An efficient system to detect protein ubiquitination by ... - CiteSeerX

The Plant Journal (2010) 61, 893–903

doi: 10.1111/j.1365-313X.2009.04109.x

TECHNICAL ADVANCE

An efficient system to detect protein ubiquitination by agroinfiltration in Nicotiana benthamiana Lijing Liu1,†, Yiyue Zhang1,†, Sanyuan Tang1, Qingzhen Zhao1,2, Zhonghui Zhang1, Huawei Zhang1, Li Dong3, Huishan Guo3 and Qi Xie1,* 1 State Key Laboratory of Plant Genomics, National Center for Plant Gene Research, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Datun Road, Beijing 100101, China, 2 School of Life Science, Liaocheng University, Liaocheng 252059, China, and 3 State Key Laboratory of Plant Genomics, National Center for Plant Gene Research, Institute of Microbiology, Chinese Academy of Sciences, Datun Road, Beijing 100101, China Received 6 October 2009; revised 2 December 2009; accepted 8 December 2009; published online 18 January 2010. * For correspondence (fax +86 10 64889351; e-mail [email protected]). † These authors contributed equally to this work.

SUMMARY The ubiquitination proteasome pathway has been demonstrated to regulate all plant developmental and signaling processes. E3 ligase/substrate-specific interactions and ubiquitination play important roles in this pathway. However, due to technical limitations only a few instances of E3 ligase–substrate binding and protein ubiquitination in plants have been directly evidenced. An efficient in vivo and in vitro ubiquitination assay was developed for analysis of protein ubiquitination reactions by agroinfiltration expression of both substrates and E3 ligases in Nicotiana benthamiana. Using a detailed analysis of the well-known E3 ligase COP1 and its substrate HY5, we demonstrated that this assay allows for fast and reliable detection of the specific interaction between the substrate and the E3 ligase, as well as the effects of MG132 and substrate ubiquitination and degradation. We were able to differentiate between the original and ubiquitinated forms of the substrate in vivo with antibodies to ubiquitin or to the target protein. We also demonstrated that the substrate and E3 ligase proteins expressed by agroinfiltration can be applied to analyze ubiquitination in in vivo or in vitro reactions. In addition, we optimized the conditions for different types of substrate and E3 ligase expression by supplementation with the gene-silencing suppressor p19 and by time-courses of sample collection. Finally, by testing different protein extraction buffers, we found that different types of buffer should be used for different ubiquitination analyses. This method should be adaptable to other protein modification studies. Keywords: ubiquitination assay, agroinfiltration, E3 ligase, substrate, in vivo.

INTRODUCTION The ubiquitin/proteasome system (UPS) is a key protein regulation mechanism in eukaryotic cells. It controls numerous cellular processes, including embryogenesis, cell cycle progression, transcriptional regulation, circadian rhythms and senescence. Defects in protein ubiquitination have been associated with large numbers of human diseases, e.g. cystic fibrosis, renal cell carcinoma and polyglutamine diseases (Corn, 2007; Turnbull et al., 2007; Dikshit and Jana, 2008). In plants, besides the general cell activities mentioned above, the UPS also affects root growth, floral ª 2010 The Authors Journal compilation ª 2010 Blackwell Publishing Ltd

homeosis, photomorphogenesis, pathogen defense and hormone regulation (Samach et al., 1999; Hellmann and Estelle, 2002; Xie et al., 2002; Devoto et al., 2003; Dreher and Callis, 2007; Stone and Callis, 2007). Thus, it is not surprising to find that about 6% of the Arabidopsis genome encodes proteins predicted to have functions related to the UPS (Downes and Vierstra, 2005). Protein degradation by the UPS pathway involves a multi-protein cascade. First, ubiquitin is activated by the ubiquitin-activating enzyme (E1) in an ATP-dependent manner. Second, the activated ubiquitin is 893

894 Lijing Liu et al. transferred to a ubiquitin-conjugating enzyme (E2). Third, with the help of a ubiquitin protein ligase (E3), ubiquitin is ligated to a specific protein target. Finally, substrates conjugated with polyubiquitin chains (of more than four ubiquitins) are targeted to the 26S proteasome for degradation. In the Arabidopsis genome, there are two E1s, 37 E2s and more than 1300 E3s. Such a large number of E3s provides specificity for recognizing target proteins (Downes and Vierstra, 2005). In mammals, the interaction and specificity of ubiquitination between E3s and substrates can be easily detected by transient expression systems in cultured cell lines, and this assay has been widely applied (Fang et al., 2000; Nie et al., 2002; Chen et al., 2008). However, in plants such a convenient system has not yet been reported. Generally, plant scientists prove a protein is the specific substrate of an E3 through evidence from three types of experiments: (i) an in vivo or in vitro pull-down assay, (ii) demonstration of E3 ubiquitination of the protein in vitro, and (iii) assessment of a change in stability of the protein in E3 overexpression or mutant lines (Xie et al., 2002; Zhang et al., 2005; Stone et al., 2006). Each of these is indispensable, but the drawbacks of these analyses are also obvious. On one hand, the in vitro ubiquitination reaction is hard to detect and possesses intrinsic limitations, including systematic ‘false-positive’ and ‘false-negative’ ubiquitination. In many cases, proteins expressed in bacteria fail in the in vitro ubiquitination reaction due to the lack of a protein modification that may be necessary for the normal activity of either the E3 ligase or the substrate. On the other hand, generation of stable transgenic plants is time-consuming, usually requiring several months. Thus, transient expression of proteins in plants has been selected as an alternative. Plant cell walls sometimes make transfection difficult, so protoplasts have been used (Abel and Theologis, 1994; Lee et al., 2009). However, this method limits the level of protein expression and the scale of the experiment. Agroinfiltration, the direct application of an agrobacterial solution to the leaves of several species of plants, has proven to be an efficient way to analyze RNA, small RNA, protein localization and antibody production (Goodin et al., 2002; Koscianska et al., 2005; Rodriguez et al., 2005). Transient expression has been described in several different species of plants. Among them, Nicotiana benthamiana is commonly utilized (Mokrzycki-Issartel et al., 2003; Wroblewski et al., 2005; Chakrabarty et al., 2007). There are several advantages to using N. benthamiana for transient protein expression. First, N. benthamiana is a commonly used model plant, and it is easily managed in all plant research laboratories. Second, proteins can be easily expressed at high levels in N. benthamiana through leaf agroinfiltration, and genesilencing suppressors, such as p19 and P1/HC-Pro, have been demonstrated to greatly enhance protein expression in the agroinfiltration system (Kapila et al., 1997; Johansen and Carrington, 2001; Voinnet et al., 2003; Ma et al., 2008).

