- © 2003 American Society of Plant Biologists
Abstract
The Arabidopsis gene AtCHIP encodes a protein with three tetratricopeptide repeats and a U-box domain, which is structurally similar to the animal CHIP proteins, a new class of E3 ubiquitin ligases. Like animal CHIP proteins, AtCHIP has E3 ubiquitin ligase activity in vitro. AtCHIP is a single-copy gene, and its transcript is up-regulated by several stress conditions such as low and high temperatures. However, increased AtCHIP expression alone was not correlated with increased stress tolerance; in fact, overexpression of AtCHIP in Arabidopsis rendered plants more sensitive to both low- and high-temperature treatments. Higher electrolyte leakage was observed in leaves of AtCHIP overexpression plants after chilling temperature treatment, suggesting that membrane function is likely impaired in these plants under such a condition. These results indicate that AtCHIP plays an important role in plant cellular metabolism under temperature stress conditions.
In an effort to identify proteins that bind to GF14λ (a 14-3-3 protein) in a yeast two-hybrid screening, an Arabidopsis gene AtCHIP that encodes a protein with three tetratricopeptide repeats (TPRs) and a U-box domain was isolated. The TPR is a 34-amino acid repeat that participates in protein-protein interactions (Blatch and Lassle, 1999). Many genes encoding TPR-containing proteins were identified in the Arabidopsis genome (Arabidopsis Genome Initiative, 2000), but their functions are largely unknown except a few such as SPINDLY, which is involved in the GA response (Tseng et al., 2001). The U-box domain is a modified ring finger motif and was shown to be the active site of a new class of E3 ubiquitin ligases (Demand et al., 2001; Hatakeyama et al., 2001; Jiang et al., 2001). There are about 63 U-box proteins found in the Arabidopsis genome, yet their functions are unknown (Patterson, 2002). AtCHIP was named for its structural similarity and high sequence homology to animal CHIP proteins. The name CHIP was initially given to a protein that binds to the carboxyl terminus of the mouse (Mus musculus) Hsc70 (Ballinger et al., 1999). Later CHIP homologs were found in other organisms including humans (Homo sapiens), fruitfly (Drosophila melanogaster), and Caenorhabditis elegans (Patterson, 2002). Animal CHIPs are U-box-containing, chaperone-dependent E3 ligases that interact with molecular chaperones (i.e. Hsp70 and Hsp90) and proteasome in the degradation of a selective group of nonnative proteins (Connell et al., 2001; Höhfeld et al., 2001; Murata et al., 2001; Cyr et al., 2002). Animal CHIP proteins play a key role in protein quality control in cells where misfolded polypeptides are prevented from accumulation to toxic level through refolding (via chaperones) or degradation (via proteasome; Cyr et al., 2002).
In the eukaryotic ubiquitylation system, there are only one or two ubiquitin-activating enzymes (E1), a few dozens of ubiquitin-conjugating enzymes (E2), and a large number of ubiquitin ligases (E3). E3 ligases provide the specificity on which proteins to be ubiquitylated and degraded through the ubiquitylation/proteasome pathway (Vierstra and Callis, 1999; Ciechanover et al., 2000). E3 ligases play important roles in many cellular processes such as hormone metabolism and photomorphogenesis in plants (Callis and Vierstra, 2000; Hellmann and Estelle, 2002). They fall into one of the following three classes: HECT domain proteins, RING domain proteins, and U-box domain proteins (Cyr et al., 2002). U-box proteins account for only 3% to 4% of the annotated ubiquitin ligases in humans, which is far less than those found in plants (Patterson, 2002). Although a few U-box-containing proteins from animals were characterized recently, little is known about the cellular functions of most U-box E3 ligases (Patterson, 2002). Because there was no study on plant U-box E3 ligases, AtCHIP appears to be an excellent model for studying the physiological functions of U-box proteins in plants. The high sequence similarity between AtCHIP and animal CHIPs suggests that AtCHIP is an E3 ubiquitin ligase. In this study, we confirmed that AtCHIP has ubiquitin ligase activity by in vitro ubiquitylation experiments. Furthermore, we studied the biological consequences of overexpression of AtCHIP in Arabidopsis. Our results indicate that overexpression of AtCHIP leads to increased sensitivity to both low- and high-temperature conditions, suggesting that expression of AtCHIP is under tight regulation and any disturbances in AtCHIP expression would lead to serious consequences in temperature stress tolerance. Our study highlights the importance of AtCHIP in plant cellular metabolism under stressful conditions.
RESULTS
AtCHIP Is an E3 Ubiquitin Ligase
Sequence comparison of AtCHIP with animal CHIPs from human, mouse, and fruitfly is shown in Figure 1. Like animal CHIPs, AtCHIP contains three TPRs at N-terminal side and a U-box domain at the C-terminal side. They generally share over 54% homology within TPRs and 75% homology within the U-box domains. