| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biochemistry and Biophysics |
Departments of Biochemistry [O. T., G. C.] and Medicine [G. C.], Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
50 kDa containing the COOH-terminal region. On the basis of independent means of identification, two adjacent sequence motifs have been identified in ErbB-2 that are required for GA-induced degradation. These motifs encompass residues 776–783 and 784–786 within the NH2-terminal lobe of the ErbB-2 kinase domain. This is also a region in which the epidermal growth factor receptor and ErbB-2 kinase domains differ significantly in sequence. Although mutations in this region abrogate GA-induced ErbB-2 degradation, the tyrosine kinase activity of ErbB-2 is not disrupted. Interestingly, these ErbB-2 mutants are specifically resistant to GA-induced degradation but retain sensitivity to other drugs, such as staurospore and curcumin, which are also able to provoke ErbB-2 degradation. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, heparin-binding EGF, amphiregulin, betacellulin, epiregulin, and epigean) are recognized by the EGFR (ErbB-1), whereas the heregulins bind directly to the ErbB-3 and ErbB-4 receptors (2)
. In the case of ErbB-3, dimerization with ErbB-2 is essential because ErbB-3 does not have tyrosine kinase activity in its absence (4)
. Heterodimerization and signaling through ErbB-2 enhance the strong proliferative effect of all these growth factors (5, 6, 7)
. Overexpression of ErbB-2 has been demonstrated in human breast, ovarian, prostate, and lung cancers. HER2 gene amplification is usually, but not always, responsible for a 10–100% increase in the ErbB-2 protein content of tumor cells, which results in constitutive activation of signaling pathways leading to uncontrolled proliferation, increased cell motility, invasiveness, and tumorigenesis (8 , 9) . ErbB-2 overexpression correlates with an aggressive disease form, particularly in the case of breast cancer, and a poor prognosis, making it a significant target for therapeutic intervention (10) .
Monoclonal antibodies against the ErbB-2 ectodomain have been shown to be effective against tumors that overexpress ErbB-2 and are in use clinically (11, 12, 13) . However, their mechanism of action is poorly understood but appears to involve the accelerated internalization and degradation of ErbB-2 (11) . Another potential therapeutic approach is to use specific tyrosine kinase inhibitors that block ErbB-2 kinase activity and at the same time increase the rate of ErbB-2 degradation (14, 15, 16) . ErbB-2 degradation is significantly increased in the presence of some nonselective tyrosine kinase inhibitors, such as GA or curcumin (17) .
GA was first described as an inhibitor of tyrosine kinase activity. It belongs to the group of benzoquinoid ansamycin antibiotics and is closely related in structure to herbimycin A and macbecin (18) . Ansamycin antibiotics were shown to have selective cytotoxicity against several malignant tumor cell lines, revert to normal the morphology of transformed fibroblasts, and reduce tumorigenicity in murine models. In cell lines, GA increases the rate of degradation of several protein kinases, including ErbB-2, Src family members, Raf-1, focal adhesion kinase, Met, and Bcr-Abl (19, 20, 21, 22, 23, 24) . Although the molecular mechanism of GA action is not clear, it is thought to bind to Hsp90 and inhibit the binding of this chaperone to protein kinases and thereby provoke the metabolic destabilization and degradation of Hsp90 targets (25 , 26) .
Previous studies have demonstrated that the ErbB-2 kinase domain is necessary for GA-induced degradation of this receptor (25 , 27) . It has been proposed that Hsp90 stabilizes ErbB-2 through interaction with its kinase domain, and this interaction is perturbed in the presence of GA. In this study, evidence is presented that the ErbB-2 kinase domain contains a cryptic motif that facilitates GA-induced receptor degradation.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Culture and Transfection.