Finally, the process is very simple and time-saving; no complicated equipment is necessary, and the full process requires no more than a week. We therefore chose transient protein expression in N. benthamiana as an experimental model to demonstrate the interaction, ubiquitination and in vivo degradation of E3 and target protein pairs. With this newly established method, we have provided powerful biochemical evidence for some E3/substrate pairs, including characterization of the function of the positive regulator RING finger E3 ligase SDIR1 in the ABA signaling pathway (Zhang et al., 2007); the specificity of the RHF E3 ligase towards different members of the ICK/KRP family and its role in their degradation (Liu et al., 2008); and the role of ELF3 as a substrate adaptor, which enables COP1 to modulate light-input signals to the circadian clock through targeted destabilization of the clock-associated protein GI (Hicks et al., 1996; McWatters et al., 2000; Yu et al., 2008). To provide a complete and feasible method for protein modification researchers to analyze ubiquitination in vivo, in this paper we selected a well-known ubiquitin regulation complex, the Arabidopsis E3 ligase COP1 and its substrate, the transcription factor HY5 (Deng et al., 1992; Ang et al., 1998; Ma et al., 2002; Saijo et al., 2003), as an example for studying factors in agroinfiltration protein expression, extraction and analysis. In some experiments, we extended the work to the COP1/ELF3 complex (Yu et al., 2008). We demonstrate that the physical interaction of COP1 and HY5 can easily be detected and that HY5 is significantly degraded in the presence of COP1 in this system. The in vitro protein ubiquitination assay can also be performed with proteins that are transiently expressed in N. benthamiana, especially when protein modifications are necessary for protein ubiquitination to take place. Therefore, the transient-expression ubiquitination analysis system that we have developed provides a viable, alternative experimental approach for studying ubiquitination in vivo. RESULTS Detecting the interaction of substrate and E3 ligase transiently expressed via agroinfiltration In vivo data are essential to demonstrate whether a substrate protein can be ubiquitinated and degraded in planta, as artifacts can be produced in some in vitro experiments. Getting a detectable level of stable protein is the first step for all biochemical assays. However, the currently available approaches fail to detect most proteins in vivo due to the small amount of these proteins produced in plants. To resolve this problem, we attempted to develop a method similar to the approach in the animal research field of expressing candidate proteins in cultured mammalian cells by transfection. We developed a highly efficient and reliable system for detecting protein ubiquitination in vivo by transient expression of both the substrate protein and the key determining protein for ubiquitination, the E3 ligase, in

ª 2010 The Authors Journal compilation ª 2010 Blackwell Publishing Ltd, The Plant Journal, (2010), 61, 893–903

Efficient technique for detection of ubiquitination 895

CO 5+

P1

HY

CO

5 HY

70 HY5-GFP

55

IgG H Anti-GFP

(b)

Detection of transiently expressed protein ubiquitination in vivo Next, we wanted to know whether we could detect the ubiquitinated form of the HY5 and ELF3 proteins in the sample extracted from agroinfiltrated N. benthamiana leaves. To facilitate the detection, we transfected plants with constructs expressing Myc-HY5 or Myc-ELF3. Total protein from plants inoculated with agrobacterial strains containing either 35S::Myc-HY5 or Myc-ELF3, as well as the control plasmid 35S::Myc-GFP, was extracted, immunoprecipitated with anti-Myc antibody and examined via western blot analysis with an anti-Myc antibody. In addition to bands of the expected sizes for the Myc-HY5 and Myc-ELF3 proteins, a smear of bands corresponding to larger molecules was detected in the Myc-HY5 and the Myc-ELF3 samples, and these bands showed the features of ubiquitinated forms of the Myc-HY5 and Myc-ELF3 proteins (Figure 2a). The same samples were immuno-analyzed with an anti-ubiquitin antibody, and the high molecular size bands could be recognized by the anti-ubiquitin antibody in both Myc-HY5 and Myc-ELF3 samples but not in the Myc-GFP expression

(a)

CO 5+ HY

P1 CO

HY

5

P1

IP anti-GFP

170 130

130 Myc-COP1

100 70

MycELF3

100 70 MycGFP MycHY5

IgG H Anti-Myc

Figure 1. HY5 and COP1 interact with each other in Nicotiana benthamiana. Different cell lysates or a lysate mixture (HY5-GFP, Myc-COP1, HY5-GFP and Myc-COP1) were immunoprecipitated with anti-Myc or anti-GFP antibodies. These immunoprecipitates were then separated by SDS-PAGE and immunoblotted with anti-GFP or anti-Myc antibodies. The numbers on the left show the molecular masses of marker proteins in kilodaltons. The IgG heavy chain (IgG H) is indicated by an arrow. (a) Samples were immunoprecipitated with anti-Myc antibody and immunoblotted with anti-GFP antibody. (b) Samples were immunoprecipitated with anti-GFP antibody and immunoblotted with anti-Myc antibody.

170 130 100 70 55

55

55

(b)

Myc -HY 5 Myc -GF P Myc -EL F3

IP anti-Myc P1

(a)

mixture of Myc-COP1 and HY5-GFP extracts. To verify the interaction, we performed the same assay, replacing the anti-Myc antibody with anti-GFP antibody, and then detected the results with an anti-Myc antibody. In this assay, the MycCOP1 protein could also be detected in the same sample (Figure 1b). Both results indicated that a direct physical interaction occurred between the COP1 and HY5 proteins extracted from transient expression by agroinfiltration.

Myc -HY 5 Myc -GF P Myc -EL F3

N. benthamiana. To demonstrate the validity and applicability of the approach, we selected a well-known ubiquitin regulation complex, the Arabidopsis E3 ligase COP1, which is able to ubiquitinate the transcription factor HY5 to regulate light signaling (Ma et al., 2002; Saijo et al., 2003). Two different binary vectors, one carrying a 35S-driven HY5-GFP expression cassette and the other a 35S-controlled Myc-COP1 expression cassette, were transformed into Agrobacterium tumefaciens EHA105. The two different A. tumefaciens strains were then infiltrated into N. benthamiana separately to express the COP1 or HY5 proteins. The physical interaction between the E3 ligase and the substrate is necessary in the first step to detect ubiquitination activity. Thus, we first detected the interaction between COP1 and HY5 expressed in N. benthamiana by an immunoprecipitation assay. The cell extracts were immunopurified with antiMyc antibody to pull down Myc-COP1 proteins, and then the immunoprecipitated products were fractionated via SDSPAGE and detected with anti-GFP antibody. As shown in Figure 1a, the HY5-GFP protein can only be detected in the

40

Anti-Myc

40

Anti-Ub

Figure 2. Polyubiquitinated Myc-HY5 and Myc-ELF3 can be detected in vivo. Myc-HY5, Myc-ELF3 and Myc-GFP (control) samples were extracted with native extraction buffer 1. The cell lysates were immunoprecipitated with antiMyc antibody. The numbers on the left show the molecular masses of marker proteins in kilodaltons. (a) Immunoblots of immunoprecipitated samples with anti-Myc antibody. (b) Immunoblots of immunoprecipitated samples with anti-ubiquitin (Ub) antibody.


896 Lijing Liu et al. sample (Figure 2b), indicating that these additional bands were ubiquitinated forms of Myc-HY5 or Myc-ELF3. This result demonstrates that this method is suitable to specifically detect the ubiquitination of protein. In this case, the protein(s) functioning as the E3 ligase(s) could be the ortholog of COP1 or another E3 complex in N. benthamiana, which could also ubiquitinate the Myc-HY5 or Myc-ELF3 proteins. Following protein ubiquitination, 26S proteasome-dependent degradation of the polyubiquitinated substrate proteins and possibly the self-ubiquitinated E3 ligases should take place in some cases. To test whether agroinfiltration expression could be used to detect this degradation, two known substrates of the COP1 E3 ligase, HY5 (Ma et al., 2002; Saijo et al., 2003) and ELF3 (Yu et al., 2008), and two known functional E3 ligases, COP1 (Deng et al., 1992; Ang et al., 1998) and SIDR1 (Zhang et al., 2007), were agroinfiltrated into the leaves of N. benthamiana. The construct HA-GFP was co-infiltrated as an internal control. Twelve hours before sample collection, the 26S proteasome inhibitor MG132 was infiltrated into the same region. Then proteins were isolated to perform western blot analysis. Reverse transcriptase-polymerase chain reaction (RT-PCR) was performed for each individual gene that was transiently expressed to avoid experimental error in parallel experiments. As shown in Figure 3(a,b), the obvious effects of the proteasome inhibitor MG132 were detected for the two substrate proteins, HY5 and ELF3, which were similar to the results demonstrated in transgenic Arabidopsis (Osterlund et al., 2000; Yu et al., 2008). For the two E3 ligases, COP1 and

(a)

(c)

(b)

HY5-GFP – +

MG132

MG132

HA-ELF3 – +

HY5-GFP

HA-ELF3

HAGFP

HAGFP

HY5

ELF3

ACT1

ACT1

(d)

Myc-COP1 – +

MG132

MG132

GFP-SDIR1 –

+

Myc-COP1

GFP-SDIR1

HAGFP

HAGFP

COP1

SDIR1

ACT1

ACT1

Figure 3. Effects of MG132 on the stabilities of proteins. Proteins were extracted from plant leaves 3 days after infiltration, and MG132 was infiltrated 12 h before the samples were harvested. Target proteins were detected using the corresponding antibody and hemagglutinin (HA)-GFP (input control; middle lower panel) using an anti-HA antibody. Expressed target genes and ACTIN1 (ACT1) mRNA expression levels were analyzed by RT-PCR (bottom panels). (a) HY5-GFP stability. (b) HA-ELF3 stability. (c) Myc-COP1 stability. (d) GFP-SDIR1 stability.

SDIR1, the effects of MG132, if any, were minor (Figure 3c,d). Because there are no reports on the stability of COP1 or SDIR1, the regulation of COP1 and SDIR1 by the 26S proteasome pathway is not known. A time-course experiment may be needed to give a reliable answer (see below). Nevertheless, the above results for the substrates HY5 and ELF3 indicate that agroinfiltration expression can be applied to study the stability of plant proteins that are specifically controlled by the 26S proteasome pathway. Analysis of E3 ligase-promoted substrate degradation by an in vivo and a semi-in vivo technique Because HY5 and ELF3 can be ubiquitinated in transient expression samples, we wanted to measure whether the degradation of each of these substrates was promoted in the presence of increasing amounts of its E3 ligase or along a time-course. First, we chose the COP1 and ELF3 pair for the in vivo assay to detect E3 ligase-promoted substrate degradation, since the degradation of ELF3 is relatively fast (Yu et al., 2008). We co-infiltrated agrobacterial host constructs expressing HA-ELF3 and FLAG-COP1 together with controls to the same leaf area of N. benthamiana. Samples were then collected for detection of both protein and RNA levels of transfected constructs. As shown in Figure 4(a), as the amount of FLAG-COP1 increased, the protein level of ELF3 decreased. In this experiment, HA-GFP was detected as an internal control, and mRNA of both Actin and ELF3 genes was analyzed by RT-PCR to ensure equal amounts of ELF3 were expressed in different co-infiltrations. Next, we wanted to measure E3-promoted substrate degradation in a timely manner. We selected the COP1 and HY5 pair since the degradation of HY5 is relatively slow (Osterlund et al., 2000), and E3 ligase-promoted substrate degradation is difficult to check in co-infiltration. To achieve this, we expressed both the E3 ligase COP1 and the substrate protein HY5 separately via different agroinfiltrations, and the samples were then mixed together to perform the degradation assay. Reaction samples were removed at different time points, transferred to a loading buffer to stop the reaction and analyzed with an anti-GFP antibody to determine the status of the HY5 protein. As shown in Figure 4(b), intact HY5-GFP was slightly reduced at 2 h after the addition of COP1 to the reaction, and a dramatic reduction was found in the 4- and 6-h reaction samples in the presence of COP1 protein. As a consequence, the degraded form of the HY5-GFP protein increased with treatment time. Degraded HY5-GFP protein could also be detected in the control sample without COP1, presumably because N. benthamiana contains similar E3 complexes. To demonstrate that the degradation of the HY5-GFP protein took place via the ubiquitination pathway rather than another pathway, the ubiquitinmediated degradation-specific inhibitor MG132 was added to the in vivo degradation assay. We first performed this



(b) 10

10

10

10

FLAGCOP1 0 pBAMyc 10 HAGFP 1

1 9 1

2 8 1

5 5 1

10 0 1 HAELF3

HAGFP

40

ACT1

Loading

2

FLAG-COP1

+M G1 3

(d)

HY

5+

CO

P1

6h

6h

HY5-GFP

ELF3

+M G1 3 6h

P1 CO

HY

5+

ck mo

5+

HY

HY

5+

mo

ck

6h

0h P1

0h ck

CO

5+

mo

HY

5+ HY

55

55

2

(c)

*FLAGCOP1

HY5-GFP

HY 5+ mo ck HY 0h 5+ CO P1 HY 0h 5+ mo ck HY 2h 5+ mo c k2 HY h+ 5+ CO MG P1 13 HY 2 2h 5+ CO P1 2h +M G1 32

HAELF3 10

HY 5+ mo HY ck 5+ 0h CO HY 5+ P1 0 HY moc h k 5+ CO 2h HY P1 5+ mo 2h HY 5+ ck 4 h CO HY P1 5+ 4 h HY moc k 5+ CO 6h P1 6h

(a)

55

HY5-GFP

40 40

FLAG-COP1

FLAG-COP1

Loading

Loading

Figure 4. COP1-promoted HY5 and ELF3 degradation and MG132 inhibition can be detected in vivo. (a) In vivo degradation of ELF3 was carried out by detecting the hemagglutinin (HA)-ELF3 protein level in co-infiltration experiments with increasing amounts of COP1. The HA-ELF3 and FLAG-COP1 proteins were detected using the corresponding antibodies and HA-GFP (input control; middle panel) using an anti-HA antibody. Target gene ELF3 and ACTIN1 (ACT1) mRNA expression levels were analyzed by RT-PCR (bottom panels). Numbers indicate the ratio of the concentrations of agrobacteria used in co-infiltration. *, endogenous non-specific protein of Nicotiana benthamiana. (b) Time-course of COP1-promoted HY5 degradation. HY5 degradation was performed by mixing cell extracts from separately infiltrated HY5-GFP and FLAGCOP1samples. The HY5-GFP extract was mixed with FLAG-COP1 extract or mock control extract and then incubated at 4C with gentle shaking. Samples were collected at different time points for the assay. HY5-GFP was detected by anti-GFP antibody and FLAG-COP1 by anti-FLAG antibody. Ponceau S staining (bottom panel) of the rubisco protein is shown as a loading control. The numbers on the left show the molecular masses of marker proteins in kilodaltons. w, the degraded form of HY5-GFP. (c) The effect of proteasome inhibitor MG132 on HY5 degradation by COP1. Similar to (b) except MG132 was added to the corresponding protein mixture samples to a final concentration of 50 lM to prevent protein degradation through 26S proteasome. The reaction was carried out at 4C. (d) Same as (c) except the assay was performed at room temperature (25C).

