous findings of the present invention that Smurf1 does not target Smad2 in cultured cells.

The dorsalizing effect of Smurf1 was further characterized using VMZ explants. In this case, Smu
Expression of rf1 causes a reduction in blood cell differentiation and concomitant muscle differentiation, and Smurf1
(Fig. 7A right panel). Smad1 for axis formation assay
Co-expression of Smurf1 and Smurf1 revealed these dorsalizing effects, demonstrating BMP pathway specificity.

In the ectoderm origin, endogenous BMP expression specifies the epidermis, but when the BMP signal is reduced or abolished, the ectoderm differentiates into cement glands or neural tissue, respectively (4
0, 41). Smurf1 mRN in the ectoderm of Xenopus blastula and early gastrula
Localization of A suggested that Smurf1 regulates ectodermal patterning by BMP signaling. Therefore, Smurf1 was tested for its presence in ectodermal tissue. Overexpression of Smurf1 in the animal pole cap induces neuronal and cement gland differentiation that is characteristic of reduced BMP signaling. These effects were reversed by co-expression of Smad1 in the animal pole cap (FIG. 7B). Together, these results demonstrate that Smurf1 can inhibit BMP signaling in ectoderm and mesoderm, which impairs BMP signaling in these tissues and affects the patterning of embryogenesis. Suggest to do.

Smurf1 Inhibits Smad1 Activity in Embryogenic Cells, but in Cultured Cells That Enhance Smad2 Activity , Smurf1 Induces BMP Pathway-Specific Smad Ubiquitination and Degradation, and Smurf1 Endogenous in Xenopus Embryos It antagonizes the BMP signal. Overexpression of Smad is specific for R-Sm in Xenopus animal pole caps
Mimics the action of TGFβ factor signaling through ad. Therefore, Smad1 is a BMP
Like ligand, it induces exclusively ventral / posterior mesoderm, but Smad2 is activin Vgl
Similarly, it induces the dorsal (Speman organizer) mesoderm and nodes. We therefore investigated whether Smurf1 could directly antagonize the mesoderm-inducing activity of Smad1 or Smad2 by overexpressing each Smad with varying amounts of Smurf1 in the animal pole cap. The inventors have found that expression of Smad1 alone (1 ng mRNA) induces ventral mesoderm, as evidenced by expression of ventral / posterior mesoderm markers Xhox3 and Xcad1. However, co-expression of Smurf1 and Smad1 was found in all Smu tested.
The amount of rf1 inhibited the induction of these markers (Fig. 8A), which indicates that Smurf1 caused Smad1.
This proves that it can antagonize the activity.

To test whether Smurf1 could be inhibited by Smad2, a fixed amount (50 pg mRNA) of Smad
2 was effective in inducing myoD (dorsal / lateral mesoderm vertebrate muscle marker), but was used in an insufficient amount for inducing goosecoid (Speman organizer dorsal mesoderm frog marker) (Fig. 7B, C). When Smurf1 co-expressed with Smad2,
No inhibition of myoD induction was seen with any of the Smurf1 doses tested (Fig. 8B). This was consistent with other findings that Smurf1 does not target Smad2. Interestingly, induction of goosecoid gene expression was observed with increasing Smurf1 levels in the presence of this limited amount (50 pg) of Smad2. This induction was dependent on Smad2 expression, as Smurf1 alone did not induce goosecoid. Limited Smad 2
The response of cap cells to increasing the amount of Smurf1 in the presence of A had similar effects to increasing the amount of Smad2 alone (FIG. 8C). in this way,
The combination of 50 pg Smad2 and 100 pg Smurf1 induced goosecoid expression to levels comparable to 5-fold higher amounts of Smad2 alone (250 pg) (FIG. 8C). Thus, rather than inhibiting Smad2 activity, Smurf1 appears to enhance the sensitivity of the animal pole cap to Smad2. These experiments inhibit the response of animal polar cells to the BMP pathway, as well as S
murf1 proves to enhance the response of these cells to the activin pathway. Although the present invention does not depend on a particular mechanism, we found that a Smurf1-mediated shift in animal polar cell responsiveness is caused by the endogenous ubiquitination of Smurf1 substrates, Smad1 and 5. We propose that it is due to a decrease in BMP signaling. Thus, Smurf1 functions during development to alter the responsiveness of cells to multiple TGFβ ligands by selective inactivation of specific Smad pathways.

Discussion Our findings provide a clear role, pathway and mechanism for selective ubiquitination in regulating developmental patterning. In the example, TGFβ signaling is Sma
It can be regulated by ubiquitination of d signal transducing molecules, providing evidence that this activity has developmental consequences. Specifically, it was shown that the Hect family E3 ubiquitin ligase, Smurf1, selectively and directly interacts with BMP pathway-specific Smads, Smad1 and Smad5, to induce their ubiquitination and degradation. However, Smurf1 does not interact with, or are affected by, the TGFβ / activin pathway-specific Smad2, or a common partner in the Smad signal transduction complex Smad4. Targeting Smads with Hect ubiquitin ligase is not a common feature of E3 ligases, as Nedd4 does not interact with Smads.
Interestingly, these results indicate that Smurf1-mediated Smad turnover is unaffected by BMP signaling, but this is not required for activated forms of Smad1 or Smad5 to function as Smurf1 substrates. Is meant. Thus, Smurf1 does not act downstream of activated Smads to produce BMP signals, but by regulating the steady-state level of Smad proteins in cells, BM
Controls the responsiveness of cells that respond to P.

Although the mSmurf1 gene is expressed in early developmental stages, the inventors speculate that it regulates the cellular response to BMP signaling. Smurf1 is present in adult tissues, implying that mSmurf1 regulates late BMP signaling. Perhaps these signals act on tissue quiescence or on ongoing differentiation associated with growth. The high degree of expression in developing mice after the neuroembryonic stage is in good agreement with the expression pattern seen in frog embryos at the same stage.

The phenotypic effect of Smurf1 in Xenopus embryos refers to selective ubiquitination as a key regulator of inducible signals during embryogenetic development. Smurf1 backs the mesoderm and abrogates the ectoderm by inhibiting the BMP signals that control the patterning of these progenitor layers during normal development. Currently, BMP
Dependent patterning regulation of BMP extracellularly by direct binding to BMP ligands
It is believed to be achieved by secretory BMP binding proteins such as cordin, noggin and follistatin that inhibit the signal. However, Smurf1 provides a new mechanism to regulate the BMP signal, acting on signal transduction levels,
Since urf1 is an intracellular protein, it probably functions autonomously. In addition, unlike secretory BMP-binding proteins with limited ligand specificity, S
murf1 confers broad inhibitory activity on the BMP pathway as many type I receptors (ALK1, ALK2, ALK3, and ALK6) signal through Smad1 and Smad5 (42).
Thus, in developing embryos, Smurf1 cooperates with extracellular BMP inhibitors to direct patterning. Whether Smurf1 is regulated in several ways is still an open question.

