Geldanamycin

Geldanamycin-induced degradation of Chk1 is mediated by proteasome

Abstract

Checkpoint kinase 1 (Chk1) is a cell cycle regulator and a heat shock protein 90 (Hsp90) client. It is essential for cell proliferation and survival. In this report, we analyzed the mechanisms of Chk1 regulation in U87MG glioblastoma cells using Geldanamycin (GA), which interferes with the function of Hsp90. GA reduced Chk1 protein level but not its mRNA level in glioblastoma cells. Co-treatment with GA and cycloheximide (CHX), a protein synthesis inhibitor, induced a decrease of half-life of the Chk1 protein to 3 h and resulted in Chk1 down-regulation. CHX alone induced only 32% reduction of Chk1 protein even after 24 h. These find- ings indicated that reduction of Chk1 by GA was due to destabilization and degradation of the protein. In addition, GA-induced down-regulation of Chk1 was reversed by MG132, a specific proteasome inhibitor. And it was revealed that Chk1 was ubiquitinated by GA. These results have indicated that degradation of Chk1 by GA was mediated by the ubiquitin–proteasome pathway in U87MG glioblastoma cells.

Keywords: Checkpoint kinase 1; Glioblastoma; Geldanamycin; Ubiquitin; Proteasome

To maintain genomic integrity, cells have evolved mechanisms known as checkpoints that detect genomic abnormality and initiate a response to cope with it through a complex network of signal transduction path- ways [1]. Checkpoint kinase 1 (Chk1) was first identified in fission yeast and functions as a DNA damage check- point causing cell cycle arrest upon DNA damage [2,3]. Chk1 is expressed in a cell cycle-dependent manner at both mRNA and protein levels. Chk1 shows almost no expression at G0 to G1 and high expression at S to M phases in human cells [4]. In mammalian cells, Chk1 participates in G2/M DNA damage checkpoints by phosphorylating and modulating the activity of Cdc25C phosphatase and Wee1 [5]. It was reported that knockout of Chk1 gene in mice was embryonically lethal and also induced apoptosis in embryonic stem cells [6,7]. Thus, Chk1 is an evolutionarily conserved protein ki- nase that functions to ensure genomic integrity upon genotoxic stress [1].

Heat shock protein 90 (Hsp90) is an abundant and ubiquitously expressed chaperone protein that accounts for 1–2% of total cellular protein in unstressed mamma- lian cells [8]. It is highly conserved and thought to be essential for eukaryotic cell viability [9]. A number of proteins require interaction with Hsp90 and its chaper- one to acquire proper protein function. Such proteins in- clude cyclin-dependent kinases [10] and protein kinases [11,12]. Recently, Arlander et al. [13] reported that Chk1 is an Hsp90 client protein.Geldanamycin (GA) has recently become an impor- tant tool to specifically interfere with the function of Hsp90 [14,15]. GA-mediated Hsp90 dissociation from client proteins results in their ubiquitination and subse- quent degradation by proteasome [16–18].To date, little is known about the mechanisms for degradation of Chk1. In this report, we analyzed the mechanisms of Chk1 regulation in glioblastoma cells using GA.

Materials and methods

Reagents. GA, MG132, and cycloheximide (CHX) were purchased from Sigma Chemical (St. Louis, MO). GA was prepared as a 1 mM stock in dimethyl sulfoxide (DMSO). MG132 was prepared as a 10 mM stock in DMSO and stored at —80 °C until use. CHX was dissolved in ethanol as a stock solution of 100 mg/ml concentration.

Cell lines. The U87MG glioblastoma cell line was used for all experiments. Various other tumor cell lines, murine glioma (GL261), malignant meningioma (HKBMM), cervical cancer (HeLa), epider- moid carcinoma (A431), and embryonal carcinoma (NEC14), were also used. The cells were maintained in appropriate culture medium supplemented with fetal bovine serum (Gibco, Grand Island, NY) and antibiotics in a humidified atmosphere containing 5% CO2 and 95% air at 37 °C. Cells were split every 3–4 days to ensure logarithmic growth according to the manufacturer’s instruction.