assay at 4C because we and others have found the degradation of most substrate proteins to be very fast and difficult to manage at higher temperatures. As shown in Figure 4(c), without the addition of the E3 ligase COP1, the degradation of HY5 was substantially blocked by MG132. Degradation of the HY5 protein in the sample containing COP1 was also greatly reduced after MG132 treatment. Because HY5 is a relatively slowly degraded protein, we then conducted the same experiment at room temperature to check whether higher temperature could promote the degradation process. Indeed, higher temperature promoted the rapid degradation of HY5-GFP protein, which was slowed by MG132. Taken together, the above results demonstrate that we were able to use agroinfiltration to

detect specific ubiquitination events as well as ubiquitindependent degradation in vivo and semi-in vivo. For E3 ligases and substrates whose interactions are only detectable under certain conditions or in the presence of other components of the degradation process, we recommend the semi-in vivo assay that we have described. Proteins expressed via agroinfiltration can be used for in vitro ubiquitination An in vitro ubiquitination assay for a substrate is essential to demonstrate specificity between the particular E3 ligase and the substrate, as in vivo ubiquitination could be due to an indirect effect. It is well known that, due to a lack of additional modifications, some substrate proteins


898 Lijing Liu et al. expressed in bacterial expression systems cannot be ubiquitinated in vitro. In the animal research field, purified substrates expressed from cell cultures can be ubiquitinated by E3 ligases in vitro (Fang et al., 2000; Chen et al., 2008). Thus, we mixed the HY5-GFP and Myc-COP1 protein extracts together and purified the HY5-GFP-Myc-COP1 complex by immunoprecipitation with anti-Myc antibody. The immunoprecipitated products were used in an in vitro ubiquitination reaction in the presence of E1, E2 (UbcH5B) and His-tagged Ubiquitin (His-Ub). The E1, E2 (UbcH5B) and Ub proteins were expressed as recombinant proteins in Escherichia coli and purified. The ubiquitination reaction mixture was resolved via SDS-PAGE and immunoblotted with an antiGFP antibody. Higher-molecular-weight species indicative of the addition of multiple Ub moieties were seen only in the presence of added E1 and other ubiquitination components (Figure 5a). Therefore, it is evident that COP1-mediated HY5 ubiquitination also specifically depends on E1 and E2. The same reaction samples were detected by nickel-horseradish peroxidase (HRP) for His-Ub, and higher-molecular-weight species were only detected in the reaction with E1 and E2 (Figure 5b). These high-molecular-weight bands reflect ubiquitinated HY5-GFP, self-ubiquitinated Myc-COP1 and perhaps some other ubiquitinated proteins from the MycCOP1/HY5-GFP complex. These results indicate that proteins transiently expressed via agroinfiltration can be used in the in vitro ubiquitination assay to detect E3 ligase/substrate specificity.

(a)

(b)

HY5+COP1 E1 – + + E2 + – + Ub + + + 170 130 100

E1 E2 Ub 170 100 55 40 35 25

70 HY5 IgG H

55 Anti-GFP

HY5+COP1 – + + + – + + + +

15

His-Ub Nickel-HRP (Ub)

Figure 5. Substrate proteins expressed via agroinfiltration can be used for ubiquitination reactions in vitro. HY5-GFP and Myc-COP1 samples were extracted by native extraction buffer 1. Then, the cell lysates were mixed and immunoprecipitated with anti-Myc antibody. The immunoprecipitated product was applied for a further in vitro ubiquitination assay. E1 (from wheat), E2 (UBCh5b), and 6 · His-tagged ubiquitin (Ub) were added to the reaction. The numbers on the left show the molecular masses of marker proteins in kilodaltons. IgG heavy chain (IgG H) is also shown. w, mixture of poly-ubiquitinated HY5-GFP and Myc-COP1. , mono-ubiquitinated E2. (a) Immunoblots of the in vitro ubiquitination assay samples with anti-GFP antibody. (b) Immunoblots of the in vitro ubiquitination assays samples with nickelhoreseradish peroxidase (HRP) to detect His-ubiquitin.

Optimal conditions for substrate and E3 ligase protein expression It has been reported that different agrobacterial strains carrying the same binary vector could produce different expression levels of proteins (Wroblewski et al., 2005). Additionally, gene-silencing suppressors have been shown to enhance protein expression levels (Voinnet et al., 2003; Ma et al., 2008). To avoid such complications and to select the optimal conditions for protein expression by agroinfiltration, we selected the commonly used Agrobacterium strain EHA105 as a host for most of our constructs because it works well with many binary vectors. Since we found that expression levels varied widely between the different gene constructs (Figure 6 and unpublished data), we tested samples at different times post-inoculation in the presence or absence of the co-infiltrated strong gene-silencing suppressor p19. Four different types of proteins, representing a wide range of sizes and different expression tags, were collected over 10 days. Because in most cases proteins are rarely detectable after 5 days, we show the results within 5 days of inoculation. As shown in Figure 6(a), HY5 fused to GFP gave a high expression level in the 1-day samples, and there was no dramatic difference in the presence or absence of p19. However, in the 3-day samples, the HY5-GFP protein could only be detected in samples co-inoculated with p19. In the 5-day samples, no HY5-GFP protein was detected in either the presence or absence of p19. Therefore, for high levels of expression of HY5-GFP, 1 day is the best time for sample collection (Figure 6a). In the case of HA-tagged ELF3, the HA-ELF3 protein was detected at all three time points when co-inoculated with p19, but reduction of the protein expression level was observed post-agroinfiltration. Without the help of p19, HA-ELF3 protein could only be detected in the 1-day sample. Thus, the best expression conditions for HA-ELF3 are very similar to those for HY5-GFP agroinfiltration (Figure 6b). The opposite result was observed in Myctagged COP1; no Myc-COP1 protein was detected in 1-day samples with or without p19. An increasing amount of MycCOP1 protein was found following the post-agroinfiltration period, and the greatest expression efficiency was observed 5 days after co-infiltration with p19 (Figure 6c). A very diffident expression pattern was observed for GFP-tagged SDIR1 (GFP-SDIR1, about 57 kDa), a membrane-bound RING finger E3 ligase in the ABA signaling pathway (Zhang et al., 2007). GFP-SDIR1 was not detected at any of the three time points of agroinfiltration without the addition of p19, and the protein could only be detected in the 3-day and 5-day samples, with a slight increase observed at the later time point. The best expression conditions for GFP-SDIR1 were similar to those of Myc-COP1, but p19 was essential for expression by agroinfiltration (Figure 6d). Together with the results we obtained for the expression of other proteins of the ubiquitination complex, such as RHF E3 ligase and KRP/ICKs



(a) DAI P19 70

HY5-GFP 1 –

1 +

3 –

3 +

(a)

5 –

5 + WT

HY5-GFP DB NB1

(b)

Myc-COP1 DB

NB2

70

130

55

100

NB1 NB2

55

(c)

HA-ELF3

(b) DAI P19 130 100

1 –

1 +

1 –

1 +

5 –

5 +

3 –

3 +

5 –

5 + WT

1 –

1 +

3 –

3 +

5 –

5 + WT

Figure 6. The effects of expression time and p19 on substrate and E3 ligase protein levels. After different inoculation times, proteins were extracted from samples with or without p19 with a denaturing buffer and resolved by 10% SDS-PAGE. Antibodies corresponding to protein tags were applied to detect individual proteins. Mock (wild type, WT) was used as a negative control. Ponceau staining (lower panel) was done as a loading control. DAI, days after infiltration. The numbers on the left show the molecular masses of marker proteins in kilodaltons. (a) The expression of HY5-GFP, detected by anti-GFP. (b) The expression of HA-ELF3, detected by anti-HA (HA, hemagglutinin). (c) The expression of Myc-COP1, detected by anti-Myc. (d) The expression of GFP-SDIR1, detected by anti-GFP.