Of particular note is the localization of Smurf1 mRNA in the animal poles of Xenopus eggs and blastulas. This region forms ectodermal tissues such as the epidermis, cement gland and nervous system, and the fate of the tissue is controlled by the level of BMP signaling received by the cells. Neuronal differentiation occurs spontaneously when the animal polar ectoderm is blocked from the endogenous BMP signal, but cell fate increases with increasing levels of BMP ligand applied to cells or the amount of Smad1 expressed intracellularly. The spectrum of is gradually specified from the nerve to the cement gland to the epidermis. Also, if BMP signaling is reduced in the scheduled ventral mesoderm of Xenopus embryos,
Tissue redesignation occurs as dorsal mesoderm. Since Smurf1 can regulate the responsiveness of cells to BMP signals, it can act on the fate specification of both ectoderm and mesoderm by controlling the levels of Smad1 or Smad5. High levels of Smurf1 can completely block the ability of cells to respond to BMP, whereas moderate or low levels of Smurf1 can
The size of the BMP signal can be regulated through the regulation of ad1 protein levels and alter the nature of the cellular response. As a result, Smurf1 can regulate cell fate decisions in response to BMPs by establishing an intracellular morphogenetic gradient of Smad1 activity. In various animals, maternal mRNA was localized in the egg, which functions as a determinant of cell fate (43). In Xenopus, mRNAs encoding Vg1 (TGFβ member) and Brat / VgT are localized to the egg plant poles, where they specify endoderm and mesoderm (44, 45).
The results presented in this example suggest that Smurf1 localized in the animal pole of the egg may function as a determinant of ectodermal cell fate.

An interesting consequence of overexpression of Smurf1 in the Xenopus polar cap was altered responsiveness of cells to the activin / TGFβ pathway-specific Smad, Smad2. Smurf1 Enhances Cell Sensitivity to Fixed Levels of Ectopic Smad2, Increases Smurf1 Levels, and Induced Mesoderms Are gradually Characterized by the Spemann Organizer, Most Dorsal Mesodermal Species .. This response mimics the effect of elevated levels of Smad2 alone in the animal pole cap. Thus, Smurf1 can simultaneously alter the ability of cells to respond to various TGFβ signals, inhibiting the response to the Smad1 pathway, while stimulating the response to the Smad2 pathway. The different actions of Smurf1 on various TGFβ signaling pathways have important consequences for embryogenetic development, cell proliferation, tissue quiescence and other biological functions regulated by the TGFβ superfamily.

Example 2 Identification and Cloning of Human Smurf2 Human Smurf2 was identified and cloned using the Xenopus Smurf1 sequence in the EST database. Two overlapping EST clones corresponding to HSmurf2 were obtained and used to construct the full length sequence of hSmurf2. Smurf2 is closely related to Smurf1 and shows about 75% homology with the amino acid sequence of hSmurf1. Smurf2 contains a C2 domain at the amino terminus followed by three WW domains and a HECT ubiquitin ligase domain (Figure
12). Smurf2 is closely related to Smurf1 but has an exogenous WW domain downstream of the C2 domain (Figure 13). Smurf2 has been shown to be expressed throughout early development and is present in most adult tissues, but its level is low in spleen and skeletal muscle (Fig. 14A).
And 14B). Furthermore, RT-PCR analysis revealed that Smurf2 was expressed in various cell lines including P19, HepG2 and 293T.

Isolation of human and mouse Smurf2 Several overlapping human clones showing similarities to Smurf1 were identified from the expression sequence tag (EST) database and two overlapping EST clones were used to perform PCR using full-length Smuf2.
rf2 was constructed. For Northern blot analysis, including stop codon, human Smu
A mouse Smurf2 cDNA partial clone encoding 225 amino acids in the open reading frame showing 96% amino acid identity with rf2 was identified in the EST database (LMAGE clone ID 638876).

Construction of plasmids For mammalian expression constructs of Smurf2, the open reading frame was amplified by the polymerase chain reaction (PCR) and in-frame with the amino-terminal Flag or My.
Subcloned into pCMV5 with c-tag (69). For the Smurf2 WW domain, amino acids 163-185 in DWW1, 257-279 in DWW2 and 303-325 in DWW3 were deleted. To generate a non-catalytic ubiquitin ligase mutant of Smurf1, cysteine 716 was replaced with alanine. Tyrosine 2 to generate Smad7 PY mutants
11 was replaced with alanine (Y211A) or the PPPPPY sequence between amino acid sequences 206-212 was deleted (ΔPY). For TβRI-Flag, a Flag tag was introduced at the carboxy terminus of the receptor. All constructs were made by PCR and confirmed by sequencing. The bacterial expression vectors pET15-Smad7-HA and pGEX4T-1-Smurf2 were constructed using convenient restriction sites.

For studies in immunoprecipitation, immunoblotting and affinity labeled mammalian cells, 293T and COS-1 cells were transiently transfected using the calcium phosphate precipitation method or the DEAE-dextran method, respectively. Immunoprecipitation and immunoblotting were performed using anti-HA monoclonal antibody (12CA5, Boeh
ringer), anti-HA rabbit polyclonal antibody (Santa Cruz), anti-Myc monoclonal antibody (9E10 ascites, Developmental Studies Hybridoma Bank), anti-Flag M2 monoclonal antibody (Sigma) or anti-Smad7 rabbit polyclonal antibody.
For anti-Smad7 antibody, GS produced by bacteria encoding amino acids 202-260
Rabbits were immunized with T-Smad7. After adsorbing the antibody to either G protein or A protein-Sepharose, the precipitate was pelleted with TNTE 0.1% (50 mM Tris, pH 7.4, 150 mM Na
Cl, 1 mM EDTA, 0.1% Triton X-100) was washed 5 times, separated by SDS-PAGE, transferred to nitrocellulose, and immunoblotted with an appropriate antibody. Detection was performed using an appropriate horseradish peroxidase (HRP) -conjugated sheep anti-mouse or anti-rabbit secondary antibody and chemiluminescence enhancer (Amersham) to produce His-produced bacteria.
Smad7-HA was incubated with Ni ²⁺ -NTA beads (Quagen) or GST or GST-Smad2 conjugated glutathione beads (Amersham) and TNTE (0.5% Triton X-100)
The precipitate was analyzed by washing three times with and immunoblotting with anti-HA antibody. For affinity labeling, transfected COS-1 cells at 4 ° C
Incubation with 250 pM [ ¹²⁵ I] TGF-b1 for 1 h allowed cross-linking of the receptor to ligands with DSS as described (70). TβRI bound to Smurf2 or Smad7
Was quantified using phosphor imaging (Molecular Dynamics).

Pulse Chase Analysis and Ubiquitination Assay COS-1 cells were transfected with the indicated constructs and 2 days after transfection 15 minutes at 37 ° C. in methionine-free Dulbecco's modified Eagle's medium (DMEM), Cells were treated with 50 mCi / ml [ ³⁵ S] -methionine (Trans [ ³⁵ S] label; Figure 2B) or 150 mCi / ml.
Labeled with [ ³⁵ S] -methionine (FIGS. 5C and D). The cell lysate is then washed twice and 30
Incubation was performed in DMEM with 10% fetal bovine serum in the presence or absence of mM lactacystin (obtained from EJ Corey, Harvard University) or 0.4 mM chloroquine for the indicated times. At each time point, cell lysates were immunoprecipitated with anti-HA monoclonal antibody, separated by SDS-PAGE and visualized by autoradiography. The metabolically labeled proteins were quantified by phosphorimaging. For the ubiquitination assay, use 293T cells with HA-tagged ubiquitin and the receptor Smad7.
And Smad2 combination. The cell lysate was subjected to anti-Smad7 immunoprecipitation, the immunoprecipitate was boiled in 1% SDS for 5 minutes, diluted with TNTE 0.1%, reprecipitated with anti-Smad7 antibody, and then subjected to anti-HA immunoblotting.