Treatment of the cells with proteasome inhibitor. U87MG glioblas- toma cells were initially incubated with 2 lM MG132 or vehicle for 1 h, and then GA or DMSO was added to the culture medium. The cells were treated with GA or DMSO for 24 h in the presence or ab- sence of MG132.
Western blot analysis. Total cells were harvested from each culture condition at the appropriate time interval, washed with ice-cold PBS, and the protein was extracted using a lysis buffer containing 1% Triton X-100, 150 mM NaCl, 5 mM EDTA, 50 mM sodium fluoride, 1 mM sodium orthovanadate, 5 lg/ml phenylmethylsulfonyl fluoride, 2 lg/ml aprotinin, 5 lg/ml leupeptin, and 2 lg/ml pepstatin. The samples were centrifuged at 14000 rpm for 30 min at 4 °C. Extracts were stored at —80 °C until use. Protein concentration was determined using the BCA assay (Pierce, Rockford, IL).

For Western blot analysis, equal amounts of protein (30 lg) were electrophoresed on 8% (Hsp90, Hsp70) or 10% (Chk1) SDS–PAGE gels, transferred to nitrocellulose membrane (Trans-Blot Transfer Medium 0.45 micron, Bio-Rad, Hercules, CA) by electroblotting at 4 °C for 4 h at 60 V, and stained with Ponceau S (Sigma). After con- firmation of protein transfer, specific proteins were detected with the following antibodies: mouse anti-human Hsp90 monoclonal antibody used at 1:2000 (SPA-830, StressGen Biotechnologies, San Diego, CA); rabbit anti-human Hsp70 polyclonal antibody used at 1:2000 (SPA- 810, StressGen Biotechnologies); mouse anti-human Chk1 monoclonal antibody used at 1:1000 (G-4, SantaCruz, CA); mouse anti-human b- actin monoclonal antibody used at 1:10000 (Chemicon, Temecula, CA). Sheep anti-mouse IgG and donkey anti-rabbit IgG horseradish peroxidase-linked secondary antibodies (Amersham Biosciences, Pis- cataway, NJ) at 1:4000 dilution were used for 1 h at room temperature. Protein detection was performed using a SuperSignal West Femto Maximum Sensitivity Substrate (Pierce) and visualized using Hyper- film ECL (Amersham Biosciences).

Inhibition of protein synthesis and half-life determination. Analysis of the half-life of a Chk1 protein was carried out as described previ- ously [19,20]. U87MG glioblastoma cells were preincubated with 500 nM GA or DMSO for 1 h. Then, cells were treated with 100 lg/ml CHX or a combination of CHX and GA, and harvested at different time points. Whole cell lysates were subjected to Western blot analysis [21,22]. The nitrocellulose membrane was successively probed with mouse anti-human Chk1 monoclonal antibody at 1:1000 dilution, and mouse anti-human b-actin monoclonal antibody at 1:10,000 dilution. The amount of immunoblotted protein was determined by scanning of autoradiograph and analyzed using the NIH Image software. The amount of Chk1 protein relative to b-actin was plotted against the time course of culture.

Immunoprecipitation. Immunoprecipitation was performed as de- scribed previously [17,23]. First, protein A–Sepharose (Pharmacia Biotech, Uppsala, Sweden) was preincubated with anti-Chk1 antibody (rabbit anti-Chk1 polyclonal antibody, FL-476, Santa Cruz). Then the cell lysate (750 lg) was incubated with protein A–Sepharose at 4 °C with constant rotation. The beads were washed seven times with lysis buffer. Then, the beads were boiled at 95 °C for 5 min. The sample buffer containing immunoprecipitated protein was electrophoresed on a 8% SDS–PAGE gel and the protein was transferred to nitrocellulose membrane by electroblotting. After confirmation of protein transfer, specific proteins were detected with the following antibodies: mouse anti-ubiquitin antibody used at 1:1000 (13-1600, Zymed, South San Francisco, CA). Then the membrane was stripped and reprobed with mouse anti-Chk1 monoclonal antibody used at 1:1000. RT-PCR. To analyze the expression of mRNA, genes were amplified by a semiquantitative RT-PCR technique. U87MG glio- blastoma cells were treated with 500 nM GA and total RNA was extracted at various time points. RT-PCR was performed using Ready- To-Go RT-PCR Beads (Amersham Biosciences). The primer sequences were as follows: forward, 50-CTTTGGCTTGGCAACAG T-30; and reverse, 50-CCAGTCAGAATACTCCTG-30, for Chk1 (expected length of product: 220 bp) [24]; forward, 50-ATCACCATTGGCAATGAGCG-30; and reverse, 50-TTGAAGGTAGTTTCGTGGAT-30, for b-actin (expected length of product: 98 bp) [25]. The cDNA templates for RT-PCR were synthesized from 1 lg of total RNA for Chk1 or 500 ng for b-actin. PCR amplification was performed for 35 cycles for Chk1 or 30 cycles for b-actin. Five microliters of each RT-PCR product was electrophoresed on a 2.0% agarose gel.