(Liu et al., 2008), and other unpublished results, we found that in most cases the gene-silencing suppressor p19 was able to increase the expression levels of proteins. We suggest that it is better to test the time-course of a particular protein’s expression with p19. Furthermore, we noticed in many cases that substrate proteins had high expression levels at short time points after agroinfiltration, while E3 ligases accumulated at high levels much later, although this

(d)

GFP-SDIR1 DB

NB1 NB2

130

70

100

55

WT

GFP-SDIR1

(d) DAI P19 70 55

3 +

Myc-COP1

(c) DAI P19 130 100

3 –

HA-ELF3 DB

NB1 NB2

Figure 7. Extraction buffers for different detections. Total proteins were extracted from samples 3 days after infiltration. For each extraction, the same weight samples and same volumes of different extraction buffers were used. Proteins were resolved by 10% SDS-PAGE. Antibodies corresponding to protein tags were applied for analysis. DB, denature buffer. NB1, native extraction buffer 1. NB2, standard native extraction buffer 2. Ponceau staining (lower panel) was done as a loading control. The numbers on the left show the molecular masses of marker proteins in kilodaltons. (a) HY5-GFP protein detected by anti-GFP. (b) HA-ELF3 protein detected by anti-HA (hemagglutinin). (c) Myc-COP1 protein detected by anti-Myc. (d) GFP-SDIR1 protein detected by anti-GFP.

pattern was not always observed. Three days post-agroinfiltration, proteins could be detected in most samples we tested. We also found that different agrobacterial strains could be co-infiltrated together at the same site on N. benthamiana plants to express different proteins. Also, there were no significant differences in expression levels resulting from the same Agrobacterium strain hosting different binary vectors that were co-infiltrated together. Up to five different protein expression constructs could be co-agroinfiltrated, and all proteins were well expressed. This indicates the feasibility of using one construct, such as a GFP or GUS protein, as an internal control to monitor expression efficiency in different agroinfiltrations. Protein extraction buffers for different detections Proteins expressed in agroinfiltrated samples can be used to detect ubiquitination or the results of ubiquitination reactions; thus, different protein extract buffers might be necessary for these different purposes. To verify the in vivo stability of substrate proteins or the effects of E3 ligases or other components on the co-agroinfiltrations, a standard denaturing protein extraction buffer worked well for all proteins we tested (Figure 7). If the extracted protein is to be used for detecting protein–protein interactions, immunoreactions or other interactions or chemical reactions, a native extraction buffer is necessary to preserve the activity of the protein. However, the standard native sodium-containing extraction buffer failed to isolate a large number of proteins due to the dissolution of certain


900 Lijing Liu et al. proteins in this buffer. As shown in Figure 7, proteins were difficult to detect in either the HA-ELF3 or the Myc-COP1 samples, and only a small amount of protein was found in the HY5-GFP sample. This result indicates that the standard native extraction buffer was not suitable for our purposes. To resolve this problem, different native protein extraction buffers were tested, and we found that native extraction buffer 1 worked well with all proteins tested (Figure 7). DISCUSSION Ubiquitination is a complex process of protein modification. It plays important roles in almost all plant developmental processes and plant–environment interactions. In the Arabidopsis genome, nearly 6% of all proteins are predicted to be involved in protein ubiquitination (Downes and Vierstra, 2005). Recently, it was reported that the functions of E3 ligases and the stability of substrates are central to biological processes in both model and cultivated plants. To complement the genetic and physiological analyses, it is important to provide biochemical and molecular data to demonstrate the functions of E3 ligases or substrates in depth. Even though in some cases abundant proteins can be detected directly in vivo or by transient expression in protoplasts, the low expression of many different regulator proteins in planta makes it difficult to detect the ubiquitination reaction process and its products in vivo. To resolve this problem, we used the knowledge that agroinfiltration (Kapila et al., 1997), together with the effects of gene-silencing suppressors, can dramatically increase protein expression levels in planta (Voinnet et al., 2003; Ma et al., 2008) to develop an efficient system for the analysis of protein ubiquitination in planta. Agroinfiltration expression of proteins for the ubiquitination assay has the same advantages as agroinfiltration for other experimental assays, such as detecting the subcellular localization and function of proteins or RNAs (Johansen and Carrington, 2001; Goodin et al., 2002): it is fast, no transgenic plants need to be created, and it is also convenient to increase the sample numbers because the experimental process is simple. It is an especially good technique for observing the functional phenotypes of low-abundance and unstable proteins because those proteins are usually undetectable even in transgenic plants (Zhang et al., 2007). Transient expression in protoplasts has been used as an alternative way to express proteins for further assays, and it was also successful for most types of analysis usually performed in transgenic plants. However, the quantity of protein that can be produced is still a limiting factor because it is difficult to prepare protoplasts on a large scale and contamination often happens during plasmid DNA transfection. Unlike the two aforementioned systems, transient expression of proteins in N. benthamiana by agroinfiltration has been proven to yield high levels of proteins in a short time. Gene-silencing suppressors have been demonstrated

to enhance protein expression levels up to 100-fold (Voinnet et al., 2003). A high level of protein expression provides the essential starting material for further protein modification analysis. The disadvantage is that N. benthamiana is a heterologous system for most proteins from other plant species. There is no mutant collection available for deep analysis of gene or protein functions. Nonetheless, N. benthamiana is a well-studied model plant species, and it possesses every kind of protein modification known in Arabidopsis. Thus, proteins expressed in N. benthamiana should have the necessary modifications for correct folding and activity. In most cases, phosphorylation is necessary prior to protein ubiquitination (Hershko and Ciechanover, 1998), and N. benthamiana fully supports the phosphorylation function. Ideally, if agroinfiltration transexpression in Arabidopsis were as efficient as it is in N. benthamiana it would combine all advantages of transexpression in N. benthamiana with the collection of mutants in Arabidopsis. However, it was reported that agroinfiltration did not work well enough in Arabidopsis to give sufficient amounts of proteins for further analysis (Ueki et al., 2009). We also found that different Agrobacterium strains, or the same strain carrying different constructs, can be co-infiltrated. We observed similar expression levels of proteins between the individual infiltrations, which might be due to homogeneous integration of the binary vectors into the genome. Up to five Agrobacterium strains carrying different plasmids were tested in our experimental assay, and all proteins were well expressed. This is important, because in addition to a construct for the gene-silencing suppressor, an internal control construct is also necessary for quantitative analysis. Moreover, in some ubiquitination reactions, different subunits, such as the F-box of the E3 complex, are necessary for efficient ubiquitination. Importantly, we found that the ubiquitination 26S proteasome inhibitor MG132 works very well in the agroinfiltration transexpression assay. This may provide additional opportunities to use this ubiquitination assay to isolate many E3 substrate proteins from N. benthamiana for further analysis. The different protein levels we detected at different time points after the agroinfiltration of four constructs might have been due to inherent differences between the binary vectors or to the stability of the individual proteins. We recommend pilot expression experiments be conducted to measure the optimal expression times for particular protein sets. Different gene-silencing suppressors were reported to greatly enhance protein expression levels in N. benthamiana (Johansen and Carrington, 2001; Voinnet et al., 2003). We found that a gene-silencing suppressor was not essential for proteins with high expression levels in transient agroinfiltration, but it was indeed helpful for proteins with low expression levels, especially for slowly expressed proteins, such as COP1 and SDIR1 (Figure 6). Generally, 3 days postinfiltration is a good time point for most sample collections,