Subcellular localization by immunofluorescence deconvolution microscopy Mv1L plated on gelatin-coated Permanox chamber slides (Nunc)
u cells were transfected with the indicated constructs by the calcium phosphate precipitation method. These cells were fixed, permeabilized and reacted with the indicated primary and secondary antibodies (69). Images were acquired on an Olympus 1X70 inverted microscope equipped with fluorescence optics and Deltavision Convolution microscope software (Applied Precision).

Transcriptional Response Assay HepG2 cells were subjected to CMV-βgal as indicated using the calcium phosphate DNA precipitation method.
, 3TP-Lux receptor construct and Smad7-HA construct were transiently transfected. Total DNA was kept constant by adding PCMV5 empty vector. The next day, cells were incubated overnight with or without 100 pM TGFβ. Luciferase activity was measured using the luciferase assay system (Promega) using Berthold
Measured with a Lumat LB 96V luminometer and normalized to β-galactosidase activity.

Results Smurf2 Does Not Regulate Steady Levels of Smad Smads were expressed in 293T cells in the presence or absence of Smurf2. Surprisingly, unlike Smurf1 that directs Smad1 to ubiquitin-mediated proteolysis, expression of Smurf2 did not alter the steady-state levels of Smad1, Smad2, Smad4 or Smad7 (FIG. 15A). To determine if Smurf2 could associate with any of these Smads, cell lysates were subjected to immunoprecipitation of anti-Smarf2 antibody and associated Smads were examined by immunoblot. In cells co-expressing Smad1, Smad2 or Smad4, Smad was found not to co-precipitate with Smurf2 under these conditions
(FIG. 15B). In contrast, a strong interaction between Smurf2 and Smad7 was detected in cells expressing Smad7. To confirm that expression of Smurf2 does not alter Smad7 turnover, a Pulse chase analysis of Smad7 was also performed. Smad7 expressed in the presence or absence of Smurf2 also showed a half-life of approximately 4 hours (
Figure 15C). To analyze the Smad7-Smurf2 association, the GST-Smurf2 fusion protein was purified from bacteria and incubated with bacterially produced Smad7. Take in
It was suggested that Smad7 binds to Smurf2 efficiently under in vitro conditions, and that Smurf2 directly associates with Smad7 (FIG. 15D). We also analyzed the determinants on Smad7 and Smurf2 that mediate this interaction. Smad7 is a PPXY sequence (PY motif
) In its linker region and the sequence is a WW as found in Smurf2.
It may mediate interactions with domains (71). The PY motif of Smad7 replaces a tyrosine residue with alanine (Smad7 (Y211A)) or deletes the motif entirely (S
It was modified by using mad7 (ΔPY). In 293T cells, each Sma of Smurf2
By analyzing the binding of d7 to the mutant, it was revealed that the interaction with Smurf2 was reduced, but all the interactions were not lost (FIG. 15E). This indicates that the PY motif makes an important contribution to the Smad7-Smurf2 interaction, but it is not the only determinant, and other regions within Smad7 may also contribute to these associations. Suggest that there is. We have created mutants of Smurf2 that lack each of the three WW domains. The interaction of Smad7 with a Smurf2 mutant lacking the first WW domain was comparable to that of wild-type Smurf2, but W
Deletion of either W2 or WW3 abolished the interaction with Smad7 and became undetectable (Fig. 15F). Therefore, both WW2 and WW3 domains within Smurf2
It is required to mediate binding to Smad7. Along with this, from these results, S
murf2 binds directly to Smad7 via the WW2 and WW3 domains, whereas Smurf2 does TGFβ
In the absence of signaling, Smad7 is shown to be unsuitable for ubiquitin-mediated proteolysis.

Smad7 Assembles Smurf2 in Complex with the TGFβ Receptor Smad7 interacts with activated type I receptor subunits through the type II TGFβ receptor (TβRII) and type ITGFβ receptor. It binds to the heteromeric complex of (TβRI) (72, 56). The constitutive association between Smad7 and Smurf2 supports the interesting possibility that Smad7 might act to recruit Smurf2 to the TGFβ receptor complex. To investigate this, TGFβ receptors were expressed in COS-1 cells in the presence or absence of Samd7 and Smurf2. Then [ ¹²⁵ I
] -TGFβ was cross-linked to label the receptor. The affinity labeled receptor complex co-precipitated with Smurf2 was visualized by autoradiography.
In the absence or presence of Smad7, little or no coprecipitation of the TGFβ receptor complex with wild-type Smurf2 was observed (FIG. 16). Smurf2 catalytic mutant, 716
It was constructed as Smurf2 in which the cysteine at position was converted to an alanine residue (C716A). This mutation targets a cysteine residue within the HECT ubiquitin-ligase domain that is believed to form a thiol ester bond with ubiquitin. Smurf
A weak interaction with the TGFβ receptor was detected when 2 (C716A) was expressed alone.
(Figure 16). However, in the presence of Smad7, the interaction of Smurf2 (C716A) with the receptor was
Significantly increased. Thus, Smad7 mediates the interaction of Smurf2 with the TGFβ receptor TGFβ receptor.

We also investigated the binding of Smad7 to the receptor. Smad7 interacts with activated type I receptor subunits to induce heteromeric TG
It binds to the Fβ receptor complex (72, 56). In the presence of wild-type Smurf2, it was observed that the amount of TGFβ receptor complex co-precipitating with Smad7 was significantly reduced (FIG. 16, lane).
3). This correlates with a decrease in the total amount of type I receptors present in these transfectants. Since Smad7 binds to the receptor complex through interaction with activated type I receptors, these results suggest that Smurf2 reduces levels of the Smad7-bound receptor complex. It is a thing. Consistent with this finding, no reduction in Smad7-bound receptor levels was observed when Smad7 was co-expressed with a catalytically inactive mutant of Smurf2 (FIG. 16, lane 5). These results suggest that the catalytic activity of Smurf2 mediates the down-regulation of TGFβ receptor bound to Smad7.

Smad7 resides in the nucleus of Smad7, which regulates intracellular localization of Smurf2, and is transported into the cytoplasm by activation of TGFβ signaling, where it interacts with the receptor I subunit to TGFβ. Binds to the receptor (73)
. To determine whether Smad7 recruits Smurf2 to its receptor in intact cells, we examined whether Smad7 could regulate Smurf2 localization. For this, Sma
TβRII, TβRI and Smurf2 (C716A) were expressed in the absence or presence of d7, and the intracellular distribution of the protein was examined by immunofluorescence microscopy.

Smad7 was localized in the nucleus, but in the presence of the signal transduction system, Smad7 was mainly observed in the cytoplasm, and coexisted with the transiently expressed TGFβ receptor in the cytoplasm, resulting in punctate formation. The pattern was shown. Such a distribution of TGFβ receptors has been previously observed by the inventor and others (74-76). Smurf2 was also found predominantly in the nucleus, but its localization was unchanged in the presence of the TGFβ signaling system. But Sma
When d7 was co-expressed, the intracellular distribution of Smurf2 was dramatically altered, the protein was found predominantly in the nucleus and showed a significant coexistence with the TGFβ receptor. These results are S
We suggest that expression of mad7 leads to export of Smurf2 to the nucleus and assembly of Smurf2 into the TGFβ receptor complex.