Human mammary SKBr3 cells were grown in 5% CO2 at 37°C in McCoy medium with 10% FCS. Cos7 cells were grown in DMEM with 10% FCS. SKBr3 cells were grown to 90% confluency and treated with the indicated drugs. For transient transfection experiments, Cos7 cells were grown to 80% confluency and transfected with LipofectAMINE according to manufacturer’s recommendations. The cells were then grown for 48 h before assays were conducted.
Construction of ErbB-2 Mutants.
ErbB-2 mutants truncated at residues 718, 786, 788, 802, 808, 813, 823, and 990 were prepared by PCR. We used the pcDNA3.1 vector containing full-length ErbB-2. The cDNA contains a XhoI restriction site after the stop codon and a SacII restriction site in the transmembrane domain, which was introduced by silent mutagenesis (codon TCT corresponding to residue 656 was replaced with TCC, and natural SacII site downstream in the cytoplasmic region was removed also using silent mutagenesis). The cytoplasmic domain of ErbB-2 was excised using SacII and XhoI restriction sites. Next, cytoplasmic domain segments of ErbB-2 mutants were prepared by PCR and ligated into a vector that contained ErbB-2 ectodomain through SacII and XhoI restriction sites. Segments corresponding to the cytoplasmic domain of particular truncated mutants were amplified by PCR with 5' primer 1 TGC ACC CAC TCC TGT GTG GAC CTG and 3' primers corresponding to the last seven residues of ErbB-2 before each truncation and containing XhoI restriction site after the stop codon.
Parts of the ErbB-2 or EGFR kinase domains fused to GFP were generated by PCR using the following primers: ErbB-2, 5'primer GGG ATC CTC ATC AAG CGA GCT CAG CAG AAG ATC, 3' primers CAC GTA TGC TTC GTC TAG AAT TTC TTT, CCG CAC ATC CTC TAG ATA GCT CAT; EGFR 5' primers GCC AAC AAG GAA ATT CTA GAT GAA GCC, AAG GGC ATG AAC TAT CTA GAG GAC CGT, 3' primer GTA GAA GTT GGA CTC GAG AGG ACT TGG. PCR products were ligated through XbaI sites, reamplified using ErbB-2 5' primer and EGFR 3' primer, and inserted into pEGFP-C1 vector through SacI and XhoI restriction sites.
Site-directed mutagenesis of ErbB-2 residues 813–817 NRGRL to HKDNI (mutant 3), 785–788 LLGI to AAGA (mutant 4), and GVGSPYVS to SVDNPHVC (mutant 5) was done using the PCR megaprimer approach. In the first round of PCR, we used 5'primer 1 (above) and 3'primers ACA CCA GTT CAG CAG GTC CTG GGA GCC AAT GTT GTC TTT GTG TTC CCG GAC ATG GTC TAA GAG GCA GCC (mutant 3), TGT CAG GCA GGC GCC CGC AGC GCG GGA GAC (mutant 4), or CAG AAG GCG GCA CAC GTG TGG GTT GTC CAC ACT AGC CAT CAC (mutant 5). In the second round of PCR, we used the first-round product of PCR as megaprimer and primer ACT ACG TCC AGT CTC GAG TCA CAC TGG CAC GTC CAG ACC. PCR products from the second round were ligated through SacII and XhoI sites into pcDNA3.1 containing the ErbB-2 without cytoplasmic domain, excised by SacII and XhoI.
Immunoblotting.
At the end of each experiment, the cells were solubilized by scraping into cold lysis buffer [10 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% NP40, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1 mM Na3VO4]. The lysates were then clarified by centrifugation (14,000 x g for 10 min), and aliquots containing equal amounts of protein were subjected to SDS-PAGE. Subsequently, proteins were transferred to nitrocellulose membranes, and the membrane was blocked by incubation with 5% BSA in PBS for 1 h at room temperature. The membrane was then incubated for 1 h with the indicated antibody in TBSTw buffer [50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.05% Tween 20, 0.2% nonfat milk], washed three times in the same buffer, and incubated for 1 h with horseradish peroxidase-conjugated antimouse antibody. The membranes were then washed five times with TBSTw buffer and visualized by ECL.
Kinase Assay.
Constructs of wild-type ErbB-2, the M4 or M5 mutants, were transiently transfected into Cos7 cells. After 48 h, cell lysates were prepared, and antibody against the ErbB-2 ectodomain was used to immunoprecipitate ErbB-2 or the mutants. Recombinant GST/PLC-
1 SH2-SH2-SH3 fusion protein was used as a substrate in the in vitro kinase assay. Equal volumes of agarose beads containing equivalent amounts of protein, as confirmed by Western blotting, were incubated for 2 h with 20 µg of substrate in a kinase assay buffer (20 mM HEPES, 100 µM sodium orthovanadate, 1 mM DTT, 10 mM MgCl2, 10 mM MnCl2, and 20 µM ATP). Next, the PLC-1-derived substrate was precipitated with GST antibody and blotted with antibodies against phosphotyrosine or GST.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
show that cleavage(s) within the ErbB-2 kinase domain is observed as early as 15–30 min after the addition of GA, curcumin, or staurosporine. In this experiment, an antibody against the ErbB-2 COOH-terminal residues 1246–1255 detected three to four fragments, of 42–51 kDa, induced by incubation of SKBr3 cells with each of these drugs. The relative molecular mass of these fragments combined with the location of the antibody epitope indicates that cleavage(s) occur close to or within the kinase domain. All three drugs induce formation of the 42, 48, and 51 kDa fragments, although the time course for the appearance of each fragment and the relative amount of each fragment are different. Previously, we have demonstrated that when ectodomain antibodies are used to detect GA-induced ErbB-2 fragments, a fragment of 135 kDa is detected, which includes the ecto- and transmembrane domains plus part of the cytoplasmic domain (27)
. Therefore, the combination of this fragment and the COOH-terminal fragments detected in Fig. 1
are sufficient to represent the result of a GA-induced cleavage of ErbB-2 within its kinase domain.
|
show that an ErbB-2 mutant, truncated at residue 718 (containing the extracellular and transmembrane domains plus 42 residues of the cytoplasmic domain but not the kinase or COOH-terminal domains), has a high level of expression in Cos7 cells and is metabolically stable in the presence of GA. In contrast, a mutant truncated at the end of the kinase domain (i.e., residue 990) and lacking only the COOH-terminal domain is unstable in Cos7 cells because of rapid basal degradation. Degradation of this mutant, however, can be enhanced by the presence of GA. In contrast to the behavior of these mutants, wild-type ErbB-2 expressed in Cos7 cells is constitutively stable, but its level is significantly decreased in the presence of GA.
|
The results in Fig. 2
are consistent with an experimental system we described previously in which the ErbB-2 kinase domain was fused to GFP. Expression of this construct was stable in Cos7 cells, but the addition of GA leads to its rapid degradation (27)
. In addition, we et al. (25
, 27)
have reported that the EGFR is considerably more stable than ErbB-2 in the presence of GA. Therefore, we made GFP fusion proteins containing the ErbB-2 kinase domain, the EGFR kinase domain, or chimeric kinase domains (termed mutants M1 and M2) containing NH2-terminal portions of the ErbB-2 kinase domain fused to COOH-terminal portions of the EGFR kinase domain (Fig. 3A)
. These constructs were expressed in Cos7 cells and tested for their sensitivity to GA. The results show clearly that the ErbB-2 kinase is much more sensitive to GA than the EGFR kinase domain (Fig. 3B)
. These data also show that the kinase domain chimeric construct M1 is sensitive to GA, whereas the M2 chimera is resistant to GA. In the M1 mutant, residues 715–837 of the ErbB-2 kinase domain are followed by residues 806–958 of the EGFR kinase domain. In the M2 mutant, residues 715–767 of the ErbB-2 kinase domain are followed by residues 736–958 of the EGFR kinase domain. Both mutants contain a complete tyrosine kinase domain. These results indicate, therefore, that a motif located between residues 767 and 837 of the ErbB-2 kinase domain confers sensitivity to GA and is consistent with the data shown in Fig. 2
with ErbB-2 kinase domain truncation mutants.
|
When the sequence in ErbB-2 between residues 813 and 817 (NRGRL) was changed to that of the EGFR (HKDNI) and the mutant protein (termed M3) was expressed, it remained as sensitive to GA as the wild-type ErbB2 (Fig. 4)
. However, alteration of the ErbB-2 sequence between residues 776 and 783 (GVGSPYVS) to that present in the corresponding position of the EGFR (SVDNPHVC) had a significant influence on GA-induced ErbB-2 degradation. As shown in Fig. 4
, this mutation (termed M4) prevented GA-induced degradation of ErbB-2.
|
not only indicate that sensitivity to GA in ErbB-2 is mediated by sequence information between residues 718 and 990 of the kinase domain but also show that the basal metabolic stability (i.e., in the absence of GA) influences the same sequence information. Therefore, we constructed a series of ErbB-2 truncation mutants that encompass this region. Beginning at residue 718, successive mutations were made through residue 823 as depicted in Fig. 5
. The basal expression of these mutants was then assayed by transient expression in Cos7 cells. In each case, ALLN was added as indicated to block basal degradation of ErbB-2 mutants.
|
, show when ErbB-2 is truncated within the kinase domain at residues 718, 780, or 783, the mutant proteins are expressed at a high level that is not affected by the presence of ALLN. However, when the truncation occurs at residue 786 or beyond, the mutant proteins are expressed at much lower levels, and their level of expression can be significantly increased by the presence of ALLN, suggesting that the low expression is caused by protein degradation. This result implies that sequence information between residues 784 and 786 increases the sensitivity of these ErbB-2 mutants to metabolic degradation.
To assess the significance of this sequence information, alanine mutagenesis was used to change the native ErbB-2 sequence LLGI to AAGA within the context of the full-length ErbB-2 molecule (mutant M5) and to test the potential role of this sequence in GA-induced degradation. It should be noted that this sequence is adjacent to the sequence (residues 776–783) shown previously in Fig. 4
(i.e., mutant M4) to be necessary for GA-induced degradation of ErbB-2. In addition, this sequence is identified in the EGFR. The results, shown in Fig. 4
, demonstrate that the M5 mutant is not sensitive to GA-induced degradation. Therefore, we have identified by somewhat distinct approaches two adjacent regions in ErbB-2, represented by mutants M4 and M5, that mediate GA-dependent degradation.
It might be argued that because each of these two mutations involve multiple substitutions, they may simply interfere with proper folding of the ErbB-2 kinase domain and thereby alter the receptor’s sensitivity to GA. To assess this possibility, we have tested the tyrosine kinase activity of wild-type ErbB-2 and the M4 or M5 mutants using an in vitro kinase assay. The substrate in this kinase assay was a GST fusion protein that contains the tyrosine phosphorylation region of PLC-1 (Ref. 30
).4
The data in Fig. 6
show that both mutants phosphorylate this substrate and that the M4 mutant actually phosphorylates this substrate significantly more than wild-type ErbB-2. This assay shows that the kinase activity of these mutants is not grossly altered by these mutations. Hence, the folding of the kinase domains do not seem to be significantly altered by the M4 or M5 mutations.