Efficient technique for detection of ubiquitination 901 but we suggest testing a time-course for particular protein collections and adding p19 when the protein of interest is known to be particularly unstable. One point we should emphasize is that the growth status of N. benthamiana is very important for the protein expression level. Young plants with healthy growth at the seven- to eight-leaf stage usually give the highest protein expression. Interestingly, we found N. benthamiana plants grown at 22C gave higher protein production than at the standard growth temperature of 26C. Whether the lower temperature increased the stability of expressed proteins or had other effects was not investigated. Because the expression of different proteins is not synchronized, such as in the case of the E3 ligase and its substrate being expressed at different time points, and because E3 ligase/substrate pairs may be localized to separate compartments under certain conditions (e.g. COP1 is localized to the nucleus in the dark), it would be difficult to observe the effect of the E3 ligase on the substrate by co-infiltration in vivo. However, the reaction can be conducted by mixing two or even more independent protein extracts together to analyze the modification or degradation of the substrate protein in the presence of the E3 ligase. In this way, we can measure the degradation kinetics as well as other additional factors. Because the process of ubiquitin-mediated proteolysis is very fast for most substrate proteins at room temperature, the semi-in vivo analysis at 4C described above is a viable alternative. Even the supplementation of ubiquitination components, such as Ub, E1 and E2, may not be essential in this kind of degradation assay, but the additional ubiquitination components will promote the degradation process and facilitate the detection of results. The protein extraction buffer used was also important, especially when samples are to be used for further analysis. Native protein extraction buffer with a certain osmotic NaCl-free reagent was found to be necessary to preserve protein folding and activity. Samples should be treated as soon as possible after protein extraction, as ubiquitination-mediated degradation was observed even when the sample was stored on ice or after freezing and thawing. Because almost all transition processes in eukaryotic cells are regulated by ubiquitination pathways, and more than 1300 E3 ligases have been found in the Arabidopsis genome, most of which have unknown functions, the establishment of an efficient method to analyze the E3 ligase and substrate complex is of supreme importance for many plant scientists. Agroinfiltration expression of proteins in N. benthamiana provides a very convincing method for detecting the ubiquitination reaction in vivo and in vitro. This approach, combined with genetic and physiological analyses, can greatly improve our understanding of protein ubiquitination in plants. Even through the full genome of N. benthamiana has not been sequenced, a collection of more than 42,659 expressed sequence tags has been established (Solanaceae Genomics Resource, http://solanaceae.plantbiology.msu.

edu/species_overview.php?sp=4100). This provides enough information to use the pull-down approach to isolate interacting proteins by agroinfiltration expression of tagged proteins in N. benthamiana. Then the orthologs of the interaction partners can be identified in other plants, such as Arabidopsis, to perform further functional analysis, taking advantage of the large mutant collections available in Arabidopsis. An alternative method of identifying E3/substrate pairs would be to silence the orthologous gene, such as the E3 ligase, in N. benthamiana and then to analyze the stability of the putative substrate expressed by agroinfiltration. Because N. benthamiana exhibits every kind of protein modification thus far discovered in Arabidopsis, we believe this system can also be applied to the study of other kinds of protein modification, such as phosphorylation and glycosylation, and even to the isolation and functional examination of protein–RNA complexes. EXPERIMENTAL PROCEDURES Plant materials and growth conditions Wild-type N. benthamiana plants were selected as the host plants for this study. Plants were grown in a growth chamber at 22C and 70% relative humidity (RH) under a 16-h light/8-h dark photoperiod for about 1–1.5 months before infiltration. After infiltration, plants were kept under the same growth conditions.

Constructs used in the analysis For the 35S::Myc-COP1 construct, a PCR-amplified fragment of COP1 DNA was inserted into EcoRI/SpeI-digested pBluescript II SK+ to give pSK-COP1. A 6xc-Myc sequence was then inserted into XbaI/ BamHI-digested pSK-COP1 to give the intermediate of pSK-MycCOP1, and the fragment of XbaI/SacI derived from pSK-Myc-COP1 was cloned into pCAMBIA-1300-221 digested with XbaI/SacI to give the construct 35S::Myc-COP1. The 35S::Flag(3·)-COP1 and 35S::HA(3·)-ELF3 were gifts from Xing Wang Deng (Yu et al., 2008). The 35S::HY5-GFP construct was also kindly provided by Xing Wang Deng. 35S::GFP-SDIR1 was constructed as described previously (Zhang et al., 2007). 35S::Myc-HY5 or 35S::Myc-ELF3 was prepared by inserting a BamHI/SpeI HY5 or ELF3 fragment into the plant binary vector pBA002 with a Myc tag. For the 35S::Myc-GFP construct, the XhoI/Klenow–SacI GFP fragment derived from pGFP2 was inserted into pBA002-Myc digested with SmaI/SacI. 35S::HAGFP was prepared by inserting a XhoI/SacI GFP fragment into the plant expression vector pCanG with a hemagglutinin (HA) tag. The 35S::p19 construct was prepared by inserting a PCR-amplified fragment of p19 DNA, obtained by RT-PCR amplification using tomato bushy stunt virus (TBSV) RNA as template, into HindIII/ EcoRI-digested pBluscript II SK+ to yield the intermediate construct pSK-p19. The SalI/Klenow–SmaI fragment derived from pSK-p19 was then inserted into pCAMBIA-1300-221, digested with SmaI/SalI and treated with T4 DNA polymerase to give the construct 35S::p19.

Agroinfiltration procedure Agrobacterium tumefaciens strains EHA105 and ABI were used in these experiments. The strains were first plated on LB medium containing the appropriate selection antibiotics. After 2–3 days, a single colony was inoculated into 5 ml LB medium supplemented with the appropriate antibiotics and grown at 28C in a shaker for


902 Lijing Liu et al. 48 h. The culture was transferred to new LB medium with 10 mM 2-(N-morpholine)-ethanesulfonic acid (MES; pH 5.6) and 40 lM acetosyringone (1:100 ratio, v/v). Bacteria were grown at 28C for 16 h. When growth reached an OD600 of approximately 3.0, the bacteria were spun down gently (3200g, 10 min), and the pellets were resuspended in 10 mM MgCl2 at a final OD600 of 1.5 or 1 (OD600 = 1 only for p19). A final concentration of 200 lM acetosyringone was added and the bacteria were kept at room temperature for at least 3 h without shaking. For co-infiltration, equivolume suspensions of different Agrobacterium strains carrying different constructs were mixed prior to infiltration. Leaf infiltration was conducted by depressing a 1-ml disposable syringe to the surface of fully expanded leaves and slowly depressing the plunger. Infiltrated leaves had a water-soaked appearance.