Smurf2 Induces TGFβ Receptor and Smad7 Degradation Smad7-associated receptors are significantly reduced in the presence of wild-type Smurf2, but not in the presence of mutant Smurf2. We suggest that Smurf2 catalyzes the degradation of the receptor complex. Smurf of the TGFβ receptor complex
To analyze 2-dependent turnover, TβRII and TβRI were coexpressed in 293T cells. Under conditions where TβRII and TβRI efficiently assemble into functional heteromeric complexes, turnover of the entire pool of receptors can be investigated. First, Smur
The steady state level of the receptor in f2-expressing cells was examined (Fig.17A) and the receptor alone or S
It was observed that Smurf2 had a minimal effect on the steady-state levels of type II or type I receptors when co-expressed with murf2. However, in the presence of wild-type Smad7, increased expression of Smurf2 resulted in a large decrease in the steady-state level of type I receptor (Fig. 17B). In contrast, Smurf2 (C716A) had no effect. type
The level of II receptor was also reduced by the expression of wild-type Smurf2, but the extent of this reduction was less than that observed for the type I receptor. This difference is S
mad7 binds to the receptor complex via an activated type I receptor and is a single TβRII
It may be due to the fact that does not interact with. Furthermore,
We investigated whether Smurf2 targets the constitutively activated type I receptor, TβRI (T204D). The receptor mediates TGFβ signaling in the absence of type II receptors and also binds Smad7 (72). Like the receptor complex, wild-type Smurf2 resulted in a decrease in the steady-state level of activated type I receptor, but Smurf2 (C716
A) did not result in this reduction (Figure 17B). To confirm that changes in receptor steady-state levels reflect changes in receptor turnover, the half-life of TβRII and TβRI was analyzed by pulse chase analysis (FIG. 17C). Analysis of the type II receptor revealed that the half-life of the newly synthesized protein was about 1 hour. In contrast, TβRI was more stable, showing a half-life of approximately 4-6 hours. Moreover, the half-life of the type I receptor was unchanged when either Smad7 or Smurf2 was co-expressed with the receptor. But Smurf2 and Sm
When ad7 was co-expressed with the TGFβ receptor complex, the type I receptor half-life was reduced to approximately 1 hour. Thus, Smad7 and Smurf2 promote type I receptor turnover.

Ubiquitin-mediated proteolysis of membrane receptors can be mediated by both proteasomes and lysosomes (60). To investigate whether the Smurf2-dependent increase in TGFβ receptor degradation depends on the function of proteasomes and lysosomes, we investigated lactacystin and chloroquine, which inhibit proteasome and lysosome proteolysis, respectively. Receptor turnover in the presence and absence was investigated. By pulse chase analysis of the receptor,
Each of these inhibitors was found to result in the stabilization of part of the total TGFβ receptor pool (FIG. 17D). These results suggest that both proteasomes and lysosomes contribute to the increased receptor turnover observed in the presence of Smad7 and Smurf2.

In the course of these analyzes of TGFβ receptors, Smad7 protein levels were also measured. In the absence of the TGFβ receptor complex, steady-state levels of Smad7 and turnover were
It was not affected by Smurf2 (see Figure 15). However, in the presence of the TGFβ receptor complex, steady-state levels and half-life of Smad7 were reduced by Smurf2 expression (FIGS. 17B and 17C, respectively). Furthermore, the expression of mutants of Smurf2 (C716A) is S
Such reduction of Smad7 was dependent on the catalytic activity of the Smurf2HECT domain, as it did not change mad7 levels. Smad7 turnover is also stabilized by lactacystin and chloroquine, which, like the receptor complex, causes Sm7
It is suggested that ad7 is subject to degradation by both proteasome and lysosome pathways. Thus, in the presence of TGFβ signaling, Smurf2 induces degradation of Smad7, presumably by targeting the entire receptor-Smad7 complex.

To investigate the ubiquitination of the receptor and Smad7, a modified form of ubiquitin tagged with a HA epitope tag was expressed and ubiquitin conjugates of TβRII, TβRI or Smad7 were subjected to immunoprecipitation experiments and subsequent immunoblotting. Examined. To ensure specificity, immunoprecipitated samples were boiled in SDS and then reprecipitated with the appropriate antibody prior to immunoblotting. Analysis of ubiquitination of Smad7 revealed that ubiquitin bound to the protein in the absence or presence of Smurf2 was low (FIG. 17E). However, when Smurf2 was co-expressed with Smad7 and TGFβ receptor, a high molecular weight ubiquitin conjugate of Smad7 was observed, which was not detected when using the catalytic mutant of Smurf2. In contrast to Smad7, no apparent Smurf2-dependent ubiquitination of type I receptors could be detected. Increased receptor turnover in the presence of Smurf2 and Smad7 and failure to detect ubiquitin-bound receptor reflects rapid degradation of the receptor, which is thought to be ubiquitinated. It may be. Alternatively, ubiquitination of Smad7 may serve as a signal to guide the intact receptor Smad7 complex to the proteosomes. Taken together, the Smad7-dependent association of Smurf2 with the TGFβ receptor results in proteasome- and lysosome-mediated TGFβ receptor complexes and Sma.
It is shown that degradation of d7 occurs.

Our studies have shown that Smurf2 association enhances the inhibitory activity of Smad7 and that Smurf2 can assemble through the Smad7 into the TGFβ receptor complex, leading to the degradation of the receptor. . This suggests that ubiquitin-mediated degradation may contribute to the inhibitory activity of Smad7 in the TGFβ pathway. In order to investigate this, first, Smad7 (Y211A) (which has a weak interaction with Smurf2) (
It was examined whether the ability of Smurf2 to aggregate at the TGFβ receptor (see FIG. 15D) is changed. At this time, wild type Smad7 or Smad7 (Y211A) was replaced with Smurf2 (C71
6A) was co-expressed with TGFβ receptor in the presence or absence of 6A) and receptor interaction was examined by analyzing affinity labeled receptors co-precipitated with Smad7 or Smurf2. The interaction between the TGFβ receptor complex and Smad7 (Y211A) is associated with wild-type Smad7.
Which was comparable to that observed in (Fig. 18A). This suggests that this mutation in the PY motif does not affect the interaction of Smad7 with the TGFβ receptor. However, the ability of Smurf2 (C716A) to associate with the TGFβ receptor was
It is substantially reduced in the presence of ad7 and correlates with the reduced efficiency observed in the above-mentioned interaction between Smurf2 and Smad7 (Y211A). Then the endogenous of Smad7 (Y211A)
We examined whether the ability to inhibit TGFβ signaling in HepG2 cells expressing Smurf2 was altered. For this reason, 3TP has been characterized in detail.
The -lux reporter construct was used to investigate TGFβ signaling. Expression of wild-type Smad7 strongly reduced TGFβ-dependent induction of the reporter (Fig. 1).
8B). In contrast, the Smad7 (Y211A) mutant had a substantially reduced inhibitory activity despite being able to efficiently interact with the TGFβ receptor complex. This is
We suggest that the binding of Smurf2 to Smad7 enhances the inhibitory activity of Smad7 on the TGFβ signaling pathway.

Previous studies have shown that Smad7 binding to the TGFβ receptor complex blocks Smad2 access to the receptor. Since Smad7 (Y211A) efficiently binds to the TGFβ receptor, when the mutant protein was expressed at a higher level, TGFβ
An attempt was made to investigate whether or not the inhibitory activity on β signal pathway could be maintained. In order to investigate this, the expression level of Smad7 was changed, and Smad7 and Smad7 (Y211A
) And the efficiency of blocking TGFβ signaling relative to Wild-type Smad7 strongly inhibited the TGFβ signaling pathway at the lowest amount tested, whereas one Smad7
(Y211A) had a very low inhibition efficiency (Fig. 18C). However, at the highest dose tested, Smad7 (Y211A) was able to inhibit TGFβ responsiveness.
These results indicate that Smad7 (Y211A) retains some inhibitory activity, presumably due to its ability to bind to receptors and block Smad2 or Smad3 access. Taken together, these data suggest that Smurf2 cooperates with Smad7 to enhance its inhibitory activity.

Discussion The above examples confirmed the identification of the novel ubiquitin ligase Smur2.
The ligase acts in cooperation with Smad7 to direct the TGFβ receptor for degradation. Smur
The binding of f2 to the TGFβ receptor was very weak and could not affect receptor turnover. However, in the presence of Smad7, Smurf2 formed an effective complex with the heteromeric TGFβ receptor, which directed the receptor for degradation. Since Smad7 binds directly to Smurf2 and associates with the TGFβ receptor complex, these results suggest that Smad7 functions as an adapter protein that mediates the interaction of Smurf2 with the TGFβ receptor complex. (FIG. 18D). Consistent with this view, mutations in the PY motif of Smad7 that interfere with the interaction between Smad7 and Smurf2 also blocked Smad7's ability to assemble Smurf2 into a complex with the receptor. This mutant form of Smad7 was also confirmed to be able to block TGFβ signaling, which indicates that Smurf
It was suggested that 2 is involved in mediating the inhibitory function of Smad7. Since Smad7 binds to the activated TGFβ receptor in competition with R-Smad, the above synergism is particularly relevant to Sma.
It may be important if d7 is expressed transiently or at low levels (53-58).
Therefore, by directing the receptor for degradation, Smurf2 may provide a mechanism for persistent inactivation of the Smad7-binding receptor complex. Consistent with the Smurf2-dependent degradation of the TGFβ receptor, a TGFβ signaling-dependent degradation of Smad7 was also observed.

The present invention is not limited in scope to the particular embodiments described herein. It will be apparent to those skilled in the art from the above description and the accompanying drawings that various modifications of the present invention can be made to the description of the present specification. Such modifications are intended to fall within the scope of the appended claims.

Further, it is to be understood that all base size, amino acid size, and all molecular weight and molecular size values described herein are approximate.

All patents, patent applications, publications and other references cited herein are hereby incorporated by reference in their entireties. Cited references

[Brief description of drawings]

FIG. 1 shows the E3 ubiquitin ligase encoded by Smurf1. The protein sequences of Xenopus and human Smurf1 were cloned from the yeast (S. pombe) pub.
Compared to 1, they are shown in the figure as SMURF1, hSMURF1 and PUB1, respectively. Identical amino acids are shaded dark gray and conservative substitutions are shaded light gray. Based on the primary structure, Smurf1 and pub1 are members of the Hect family of E3 ubiquitin ligases and display multiple conserved features of the family: one lipid / Ca2 + binding domain at the N-terminus (residues). 22-37), two WW protein interaction domains are located at 236-271 and 282-311 (shown in bold) and a catalytic Hect domain starting at residue 347 and extending to the C-terminus. Alignment is Clust
Al W analysis (Mac Vector) was performed.

2A and 2B show northern blot results of mSmurf1 expression in embryonic and adult mouse tissues. Equal amounts of poly A + mRNA were analyzed from the indicated stages and tissues. (2A) Embryonic tissue-This blot showed post-mating embryo days. (2B) The mature tissues shown are T, testis; K, kidney; M, skeletal muscle; L, lung; Sp, spleen; Br, brain; H, heart.

FIG. 3A and 3B show developmental expression of Xenopus Smurf1. (A) RT-PCR of embryonic cDNA at each stage revealed that Xenopus Smurf1 is a maternal mRNA and is present at the highest level in eggs, blastula and early gastrula stage. Indicated. The zygote Smurf1 mRNA level decreases during gastrulation, but maintains stable expression until the migrating tadpole stage. Numbers on each lane correspond to Niuwkoop and Faber stages; 7 and 9
, Spores; 11 and 13, gastrula; 15 and 20, neurula; 25 and 35,
It is a tadpole. Ordinin decarboxylase (ODC) mRNA, which is universally expressed in cells, was assayed and canonicalized for RNA recovery. RT- is a fake cDN
PCR for A (non-reverse transcriptase). (B) Whole-mount in situ hybridization of Smurf1 in developing Xenopus embryos. At the egg and blastocyst stage, the Smurf1 transcript is localized to the animal pole (bracket in the egg). Expression diffuses throughout the ectoderm and is involute into the marginal zone of the gastrula; diagrams from animal poles (an) and plant poles (Vg).
At neuroembryonic stage 17, there is some enrichment of transcripts in the neural fold.
In tadpole stages 25 and 35, Smurf1 expression includes brain (b), eye (e), ear vesicles (o), segment (s), pharyngeal fossa (p) and developing kidney (k).

4A, 4B, 4C and 4D show that Smurf1 expression is Smad1 and S in mammalian cell lines.
It has been shown to result in the selective reduction of steady-state levels of mad5. Cells were transiently transfected with the indicated expression vector (DNA amount, μg) and after 2 days an aliquot representing approximately 0.4% of total cell lysate was subjected to SDS-PAGE and immunoblotting. To determine the steady-state level of Smad, blots as directed,
It was probed with the appropriate Smad antibody. Flag-hSmurf1 expression levels were confirmed by probing whole cell lysate blots with anti-Flag monoclonal antibody. (A) COS-1 or 293T cells were loaded with a constant amount of pCMV5-Smad1 and increasing concentrations of pCMV5.
-Flag-hSmurf1 was used to transfect as indicated. To determine Smad1 steady-state levels and hSmurf1 expression, Western blots of whole cell lysates from both cell lines were probed with anti-Smad1 and anti-Flag antibodies (α-Sma
d1 and α-Flag blots). (B) 293T cells were transiently transfected with pCMV5-Smad1, wild type or activated (Q203D) pCMV5-ALK6-HA and increasing concentrations of pCMV5-Flag-hSmurf1, as indicated. . Smad1 steady-state levels were tested by immunoblotting whole cell lysates with anti-Smad1 antibody on Western blots (α-Smad1). Expression levels of hSmurf1 and wild-type or activated-ALK6 were determined by immunoblotting whole cell lysates with anti-Flag (α-Flag) or anti-HA (α-HA) antibodies. (C) 293T cells were transfected with a fixed amount of pCMV5-Smad1 or pCMV5-Smad2 as well as with the indicated concentrations of pCMV5-Flag-hSmurf1. Smad1 (α-Smad1 blot), Smad2 (α-Smad2 blot) and Flag -hS in cell lysates
The murf1 (α-Flag blot) steady-state protein levels were determined as above. (D) 293T cells were treated with Smad in pCMV5 in the absence or presence of pCMV5-Flag-hSmurf1.
Transfection with 1, Smad3, Smad4 or Smad5. Steady-state levels of each Smad in whole cell lysates were analyzed by Western blot with anti-Smad against Smad1 and Smad5.
It was determined by incubating with 1 antibody, anti-Smad3 antibody against Smad3, and Smad4 antibody for Smad4 detection (α-Smads blot). Equivalent hSmurf1 expression,
Confirmed as above (α-Flag blot).