Results

GA reduced Chk1 protein level (Fig. 1)

U87MG glioblastoma cells were treated with 500 nM GA for various time intervals. As shown in Fig. 1, Chk1 protein showed a slight decrease at 4 h and declined sig- nificantly after 12 h treatment by GA. GA promoted down-regulation of Chk1 in a time-dependent manner, with the protein diminishing at 24 h. Exposure of U87MG glioblastoma cells to GA up-regulated Hsp90 protein expression. An increase of Hsp70 protein was also noted in U87MG glioblastoma cells by GA. On the other hand, expressions of Chk1, Hsp70, and Hsp90 were not changed by DMSO alone.

GA decreased half-life of Chk1 protein (Fig. 2)

To determine whether GA down-regulates Chk1 at the protein or mRNA level, we pretreated U87MG glio- blastoma cells with GA, and protein synthesis was sub- sequently inhibited with CHX. At the times indicated in Fig. 2, cellular Chk1 protein level was determined by Western blot analysis. The level of Chk1 protein in cells exposed to both GA and CHX was compared with that in cells treated with CHX alone. As shown in Fig. 2, pre- existing Chk1 protein was markedly decreased by GA and CHX, indicating a protein half-life of 3 h. On the other hand, when cells were treated with CHX alone, Chk1 protein decreased to only 67.6% of the initial amount after 24 h. These data demonstrated that down-regulation of Chk1 protein occurred at the protein level.

Chk1 degradation by GA was mediated by proteasome (Fig. 4)

Previous research of other proteins has suggested a linkage between GA and degradation of target proteins via the proteasome pathway [26,27]. We examined whether proteasomal degradation mediates the loss of Chk1 in GA-treated U87MG glioblastoma cells. As shown in Fig. 4, presence of MG132 completely recruited GA-mediated decline of Chk1 (lanes 5 and 6). The combination of proteasome inhibitor and GA (lanes 5 and 6) stabilized the Chk1 protein level, but to a degree less than that observed with the proteasome inhibitor alone (lane 4). These data demonstrated that Chk1 was degraded by proteasome.

Fig. 2. GA destabilizes Chk1 protein. The figures (A, Western blot analysis; B, densitometric analysis) show that Chk1 protein level in U87MG glioblastoma cells treated with GA and CHX declines rapidly. On the other hand, Chk1 expression shows only a slight decrease with CHX alone.

GA induced Chk1 ubiquitination (Fig. 5)

We tested whether Chk1 was ubiquitinated prior to its degradation in GA-treated cells. We immunoprecipi- tated Chk1 from whole cell lysates. The immunoprecip- itates were electrophoresed and transferred to a membrane. After probing with anti-ubiquitin antibody (upper panel), the membrane was stripped and re-probed with anti-Chk1 antibody (lower panel). As shown in Fig. 5, there was virtually no detectable ubiq- uitinated Chk1 in control and GA-treated cells (lanes 1 and 2), and only a slight increase in ubiquitinated spe- cies was noted after treatment with MG132 alone (lane 3). However, a significant increase of ubiquitinated damage and replication stress [13]. Activated Chk1 per- forms several functions that promote cell survival, such as cell cycle arrest. Disruption of Chk1 signaling may be an effective method to sensitize tumor cells to chemo- therapeutic agents, DNA damage agents or stimuli such as ultraviolet light. Therefore, to know the mechanisms in which Chk1 protein is destabilized or degraded is essential for developing a new treatment for tumors.

Fig. 4. GA-induced degradation of Chk1 proceeds via the proteasome pathway. The figures (A, Western blot analysis; B, densitometric analysis) show significant prevention of the degradation of Chk1 when the cells were simultaneously treated with both GA and MG132. Combined treatment of GA and MG132 resulted in the accumulation of Chk1 protein to a degree almost same as that in control.