|
, staurosporine and curcumin also stimulate ErbB-2 degradation in a manner that resembles that of GA in some aspects. Therefore, we tested whether the M4 and M5 mutants are resistant to degradation induced by staurosporine or curcumin as well as GA. The results, shown in Fig. 7
, demonstrate that although each of the two ErbB-2 mutants are resistant to GA-induced degradation, each remains sensitive to staurosporine- or curcumin-induced degradation.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
As shown in this manuscript, several ErbB-2 fragments of similar but distinct molecular masses (42–51 kDa) are detected after the addition of GA. It is not clear whether each of these fragments is derived from a separate cleavage event within the ErbB-2 kinase domain or whether the cleavage of ErbB-2 produces one fragment that is then subject to the proteolytic processing. If multiple cleavages at distinct sites within the kinase domain did occur, this would be expected to result in molecular heterogeneity of both the NH2- and COOH-terminal fragments. However, the 135 kDa NH2-terminal fragment does not appear to be heterogeneous. Therefore, it seems more likely that post-ErbB-2 cleavage proteolytic processing of a single COOH-terminal fragment occurs to generate the multiple fragments that we have observed.
Using two different approaches, we have identified two adjacent sequence motifs that are located within the kinase domain of ErbB-2. When mutated, each of these motifs, represented by the mutations M4 and M5, produces a form of ErbB-2 that is not degraded in the presence of GA. Structural analyses of protein kinases indicate that each kinase domain is composed of an NH2- and COOH-terminal lobe. The GA-resistant mutations that we have identified lie within the NH2-terminal lobe whose major function is to facilitate the binding of ATP for subsequent catalysis. Sequence homology analysis has divided kinase domains into 11 distinct subdomains that can be aligned with structural characteristics, such as ß sheets and helices. The M4 and M5 mutants encompass 13 residues that begin near the end of subdomain III and extend through subdomain IV. This places the M4 residues in a connecting strand between the C helix and ß4 strand, whereas the M5 residues are within the ß4 strand. This part of the NH2-terminal lobe is not directly involved in kinase catalytic function. Hence, it seems understandable that the M4 and M5 mutations do not abrogate kinase activity.
Although the M4 mutation changes a unique sequence in ErbB-2 to resemble the corresponding sequence in the EGFR, which is not sensitive to GA, the M5 mutation changes an ErbB-2 sequence that is preserved in the EGFR. Hence, although the sequence at residues 784–786 of ErbB-2 may be necessary for GA-induced degradation, it is clear that these residues are not sufficient to mediate sensitivity to this drug. It seems likely, given their proximity, that the ErbB-2 sequences defined by the M4 and M5 mutations cooperate to facilitate GA-induced degradation of this molecule.
The mechanism by which GA initiates ErbB-2 degradation has been proposed to involve the binding of Hsp90 to ErbB-2 (25) . Hsp90 also binds GA, and in the presence of GA, Hsp90 association with ErbB-2 is disrupted. The mutations to GA resistance reported in this study could represent residues involved in Hsp90 binding to ErbB-2. If these mutations did abrogate Hsp90 binding, we would expect the mutants to become more sensitive to degradation in the absence of associated Hsp90. A second possibility is that the mutations affect protease recognition of ErbB-2 or alter the kinase domain cleavage site. The identity of the protease(s) that cleaves ErbB-2 within the kinase domain in the presence of GA is unknown. However, the kinase domain M4 and M5 mutations prevent GA-induced cleavage but do not block cleavage induced by staurosporine or curcumin, which produces similar sized fragments from ErbB-2. Hence, this model would require that GA-induced degradation of ErbB-2 involves protease distinct from those involved in staurosporine- or curcumin-induced ErbB-2 degradation.
Lastly, it is possible that cleavage of the ErbB-2 molecule produces fragments that may have biological effects in cells. This has been reported for other tyrosine kinases, such as RET, that are subject to cleavage within their cytoplasmic domain by caspases (31) . This possibility is being evaluated.
|
FOOTNOTES |
|---|
1 Supported by the Department of Defense Idea Award Grant DAMD 17-00-1-0483. 
2 To whom requests for reprints should be addressed, at Department of Biochemistry, Vanderbilt University School of Medicine, Nashville, TN 37232-0146. Phone: (615) 322-6678; Fax: (615) 322-2931; E-mail: Graham.Carpenter{at}mcmail.vanderbilt.edu
3 The abbreviations used are: EGF, epidermal growth factor; GA, geldanamycin; ALLN, N-acetyl-L-leucinil-L-leucinil-L-norleucinal; ECL, enhanced chemiluminescence; GST, glutathione S-transferase; EGFR, epidermal growth factor receptor; GFP, green fluorescence protein; PLC, peritoneal lymphocyte.