5A, 5B, and 5C show that hSmurf1 regulates Smad1 turnover and ubiquitination: hSmurf1 enhances Smad1 turnover. (A) COS-1 cells were transiently transfected with pCMV5-Smad alone or Flag-labeled hSmurf1 (F / hSmur
f1) and transfected using lopofectAMINE. Two days later, the transfectants were subjected to pulse chase analysis with [ ³⁵ S] methionine. Cells were lysed and subjected to anti-Smad1 immunoprecipitation at the indicated times in the chase. Immunoprecipitate SDS-P
It was decomposed by AGE and visualized by autoradiography (left panel). Radiolabeled Smad1 was also quantified by phosphorimaging and the results were plotted as the amount of [ ³⁵ S] methionine labeled Smad1 present at each time point compared to the level at time point 0 (right panel). (B) Smad1 ubiquitination in 293T cells. Cells were transiently transfected with the indicated combinations of HA-labeled ubiquitin (HA-Ub), pCMV5-Smad1, and Flag-labeled hSmurf1 wild type (WT) or ubiquitin ligase mutant (CA). Used to transfect. Two days after transfection, lysates were subjected to α-Smad1 immunoprecipitation (α-Smad IP), followed by SDS-PAGE, and immunoblot (α-HA blot) with α-HA monoclonal antibody. It was Protein bands immunoreactive with α-HA are marked with square brackets. Smad1 or Flag-hSmurf
Expression of 1 was confirmed by subjecting whole cell lysates to immunoblotting with anti-Smad1 polyclonal antibody (α-Smad1 blot) or anti-Flag monoclonal antibody (α-Flag blot), respectively. (C) Loss of Smad1 steady-state levels by hSmurf1 requires intact ubiquitin ligase activity of the Hect domain. 293T cells were transfected with Smad1 and increasing amounts of wild type (WT) or ubiquitin ligase mutant (C710A). As shown in FIG. 3, whole cell lysates were analyzed for Smad1, hSmurf1 or hSmurf1 (C710A) protein by immunoblotting with appropriate antibodies.

6A, 6B, and 6C show the interaction of Smurf1 and Smad. (A) Smurf1 interacts with Smad1. A yeast two-hybrid assay was performed on yeast co-transformed with Xenopus Smurf1 (Xsmurf) in combination with Smad1, Smad2, lamin or vector alone. Only the combination of Smad1 and Smurf1 showed significant β-galactosidase activity (left panel, picture of stained yeast colonies double assayed). Co-immunoprecipitation for in vitro translated proteins was performed in vitro by immobilizing 35S-labeled Smurf1 on an anti-Flag affinity gel matrix.
35S-Mettrace label-Flag label obtained by tro translation Smad1 (35S-F / Smad1), Smad2 (35S-F
/ Smad2) or Smad4 (35S-F / Smad4). After washing, bound proteins were eluted and analyzed by SDS-PAGE (
Right panel). (B) hSmurf1 interacts selectively with Smad1 and Smad5. Smad1 (S1), Smad5
(S5) or Smad2 (S2) containing pCMV5 expression vector alone or wild type (WT) or ubiquitin ligase mutation (CA) Flag tagged hSmurf1 (F / h
Smurf1) were used together to transiently transfect 293T cells. To test the interaction of Smad1 or Smad5 with hSmurf1, anti-Flag immunoprecipitate blots were probed with anti-Smad1 polyclonal antibody (α-Smad1 blot). S
To test the interaction between mad2 and hSmurf1, anti-Flag immunoprecipitate blots were probed with anti-Smad2 polyclonal antibody (α-Smad2 blot). The levels of hSmurf1 in immunoprecipitates and Smad1, Smad5 and Smad2 in whole cell lysates are shown in the lower two panels. (C) Specificity of hSmurf1 action on Smad1. Nedd4, a ubiquitin ligase associated with hSmurf1, does not interact with Smad1 and does not reduce steady-state levels of Smad1. 293T cells were transiently co-transfected with Smad1 and Flag-tagged hSmurf1 (WT or CA) or Nedd4 (WT) or ligase mutants (CS) as indicated. Smad1 coprecipitation with hSmurf1 (left panel) was confirmed as described above. Ne
To confirm the Smad1 interaction with dd4, cell lysates were immunoprecipitated with Hect-Nedd4 polyclonal antibody (α-Nedd IP), followed by immunoblotting with anti-Smad1 polyclonal antibody (α-Nedd IP). Smad1 blot), shown in the right panel. To confirm the level of Nedd4 expression in the sample, each blot was WW2 Nedd4
Reprobed with a polyclonal antibody (α-Nedd4 blot). Smad1 levels in whole cell lysates are shown in the bottom panel by Western blot.

FIGS. 7A and 7B show that Smurf1 dorsalizes ventral mesoderm and neuralizes ectoderm. (A) Four cell Xenopus laevis embryos were injected with Xenopus Smurf1 mRNA into the peripheral zones of two blastomere cells (50 pg Smurf mRNA per cell). Tadpoles generated from injected embryos have an ectopic dorsal axial structure.
) Was formed (lower left panel, arrow). 100 pg Smad1 with 50 pg Smurf mRNA
Simultaneous injection rescued ectopic axial structure in all cases (lower tadpole)
. Smurf1 with ventral marginal zone (VMZ) strips for early gastrulation, wild-type embryos, or VMZ co-injected with Smurf1 mRNA alone or Smad1 at 4-cell stage
Were transplanted from another set of embryos injected with. Upon early gastrulation as described, the VMZ transplant was excised, then control embryos were cultured until they reached the medium-tadpole, stage 28, total VMZ RNA was prepared, by RT-PCR, and erythrocyte-specific by RT-PCR. Target α-
Globulin, muscle-specific actin, and general cell mRNA, were assayed for expression of eF1-a as a control for RNA recovery (lower right panel). Control PCR reaction is embryo
RNA was performed with or without reverse transcription. (B) Smurf1 and / or Smad1 mRNA was or was not injected into the fertilized egg animal pole. Animal pole caps were transplanted to blastocyst stage 8, cultured to mid-gastrointestinal embryo stage 11, then total RNA was prepared and assayed by RT-PCR to obtain cement gland (cement gland).
) Expression of markers, XAG, and general neural markers, NCAM was detected. The control was as described above.

8A, 8B, and 8C show that Smurf1 modifies the ability of embryonic cells to respond to Smad1 and Smad2. (A) Smurf1 blocks Smad1 induction of abdominal mesoderm. A fixed amount (1 ng) of Smad
1 mRNA alone or together with increasing amounts of Smurf1 was injected into the fertilized egg animal pole and induction of ventral mesoderm in animal pole cap transplants by RT-PCR for Xhox3 and Xcad3.
Assays were performed using primers for the homeodomain gene. From left to right in the lane, no or 100 injections of Smurf1 mRNA in the animal pole cap,
0, 25, 50, 100, and 200 pg. (B) Smurf1 enhances dorsal mesoderm induction by Smad2. Injecting a constant amount (50 pg) of Smad2 mRNA, alone or with increasing amounts of Smurf1, is a homeodomain gene that expresses dorsal germ induction in myoD marking muscle and most dorsal mesoderm types It was monitored by the expression of goosecoid, Spemann Organizer. Note that with increasing Smurf1 dose, goosecoid expression was triggered from undetected levels. The animal pole cap was injected as shown in panel a. (C) Dose response of animal pole caps to Smad2. The animal pole cap was injected with synthetic mRNA for Smad2. 50 pg of Smad2 induces MyoD, 100 pg or more of injected Sm
Reached the maximum level with ad2. Goosecoid was induced at a minimum Smad2 dose of 250 pg. Goosecoid levels induced by 250 pg of Smad2 alone were 50 pg of Smad2
It was identical to the level induced by the combination of 100 pg Smurf1 (panel 1b). The rightmost two lanes with and without reverse transcription (RT +), respectively, in all panels
Corresponds to PCR of (RT-) wild-type embryo RNA. The eF1-a controls for all panels were as in Figure 7.

[Figure 9]   FIG. 9 shows the cDNA sequence of human Smurf1 [SEQ ID NO: 1].