GA is an Hsp90 binding agent which has been shown previously to stimulate the proteasome-mediated degra- dation of several cellular signaling proteins including Akt [28], Plk [29], and Wee1 [30]. In the present study, we analyzed an effect of GA on U87MG glioblastoma cells focusing on destabilization mechanisms of Chk1. We showed that GA reduced Chk1 protein in U87MG glioblastoma cells with no effect on mRNA. This inhib- itory effect of GA was observed in a wide spectrum of tumor cell lines. These results indicated that the inhibi- tory effect of GA was due to degradation of Chk1 pro- tein but not decreased synthesis.
Then, we examined whether Chk1 protein degrada- tion by GA is mediated by proteasome. We used MG132, a specific proteasome inhibitor, to analyze the pathway responsible for turnover of Chk1. MG132 alone slightly increased the amount of Chk1 protein. Co-treatment of the cells with GA and MG132 completely recruited the Chk1 down-regulation by GA. These data demonstrated that GA-induced degradation of Chk1 protein was mediated by proteasome.

Fig. 5. Effect of GA on Chk1 ubiquitination. The figure shows that there is no detectable ubiquitinated Chk1 in control and GA-treated cells (lanes 1 and 2). Faint bands of ubiquitinated species are observed after treatment with MG132 alone (lane 3). A significant increase of ubiquitinated Chk1 is shown in cells treated with both GA and MG132 (lane 4). Arrow in the upper panel indicates the molecular weight of 56 kDa where non-ubiquitinated Chk1 should be electrophoresed.

Proteasomal degradation of proteins is dependent on the covalent attachment of multiple ubiquitin molecules to the target [31]. Although some reports have suggested that the proteasome could degrade selectively non-ubiq- uitinated proteins efficiently, most proteasomal degrada- tion appeared to be ubiquitin-dependent [32]. Previous reports have demonstrated that GA-mediated proteaso- mal degradation of Hsp90 client proteins was preceded by their ubiquitination [16,27,33]. Therefore, to confirm that GA-stimulated degradation of Chk1 was mediated by this system, we immunoprecipitated Chk1 protein fol- lowed by probing with anti-ubiquitin antibody. Ubiqui- tinated Chk1 from anti-Chk1 immunoprecipitates could be visualized after co-treatment with GA and MG132. This finding demonstrated that GA promoted efficient ubiquitination of Chk1, and ubiquitination of Chk1 might be followed by proteasome-dependent deg- radation in U87MG glioblastoma cells. And Chk1 might share common characteristics of degradation mecha- nisms with those of the other Hsp90 client proteins.

Fig. 6. GA stimulates Chk1 protein degradation in various tumor cell lines. The indicated cells were treated with either GA or vehicle for 24 h and then were harvested for Western blot analysis. The figure shows that GA decreases Chk1 expression in the various tumor cell lines.

Taken together, GA treatment of U87MG glioblasto- ma cells might result in dissociation of Hsp90 from Chk1. Then the dissociated Chk1 might be ubiquitina- ted followed by proteasomal degradation. To our knowledge, this is the first report that demonstrated the mechanisms of degeneration of Chk1 by GA. Our data revealed that ubiquitin-conjugated Chk1 was a sub- strate for proteasome and that GA treatment accelerat- ed the degradation of Chk1 by the proteasomal pathway. Therefore, selective destabilization of Chk1 by GA may be an attractive therapy for malignant tumors.

In conclusion, GA inhibited Chk1 expression in U87MG glioblastoma cells and various tumor cell lines. This down-regulation of Chk1 by GA was mediated via the ubiquitin–proteasome pathway. GA leads to ubiqui- tination of Chk1, followed by its proteasomal degrada- tion. A combination of a cell cycle-specific drug and a cell cycle abrogator such as GA may be an effective treatment for malignant tumors.