4 D. Tvorogov and G. Carpenter, unpublished data.
Received 7/18/02. Accepted 10/28/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 with the activated EGF receptor. J. Biol. Chem., 274: 26091-26097, 1999.This article has been cited by other articles:
|
|
 |
|
  N. M. Pedersen, I. H. Madshus, C. Haslekas, and E. Stang Geldanamycin-Induced Down-Regulation of ErbB2 from the Plasma Membrane Is Clathrin Dependent but Proteasomal Activity Independent Mol. Cancer Res., March 1, 2008; 6(3): 491 - 500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-X. Fan, L. Wong, J. Ding, N. A. Spiridonov, R. C. Johnson, and G. R. Johnson Mutational Activation of ErbB2 Reveals a New Protein Kinase Autoinhibition Mechanism J. Biol. Chem., January 18, 2008; 283(3): 1588 - 1596. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lerdrup, S. Bruun, M. V. Grandal, K. Roepstorff, M. M. Kristensen, A. M. Hommelgaard, and B. van Deurs Endocytic Down-Regulation of ErbB2 Is Stimulated by Cleavage of Its C-Terminus Mol. Biol. Cell, September 1, 2007; 18(9): 3656 - 3666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. Mandal, P. Lee, J. A. Chen, N. Nillegoda, A. Heller, S. DiStasio, H. Oen, J. Victor, D. M. Nair, J. L. Brodsky, et al. Cdc37 has distinct roles in protein kinase quality control that protect nascent chains from degradation and promote posttranslational maturation J. Cell Biol., January 29, 2007; 176(3): 319 - 328. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Citri, D. Harari, G. Shohat, P. Ramakrishnan, J. Gan, S. Lavi, M. Eisenstein, A. Kimchi, D. Wallach, S. Pietrokovski, et al. Hsp90 Recognizes a Common Surface on Client Kinases J. Biol. Chem., May 19, 2006; 281(20): 14361 - 14369. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lerdrup, A. M. Hommelgaard, M. Grandal, and B. van Deurs Geldanamycin stimulates internalization of ErbB2 in a proteasome-dependent way J. Cell Sci., January 1, 2006; 119(1): 85 - 95. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Kuwada, J. Kuang, and X. Li Integrin {alpha}5/{beta}1 Expression Mediates HER-2 Down-regulation in Colon Cancer Cells J. Biol. Chem., May 13, 2005; 280(19): 19027 - 19035. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Tsvetkov, G. Asher, V. Reiss, Y. Shaul, L. Sachs, and J. Lotem Inhibition of NAD(P)H:quinone oxidoreductase 1 activity and induction of p53 degradation by the natural phenolic compound curcumin PNAS, April 12, 2005; 102(15): 5535 - 5540. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kansra, S. W. Stoll, J. L. Johnson, and J. T. Elder Autocrine Extracellular Signal-regulated Kinase (ERK) Activation in Normal Human Keratinocytes: Metalloproteinase-mediated Release of Amphiregulin Triggers Signaling from ErbB1 to ERK Mol. Biol. Cell, September 1, 2004; 15(9): 4299 - 4309. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Giannini and M.-J. Bijlmakers Regulation of the Src Family Kinase Lck by Hsp90 and Ubiquitination Mol. Cell. Biol., July 1, 2004; 24(13): 5667 - 5676. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Ishizawar, D. A. Tice, T. Karaoli, and S. J. Parsons The C Terminus of c-Src Inhibits Breast Tumor Cell Growth by a Kinase-independent Mechanism J. Biol. Chem., May 28, 2004; 279(22): 23773 - 23781. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