[Figure 10]   FIG. 10 shows the protein sequence of human Smurf1 [SEQ ID NO: 2].

FIG. 11   FIG. 11 shows the cDNA sequence of human Smurf2 [SEQ ID NO: 3].

FIG. 12 shows the protein sequence of human Smurf2 [SEQ ID NO: 4]. Smurf2 is a member of the HECT E3 ubiquitin ligase family. The C2 (overline), WW (shaded) and HECT (double overline) domains and the Cys716Ala mutation (boxed) are indicated.

FIG. 13 shows a comparison of Smurf1 and Smurf2 proteins. Schematic representation of Smurf1 and Smurf2. The degree (%) of amino acid identity between the C2, WW and HECT domains is shown.

14A and 14B show that Smurf2 is expressed in mouse tissues. (A) Smurf2 is expressed throughout mouse embryogenesis. A mouse embryo Northern blot (Clontech) was probed with a 1 Kb XhoI / NotI fragment encompassing the 3'UTR of mouse Smurf2. (B) Expression of Smurf2 in mature mouse tissues. Multi-Tissue Northern Blot (Clontec
h) was probed with a fragment of mouse Smurf2 as in A.

15A, 15B, 15C, 15D, 15E and 15F show that Smurf2 interacts with Smad7. (A and B) Smurf2 expression does not reduce steady-state levels of Smad. 293T cells were treated with Flag labeled Smad1, Smad2, Smad4 or HA labeled Smad7 alone or Myc labeled Smurf
Used with 2 to transfect. Sm immunoblotted aliquots of whole cell lysates to detect expression of Smurf2 and Smad (Figure 15A) or immunoprecipitated with anti-Myc antibody followed by co-precipitation by anti-Flag or anti-HA immunoblot.
ad was detected (Fig. 15B). The migration of anti-Myc heavy chain (IgH) was marked. (C) Expression of Smurf2 does not alter Smad7 turnover. Smad7-HA alone or Fl
COS-1 cells transfected with agSmurf2 were pulse labeled with [ ³⁵ S] -methionine and then chased with medium containing unlabeled methionine for the times indicated. [ ³⁵ S] -labelled Smad7-HA in anti-HA immunoprecipitates was quantified by phosphorimaging, comparing levels in control (squares) and Smurf2-expressing cells (circles) with the amount present at time 0. Plotted. Data represent the mean ± SD of 2 experiments. (D) In vitro interaction between Smurf2 and Smad7 expressed in bacteria. His- produced by bacteria
Smad7-HA protein was incubated with Ni ²⁺ -NTA, GST and GST-Smurf2.
Bound material was visualized by SDS-PAGE and immunoblotted with anti-HA antibody.
GST protein levels were determined by Coomassie staining (bottom panel). (E) The PY motif of Smad7 is an important determinant mediating the interaction with Smurf2.
293T cells were treated with FlagSmurf2 alone or wild type (WT) or mutant Y211A (Y
A) or the ΔPY version of Smad2-HA was used together and transfected.
Cell lysates were subjected to anti-Flag immunoprecipitation to immunoprecipitate Smad7 protein with anti-Smad7
Detected by immunoblot with antibody. Smad7 expression was confirmed by immunoblotting an aliquot of whole cell lysate (bottom panel). (F) The Smurf2 WW domain is required for binding to Smad7. 293T cells, Smad7-HA
And wild-type (WT) or mutant versions of Flag-Smurf2 (ΔWW1, ΔWW2
Or ΔWW3). Anti-Flag Cell Lysate
Upon immunoprecipitation, the immunoprecipitating Smad7 protein was detected by immunoblotting with anti-HA antibody. Smad7 expression was confirmed by immunoblotting an aliquot of whole cell lysate (bottom panel).

FIG. 16 shows that Smad7 assembles Smurf2 into the TGFβ receptor complex.
COS-1 cells, as shown, TβRII, TβRI-HA, Smad7-HA and wild type (WT)
Alternatively, various combinations of mutant (C716A) Flag-Smurf2 were used for transfection. Cells were affinity labeled with [ ¹²⁵ I] TGFβ and lysates were immunoprecipitated with anti-Flag or anti-Smad7 antibody. The co-precipitated receptor complex was visualized by SDS-PAGE and autoradiography. The amount of co-precipitating TβRI was quantified by phosphorimaging (right panel). Receptor expression was confirmed by visualizing an aliquot of whole cell lysate by autoradiography. Smurf2 and Smad
Seven levels were confirmed by immunoblotting with aliquots of whole cell lysates using anti-Flag and anti-HA antibodies, respectively.

17A, 17B, 17C, 17D and 17E show that Smurf2 induces degradation of the TGFβ receptor and Smad7. (A) In the absence of Smad7, Smurf2 expression does not reduce receptor steady-state levels. 293T cells were shown to have TβRII-HA, TβRI-HA and varying amounts of Flag-Smur as indicated.
Transfections were carried out with various combinations of f2 (plasmid DNA, unit μg).
Protein expression levels were determined by immunoblotting, as indicated, aliquots of whole cell lysates utilizing the appropriate antibodies. (B) In the presence of Smad7, Smurf2 reduces steady-state receptor levels. 293T
Cells were treated with either Smad7-HA, TβRII-HA and TβRI-HA (left panel) or constitutively active type I receptor, TβRI-HA (T204D) (right panel), and increasing amounts of wild type. (WT) or the mutant Flag-Smurf2 (C716A). Steady-state levels of the receptors, Smad7 and Smurf2, were shown, as indicated by anti-HA.
Or determined by anti-Flag immunoblot. (C) Smurf2 increases the turnover rate of the receptor complex. TGFβ receptor (T
βRII-HA and TβRI-HA) alone or with Smad7-HA, Flag-Smurf2 or both, were pulse-labeled with [ ³⁵ S] -methionine and then shown. Chase with medium containing unlabeled methionine for hours. Cell lysates were subjected to anti-HA immunoprecipitation and the amount of labeled receptor and Smad7 was quantified by phosphorimaging and plotted relative to the amount present at time 0. (D) Proteasomes and lysosomal inhibitors block Smurf2-induced receptor complex degradation. TGFβ receptors (TβRII-HA and TβRI-HA), Smad7-
COS-1 cells transfected with HA and Flag-Smurf2 were pulse-labeled with [ ³⁵ S] -methionine and then in the absence of inhibitor or in the presence of 30 mM lactacystin or 0.4 mM chloroquine. Originally, I chase for the indicated time. Cell lysates were subjected to anti-HA immunoprecipitation and receptor and Smad7 levels were tested for SDS-PAG.
Visualized by E and autoradiography. (E) Smurf2 induces ubiquitination of Smad7 in the presence of the receptor. 293T cells were treated with HA-labeled ubiquitin in various combinations of Smad7, TβRII and TβRI-Fla.
g, and wild type (WT) or mutant (C716A) Myc-Smurf2 were transfected as indicated. Cell lysates were subjected to double immunoprecipitation with anti-Smad7 antibody and then immunoblotted with anti-HA antibody. Protein expression in aliquots of whole cell lysates was confirmed by immunoblot.