Protein extraction and immunoblot analysis The infiltrated parts of N. benthamiana leaves were harvested, and then the leaf tissue was ground in liquid nitrogen and resuspended in extraction buffer on ice. Total extract was centrifuged at 16 000g at 4C for 30 min, and supernatant was subjected to protein gel blots. Three different protein extraction buffers were used: denaturing buffer [DB; 50 mM 2-amino-2-(hydroxymethyl)-1,3-propanediol (TRIS)-HCl pH 7.5, 150 mM NaCl, 0.1% NP-40, 4 M urea, 1 mM phenylmethylsulfonyl fluoride (PMSF; optional)], native extraction buffer 1 [NB1; 50 mM TRIS-MES pH 8.0, 0.5 M sucrose, 1 mM MgCl2, 10 mM EDTA, 5 mM DTT, protease inhibitor cocktail CompleteMini tablets (Roche, http://www.roche.com/)] and standard native extraction buffer 2 [NB2; sodium phosphate 50 mM pH 7.0, NaCl 200 mM, MgCl 10 mM, glycerol, 10%, protease inhibitor cocktail tablets (Roche), PMSF 4 mM]. For immunoblot analysis, proteins were separated by SDS-PAGE in a 10% acrylamide gel and electroblotted to nitrocellulose membrane (Hybond-C, Amersham, http://www6.gelifesciences. com/) at 100 V for 75 min. The membrane was blocked with PBS containing 5% skimmed milk powder for 1 h at room temperature. The membrane was then incubated first with primary antibody and then with secondary antibody diluted in PBS containing 3% skimmed milk for 1 h at room temperature. Bands were detected with the Millipore chemiluminescent HRP substrate kit (http:// www.millipore.com/). Antibodies and the dilutions used in these experiments were as follows: anti-HA antibody (sc-7392AC Santa Cruz, 1:500; http://www.scbt.com/), anti-Myc antibody (sc-40 Santa Cruz, 1:500), anti-GFP antibody (JL-8 Clontech/NO 1814460 Roche, 1:1000; http://www.clontech.com/), anti-FLAG antibody (F3165 Sigma, 1:3000; http://www.sigmaaldrich.com/), anti-ubiquitin monoclonal antibody (raised in our laboratory) and goat antimouse HRP-conjugated antibody (00001-1 Proteintech, 1:2500; http://www.ptglab.com).

RT-PCR amplification To examine the expression of genes by RT-PCR, 2 lg total RNA was subjected to a reverse transcription reaction at 42C for 1 h. The PCR amplification was performed using gene-specific forward (FW) and reverse (REV) primers for 28 cycles. Expression levels of Actin1 were monitored with FW and REV primers to serve as an internal control. The following RT-PCR primers were used: ELF3 (FW: 5¢-GATAAATGAAGAGGCAAGTGATG-3¢, REV: 5¢-GTTGATGATGACCTTGATTTGAC-3¢), COP1 (FW: 5¢-GCTTGTGGTCATAGTTTCTGC-3¢, REV: 5¢-TTGCACCTCATTTAGTTCATC-3¢), HY5 (FW: 5¢-CCATCAAGCAGCGAGAGGTC-3¢, REV: 5¢-GCATTAGAACCACCACCACC-3¢), SDIR1 (FW: 5¢-ATGAGCTTTGTTTTCCGGGG-3¢, REV: 5¢-TCAAACCATGTCGGAAGCATC-3¢), Actin (FW: 5¢-CATCAGGAAGGACTTGTACGG-3¢, REV: 5¢-GATGGACCTGACTCGTCATAC-3¢).

In vivo and semi-in vivo protein degradation For in vivo protein degradation experiments, agrobacterial strains carrying constructs of E3 ligase, substrate and p19 genes, as well as internal control plasmids, were co-infiltrated at different ratios. Three days after infiltration, samples were collected for analysis. For semi-in vivo protein degradation analysis, 1 day after infiltration, a HY5-GFP sample was harvested. Three days after infiltration, a MycCOP1 sample and a wild-type (WT) sample were harvested. These three samples were separately extracted, as described, in native extraction buffer 1 (NB1). A final concentration of 10 lM ATP was added to the cell lysates to preserve the function of the 26S proteasome. Then HY5-GFP extract was mixed with Myc-COP1 or WT extract in a volume ratio of 1:1. MG132 was added to the corresponding mixtures to a final concentration of 50 lM. The mixtures were incubated at 4C or room temperature (25C) with gentle shaking. Samples were removed at different time points and the reaction was stopped by the addition of SDS sample buffer and boiling for 5 min before gel analysis.

Immunoprecipitation Samples were extracted with native extraction buffer 1 (NB1) as described above. Corresponding antibodies were added to the cell lysates (10 lg ml)1) and MG132 was also added at a final concentration of 50 lM to prevent protein degradation. The mixtures were kept at 4C with gentle shaking for 3 h or overnight. The immunocomplex was captured by adding 20 ll ml)1 protein G agarose beads (16-266 Millipore) and shaking at 4C for another 3 h. The agarose beads were recovered by centrifugation at 14 000 g for 5 sec and washed with cold PBS three times.

In vitro ubiquitination assay The HY5-GFP and Myc-COP1 cell lysate mixture was immunoprecipitated with anti-Myc antibody as described above. The product of the immunoprecipitation was used for an in vitro ubiquitination assay. In addition to the immunoprecipitation product, wheat (Triticum aestivum) E1 (GI: 136632), human E2 (UBCh5b) and purified, bacterially expressed, His-tagged Arabidopsis ubiquitin (UBQ14) were also used in the reaction. Reactions were carried out at 30C with agitation in an Eppendorf Thermomixer for 1.5 h. For the immunoblots, nickel–nitrilotriacetic acid agarose conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories, http://www.kpl.com/) was used for the detection of His-tagged ubiquitin, and anti-GFP antibody was used for HY5-GFP.

ACKNOWLEDGEMENTS We would like to thank Dr Nam-Hai Chua of Rockefeller University for kindly providing us with the pBA002 Myc vector and Dr Xing Wang Deng for HY-GFP, HA-ELF3 and Flag-COP1 constructs. This research was supported by grant CNSF30530400/90717006 from the Chinese Natural Science Foundation. QX is supported by grants KSCX2-YW-N-010 and CXTD-S2005-2 from the Chinese Academy of Science. HG is supported by CNSF grant 30525004 and 90919010.

REFERENCES Abel, S. and Theologis, A. (1994) Transient transformation of Arabidopsis leaf protoplasts: a versatile experimental system to study gene expression. Plant J. 5, 421–427. Ang, L.H., Chattopadhyay, S., Wei, N., Oyama, T., Okada, K., Batschauer, A. and Deng, X.W. (1998) Molecular interaction between COP1 and HY5 defines a regulatory switch for light control of Arabidopsis development. Mol. Cell, 1, 213–222.