References

[1] Y. Chen, Y. Sanchez, Chk1 in the DNA damage response: conserved roles from yeasts to mammals, DNA Repair 3 (2004) 1025–1032.
[2] F. Al-Khodairy, E. Fotou, K.S. Sheldrick, D.J.F. Griffiths, A.R. Lehmann, A.M. Carr, Identification and characterization of new elements involved in checkpoint and feedback controls in fission yeast, Mol. Biol. Cell 5 (1994) 147–160.
[3] N. Walworth, S. Davey, D. Beach, Fission yeast chk1 protein kinase links the rad checkpoint pathway to cdc2, Nature 363 (1993) 368–371.
[4] Y. Kaneko, N. Watanabe, H. Morisaki, H. Akita, A. Fujimoto, K. Tominaga, M. Terasawa, A. Tachibana, K. Ikeda, M. Nakanishi, Cell cycle-dependent and ATM-indepen- dent expression of human Chk1 kinase, Oncogene 18 (1999) 3673–3681.
[5] N. Rhind, P. Russell, Chk1 and Cds1: linchpins of the DNA damage and replication checkpoint pathway, J. Cell Sci. 113 (2000) 3889–3896.
[6] H. Takai, K. Tominaga, N. Motoyama, Y.A. Minamishima, H. Nagahama, T. Tsukiyama, K. Ikeda, K. Nakayama, M. Nakani- shi, K. Nakayama, Aberrant cell cycle checkpoint function and early embryonic death in Chk1—/— mice, Genes Dev. 14 (1993) 1439–1447.
[7] Q. Liu, S. Guntuku, X.S. Cui, S. Matsuoka, D. Cortez, K. Tamai,
G. Luo, S. Carattini-Rivera, F. DeMayo, A. Bradley, L.A. Donehower, S.J. Elledge, Chk1 is an essential kinase that is regulated by Atr and required for the G2/M DNA damage checkpoint, Genes Dev. 14 (2000) 1448–1459.
[8] T. Scheibel, J. Buchner, The Hsp90 complex: a super-chaperone machine as a novel drug target, Biochem. Pharmacol. 56 (1998) 675–682.
[9] K.A. Borkovich, F.W. Farrelly, D.B. Finkelstein, J. Taulien, S. Lindquist, hsp82 is an essential protein that is required in higher concentrations for growth of cells at higher temperatures, Mol. Cell. Biol. 9 (1989) 3919–3930.
[10] L. Stepanova, X. Leng, S.B. Parker, J.W. Harper, Mamma- lian p50Cdc37 is a protein kinase-targeting subunit of Hsp90 that binds and stabilizes Cdk4, Genes Dev. 10 (1996) 1491–1502.
[11] J.S. Brugge, E. Erikson, R.L. Erikson, The specific interaction of the Rous sarcoma virus transforming protein, pp60src, with two cellular proteins, Cell 25 (1981) 363–372.
[12] L.F. Stancato, Y.H. Chow, K.A. Hutchison, G.H. Perdew, R. Jove, W.B. Pratt, Raf exists in a native heterocomplex with Hsp90 and p50 that can be reconstituted in a cell-free system, J. Biol. Chem. 268 (1993) 21711–21716.
[13] S.J.H. Arlander, A.K. Eapen, B.T. Vroman, R.J. McDonald,
D.O. Toft, L.M. Karnitz, Hsp90 inhibition depletes Chk1 and sensitizes tumor cells to replication stress, J. Biol. Chem. 278 (2003) 52572–52577.
[14] L. Whitesell, E.G. Mimnaugh, B. De Costa, C.E. Myers, L.M. Neckers, Inhibition of heat shock protein HSP90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: essential role for stress proteins in oncogenic transformation, Proc. Natl. Acad. Sci. USA 91 (1994) 8324–8328.
[15] D.F. Smith, L. Whitesell, S.C. Nair, S. Chen, V. Prapapanich,
R.A. Rimerman, Progesterone receptor structure and function altered by geldanamycin, an hsp90-binding agent, Mol. Cell. Biol. 15 (1995) 6804–6812.
[16] E.G. Mimnaugh, C. Chavany, L. Neckers, Polyubiquitination and proteasomal degradation of the p185c-erbB2 receptor protein- tyrosine kinase induced by geldanamycin, J. Biol. Chem. 271 (1996) 22796–22801.
[17] T.W. Schulte, W.G. An, L.M. Neckers, Geldanamycin-induced destabilization of Raf-1 involves the proteasome, Biochem. Biophys. Res. Commun. 239 (1997) 655–659.