18A, 18B, 18C and 18D show that the association of Smurf2 and Smad7 enhances Smad7 inhibitory activity. (A) Smad7 (Y211A) binds to the TGFβ receptor but with a reduced ability to assemble Smurf2 into the receptor complex. TGFβ receptors (TβRII-HA and TβRI-HA) on COS-1 cells
And wild-type (WT) or mutant (Y211A) Smad7 / HA using Flag-Smurf2
Transfected in the absence and presence of (C716A). Cells to [ ¹²⁵ I] TGFβ
Was affinity labeled with and the lysates were immunoprecipitated with anti-Smad7 or anti-Flag antibodies. The co-precipitating receptor complex was visualized by SDS-PAGE and autoradiography. Total receptor expression was determined by autoradiography and Smad
7 and Smurf2 protein levels were measured in aliquots of whole cell lysate with anti-HA or anti-F.
Confirmed by lag immunoblot. (B and C) Smad7 (Y211A) is a wild-type S that inhibits TGFβ-dependent transcriptional activation.
Not as effective as mad7. HepG2 cells were transfected with the 3TP-Lux reporter alone or with varying concentrations of wild type (WT) or mutant (Y211A) Smad7-HA. In (B), 0.3 ng / ml of each Smad7 plasmid was used. Cells were incubated in the presence or absence of TGFβ1 to normalize luciferase activity to b-galactosidase activity and plot the mean ± SD of triplicates from a representative experimental value. (D) A model for degradation of the TGFβ receptor complex mediated by Smad7 and Smurf2. S
mad7 binds directly to Smurf2 and associates with the TGFβ receptor complex. Therefore, Sma
d7 functions as an adapter protein that mediates the degradation of the TGFβ receptor complex.

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) A61K 39/395 A61K 48/00 4C084 45/00 A61P 19/08 4C085 48/00 25/00 4C086 A61P 19 / 08 27/02 4H045 25/00 43/00 105 27/02 C07K 16/40 43/00 105 C12N 1/15 C07K 16/40 1/19 C12N 1/15 1/21 1/19 9/00 1/21 C12Q 1/02 5/10 G01N 33/15 Z 9/00 33/50 Z C12Q 1/02 33/53 D G01N 33/15 C12N 15/00 ZNAA 33/50 5/00 A 33/53 A61K 37/02 (81) Designated countries EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ , CF, CG, CI, CM, GA GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, MZ, SD, SL, SZ, TZ, UG, ZW), EA (AM, AZ , BY, KG, KZ, MD, RU, TJ, TM), AE, AG, AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, CR, CU , CZ, DE, DK, DM, EE, ES, FI, GB, GD, GE, HR, HU, ID, IL, IN, IS, JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MA, MD, MG, MK, MN, MW, MX, MZ, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL , TJ, TM, TR, TT, TZ, UA, UG, US, UZ, VN, YU, ZA, ZW (72) Inventor Tomsen, Gerald, H. United States 11777-1304 Port Jefferson, New York, Bayview Terase 201 (72) Inventor Wallana, Jeffrey Canada M6J2M5Ontario Toronto, Claremonto Street 106 BI F-term (reference) 2G045 AA40 DA12 DA13 DA36 FB02 FB03 4B024 AA01 AA11 BA07 CA04 CA09 DA02 EA02 EA04 GA11 4B050 CC01 CC03 CC04 DD11 LL03 4B063 QA05 QQ08 QQ13 QQ20 QQ40 QR51 QR77 QR80 QS38 4B065 AA90X AA91X AA93X AA93Y AB01 AC14 BA02 CA27 CA46 4C084 AA02 AA06 AA07 AA13 AA17 BA01 BA08 BA22 BA23 MA01 NA14 ZA012 ZA332 ZA672 ZA922 ZA962 ZB212 4C085 AA13 AA14 CC32 EE01 GG01 4C086 AA01 AA02 AA03 EA16 MA01 MA04 NA14 ZA01 ZA33 ZA92 ZA96 ZB21 4H045 AA10 AA11 AA30 FA74 FA75 DA72 FA75 DA72 FA75 DA72

Claims (36)

[Claims] 1. An isolated Smurf protein. 2. The Smurf protein according to claim 1, which is derived from human. 3. The Smurf protein according to claim 1, which is Smurf1. 4. The Smurf according to claim 3, which has a mutation corresponding to C710A.
1. 5. Smurf1 according to claim 3, comprising at least 10 consecutive amino acid residues in the sequence shown in SEQ ID NO: 2. 6. The Smurf1 according to claim 3, which comprises the amino acid sequence shown in SEQ ID NO: 2. 7. The Smurf protein according to claim 1, which is Smurf2. 8. The Smurf according to claim 7, which has a mutation corresponding to C716A.
2. 9. The Smurf2 protein of claim 7, comprising at least 10 consecutive amino acid residues in the sequence set forth in claim 4. 10. The Smurf2 protein according to claim 7, which comprises the amino acid sequence shown in SEQ ID NO: 4. 11. An isolated nucleic acid encoding the Smurf protein of claim 1. 12. The Smurf protein is a human Smurf protein.
1. The nucleic acid according to 1. 13. The nucleic acid according to claim 11, wherein the Smurf protein is Smurf1. 14. The nucleic acid according to claim 13, which has a mutation corresponding to C710A. 15. The nucleic acid of claim 13, wherein Smurf1 comprises at least 10 consecutive amino acid residues in the sequence shown in SEQ ID NO: 2. 16. The nucleic acid according to claim 13, wherein Smurf1 comprises the amino acid sequence shown in SEQ ID NO: 2. 17. The nucleic acid of claim 16, having the nucleotide sequence set forth in SEQ ID NO: 1. 18. The nucleic acid according to claim 11, wherein the Smurf protein is Smurf2. 19. The nucleic acid according to claim 18, which has a mutation corresponding to C716A. 20. The nucleic acid of claim 18, wherein the Smurf2 protein comprises at least 10 consecutive amino acid residues in the sequence shown in SEQ ID NO: 4. 21. The nucleic acid of claim 18, wherein the Smurf2 protein comprises the amino acid sequence set forth in SEQ ID NO: 4. 22. The nucleic acid of claim 18, having the nucleotide sequence set forth in SEQ ID NO: 3. 23. A vector containing the nucleic acid according to claim 11. 24. A host cell containing the vector of claim 23. 25. A method for producing Smurf protein, which comprises growing the host cell of claim 23 under conditions that allow expression of the Smurf protein from the vector. 26. A method for producing a Smurf protein, which comprises growing the host cell of claim 24 under conditions that allow expression of the Smurf protein from the vector. 27. A transgenic non-human animal expressing a human Smurf protein. 28. A method of inhibiting a bone morphogenetic protein or tumor growth factor β activation pathway in a cell, the cell allowing expression of Smurf from the vector of claim 23 introduced into said cell. Growing under the conditions described below. 29. A method of inhibiting a bone morphogenetic protein or tumor growth factor β activation pathway in a cell, the cell allowing expression of Smurf from the vector of claim 24 introduced into said cell. Growing under the conditions described below. 30. A method of promoting a bone morphogenetic protein or tumor growth factor β activation pathway in a cell, comprising suppressing the expression of endogenous Smurf in the cell. 31. A method of screening for modulators of Smurf activity, the method comprising: Smuf in the presence of a test compound for Smurf activity in the absence of the test compound.
A method comprising detecting a change in rf activity. 32. The method of claim 31, wherein the Smurf activity is ubiquitination of Smad protein in host cells. 33. The method of claim 31, wherein the Smurf activity is the interaction of the Smurf WW domain with the PPXY domain of the Smad protein. 34. The method of claim 33, wherein the test compound is screened for the ability to inhibit the interaction. 35. An antibody that specifically binds to Smurf protein. 36. An oligonucleotide or a nucleic acid that specifically hybridizes to a nucleic acid having a sequence encoding Smurf under highly stringent conditions.