Efficient technique for detection of ubiquitination 903 Chakrabarty, R., Banerjee, R., Chung, S.M., Farman, M., Citovsky, V., Hogenhout, S.A., Tzfira, T. and Goodin, M. (2007) PSITE vectors for stable integration or transient expression of autofluorescent protein fusions in plants: probing Nicotiana benthamiana-virus interactions. Mol. Plant Microbe Interact. 20, 740–750. Chen, G., Huang, H., Frohlich, O., Yang, Y., Klein, J.D., Price, S.R. and Sands, J.M. (2008) MDM2 E3 ubiquitin ligase mediates UT-A1 urea transporter ubiquitination and degradation. Am. J. Physiol. Renal Physiol. 295, F1528– F1534. Corn, P.G. (2007) Role of the ubiquitin proteasome system in renal cell carcinoma. BMC Biochem. 8(Suppl 1), S4. Deng, X.W., Matsui, M., Wei, N., Wagner, D., Chu, A.M., Feldmann, K.A. and Quail, P.H. (1992) COP1, an Arabidopsis regulatory gene, encodes a protein with both a zinc-binding motif and a G beta homologous domain. Cell, 71, 791–801. Devoto, A., Muskett, P.R. and Shirasu, K. (2003) Role of ubiquitination in the regulation of plant defence against pathogens. Curr. Opin. Plant Biol. 6, 307–311. Dikshit, P. and Jana, N.R. (2008) Role of ubiquitin protein ligases in the pathogenesis of polyglutamine diseases. Neurochem. Res. 33, 945–951. Downes, B. and Vierstra, R.D. (2005) Post-translational regulation in plants employing a diverse set of polypeptide tags. Biochem. Soc. Trans. 33, 393– 399. Dreher, K. and Callis, J. (2007) Ubiquitin, hormones and biotic stress in plants. Ann. Bot. (Lond). 99, 787–822. Fang, S., Jensen, J.P., Ludwig, R.L., Vousden, K.H. and Weissman, A.M. (2000) Mdm2 is a RING finger-dependent ubiquitin protein ligase for itself and p53. J. Biol. Chem. 275, 8945–8951. Goodin, M.M., Dietzgen, R.G., Schichnes, D., Ruzin, S. and Jackson, A.O. (2002) pGD vectors: versatile tools for the expression of green and red fluorescent protein fusions in agroinfiltrated plant leaves. Plant J. 31, 375–383. Hellmann, H. and Estelle, M. (2002) Plant development: regulation by protein degradation. Science, 297, 793–797. Hershko, A. and Ciechanover, A. (1998) The ubiquitin system. Annu. Rev. Biochem. 67, 425–479. Hicks, K.A., Millar, A.J., Carre, I.A., Somers, D.E., Straume, M., MeeksWagner, D.R. and Kay, S.A. (1996) Conditional circadian dysfunction of the Arabidopsis early-flowering 3 mutant. Science, 274, 790–792. Johansen, L.K. and Carrington, J.C. (2001) Silencing on the spot. Induction and suppression of RNA silencing in the Agrobacterium-mediated transient expression system. Plant Physiol. 126, 930–938. Kapila, J., de Rycke, R., van Montagu, M. and Angenon, G. (1997) An Agrobacterium-mediated transient gene expression system for intact leaves. Plant Sci. 122, 101–108. Koscianska, E., Kalantidis, K., Wypijewski, K., Sadowski, J. and Tabler, M. (2005) Analysis of RNA silencing in agroinfiltrated leaves of Nicotiana benthamiana and Nicotiana tabacum. Plant Mol. Biol. 59, 647–661. Lee, H.K., Cho, S.K., Son, O., Xu, Z., Hwang, I. and Kim, W.T. (2009) Drought stress-induced Rma1H1, a RING membrane-anchor E3 ubiquitin ligase homolog, regulates aquaporin levels via ubiquitination in transgenic Arabidopsis plants. Plant Cell, 21, 622–641. Liu, J., Zhang, Y., Qin, G. et al. (2008) Targeted degradation of the cyclindependent kinase inhibitor ICK4/KRP6 by RING-type E3 ligases is essential for mitotic cell cycle progression during Arabidopsis gametogenesis. Plant Cell, 20, 1538–1554. Ma, L., Gao, Y., Qu, L., Chen, Z., Li, J., Zhao, H. and Deng, X.W. (2002) Genomic evidence for COP1 as a repressor of light-regulated gene expression and development in Arabidopsis. Plant Cell, 14, 2383–2398. Ma, P., Liu, J., He, H., Yang, M., Li, M., Zhu, X. and Wang, X. (2008) A viral suppressor P1/HC-Pro increases the GFP Gene expression in agrobacterium-mediated transient assay. Appl. Biochem. Biotechnol. 158, 243–252.

McWatters, H.G., Bastow, R.M., Hall, A. and Millar, A.J. (2000) The ELF3 zeitnehmer regulates light signalling to the circadian clock. Nature, 408, 716–720. Mokrzycki-Issartel, N., Bouchon, B., Farrer, S., Berland, P., Laparra, H., Madelmont, J.C. and Theisen, M. (2003) A transient tobacco expression system coupled to MALDI-TOF-MS allows validation of the impact of differential targeting on structure and activity of a recombinant therapeutic glycoprotein produced in plants. FEBS Lett. 552, 170– 176. Nie, J., McGill, M.A., Dermer, M., Dho, S.E., Wolting, C.D. and McGlade, C.J. (2002) LNX functions as a RING type E3 ubiquitin ligase that targets the cell fate determinant Numb for ubiquitin-dependent degradation. EMBO J. 21, 93–102. Osterlund, M.T., Hardtke, C.S., Wei, N. and Deng, X.W. (2000) Targeted destabilization of HY5 during light-regulated development of Arabidopsis. Nature, 405, 462–466. Rodriguez, M., Ramirez, N.I., Ayala, M., Freyre, F., Perez, L., Triguero, A., Mateo, C., Selman-Housein, G., Gavilondo, J.V. and Pujol, M. (2005) Transient expression in tobacco leaves of an aglycosylated recombinant antibody against the epidermal growth factor receptor. Biotechnol. Bioeng. 89, 188–194. Saijo, Y., Sullivan, J.A., Wang, H., Yang, J., Shen, Y., Rubio, V., Ma, L., Hoecker, U. and Deng, X.W. (2003) The COP1-SPA1 interaction defines a critical step in phytochrome A-mediated regulation of HY5 activity. Genes Dev. 17, 2642–2647. Samach, A., Klenz, J.E., Kohalmi, S.E., Risseeuw, E., Haughn, G.W. and Crosby, W.L. (1999) The UNUSUAL FLORAL ORGANS gene of Arabidopsis thaliana is an F-box protein required for normal patterning and growth in the floral meristem. Plant J. 20, 433–445. Stone, S.L. and Callis, J. (2007) Ubiquitin ligases mediate growth and development by promoting protein death. Curr. Opin. Plant Biol. 10, 624–632. Stone, S.L., Williams, L.A., Farmer, L.M., Vierstra, R.D. and Callis, J. (2006) KEEP ON GOING, a RING E3 ligase essential for Arabidopsis growth and development, is involved in abscisic acid signaling. Plant Cell, 18, 3415– 3428. Turnbull, E.L., Rosser, M.F. and Cyr, D.M. (2007) The role of the UPS in cystic fibrosis. BMC Biochem. 8(Suppl 1), S11. Ueki, S., Lacroix, B., Krichevsky, A., Lazarowitz, S.G. and Citovsky, V. (2009) Functional transient genetic transformation of Arabidopsis leaves by biolistic bombardment. Nat. Protoc. 4, 71–77. Voinnet, O., Rivas, S., Mestre, P. and Baulcombe, D. (2003) An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus. Plant J. 33, 949– 956. Wroblewski, T., Tomczak, A. and Michelmore, R. (2005) Optimization of Agrobacterium-mediated transient assays of gene expression in lettuce, tomato and Arabidopsis. Plant Biotechnol. J. 3, 259–273. Xie, Q., Guo, H.S., Dallman, G., Fang, S., Weissman, A.M. and Chua, N.H. (2002) SINAT5 promotes ubiquitin-related degradation of NAC1 to attenuate auxin signals. Nature, 419, 167–170. Yu, J.W., Rubio, V., Lee, N.Y. et al. (2008) COP1 and ELF3 control circadian function and photoperiodic flowering by regulating GI stability. Mol. Cell, 32, 617–630. Zhang, X., Garreton, V. and Chua, N.H. (2005) The AIP2 E3 ligase acts as a novel negative regulator of ABA signaling by promoting ABI3 degradation. Genes Dev. 19, 1532–1543. Zhang, Y., Yang, C., Li, Y., Zheng, N., Chen, H., Zhao, Q., Gao, T., Guo, H. and Xie, Q. (2007) SDIR1 is a RING finger E3 ligase that positively regulates stress-responsive abscisic acid signaling in Arabidopsis. Plant Cell, 19, 1912–1929.