[18] L. Whitesell, P. Sutphin, W.G. An, T. Schulte, M.V. Blag- osklonny, L. Neckers, Geldanamycin-stimulated destabilization of mutated p53 is mediated by the proteasome in vivo, Oncogene 14 (1997) 2809–2816.
[19] P. Riou, M. Vandromme, L. Gazzolo, Human T-cell leukemia virus type 1 tax protein inhibits the expression of the basic helix- loop-helix transcription factor MyoD in muscle cells: maintenance of proliferation and repression of differentiation, Cell Growth Differ. 12 (2001) 613–622.
[20] S.X. Zeng, Y. Jin, D.T. Kuninger, P. Rotwein, H. Lu, The acetylase activity of p300 is dispensable for MDM2 stabilization, J. Biol. Chem. 278 (2003) 7453–7458.
[21] D. Zagzag, M. Nomura, D.R. Friedlander, C.Y. Blanco, J.P. Gagner, N. Nomura, E.W. Newcomb, Geldanamycin inhibits migration of glioma cells in vitro: a potential role for hypoxia- inducible factor (HIF-1alpha) in glioma cell invasion, J. Cell. Physiol. 196 (2003) 394–402.
[22] M. Nomura, N. Nomura, E.W. Newcomb, Y. Lukyanov, C. Tamasdan, D. Zagzag, Geldanamycin induces mitotic catastrophe and subsequent apoptosis in human glioma cells, J. Cell. Physiol. 201 (2004) 374–384.
[23] J.S. Isaacs, Y.J. Jung, E.G. Mimnaugh, A. Martinez, F. Cuttitta,
L.M. Neckers, Hsp90 regulates a von Hipple Lindu-independent hypoxia-inducible factor-1a-degradative pathway, J. Biol. Chem. 277 (2002) 29936–29944.
[24] N. Haruki, H. Saito, Y. Tatematsu, H. Konishi, T. Harano, A. Masuda, H. Osada, Y. Fujii, T. Takahashi, Histological type- selective, tumor-predominant expression of a novel CHK1 isoform and infrequent in vivo somatic CHK2 mutation in small cell lung cancer, Cancer Res. 60 (2000) 4689–4692.
[25] M. Nomura, S. Yamagishi, S. Harada, Y. Hayashi, T. Yamashi- ma, J. Yamashita, H. Yamamoto, Possible participation of autocrine and paracrine vascular endothelial growth factors in hypoxia-induced proliferation of endothelial cells and pericytes, J. Biol. Chem. 270 (1995) 28316–28324.
[26] B. Segniz, U. Gehring, The function of steroid hormone receptors is inhibited by the hsp90-specific compound geldanamycin, J. Biol. Chem. 272 (1997) 18694–18701.
[27] M.A. Loo, T.J. Jensen, L. Cui, Y. Hou, X.B. Chang, J.R. Riordan, Perturbation of Hsp90 interaction with nascent CFTR prevents its maturation and accelerates its degradation by the proteasome, EMBO J. 17 (1998) 6879–6887.
[28] J.Y. Le Brazidec, A. Kamal, D. Busch, L. Thao, L. Zhang, G. Timony, R. Grecko, K. Trent, R. Lough, T. Salazar, S. Khan, F. Burrows, M.F. Boehm, Synthesis and biological evaluation of a new class of geldanamycin derivatives as potent inhibitors of Hsp90, J. Med. Chem. 47 (2004) 3865–3873.
[29] S. Simizu, H. Osada, Mutations in the Plk gene lead to instability of Plk protein in human tumour cell lines, Nat. Cell Biol. 2 (2000) 852–854.
[30] F.S. Goes, J. Martin, Hsp90 chaperone complexes are required for the activity and stability of yeast protein kinases Mik1, Wee1 and Swe1, Eur. J. Biochem. 268 (2001) 2281–2289.
[31] D. Kornitzer, A. Ciechanover, Modes of regulation of ubiquitin- mediated protein degradation, J. Cell. Physiol. 182 (2000) 1–11.
[32] I. Jariel-Encontre, M. Pariat, F. Martin, S. Carillo, C. Salvat, M. Piechaczyk, Ubiquitinylation is not an absolute requirement for degradation of c-Jun protein by the 26S proteasome, J. Biol. Chem. 270 (1995) 11623–11627.
[33] L. Sepp-Lorenzino, Z. Ma, D.E. Lebwohl, A. Vinitsky, N. Rosen, Herbimycin A induces the 20 S proteasome-and ubiquitin- dependent degradation of receptor tyrosine kinases, J. Biol. Chem. 270 (1995) 16580–16587.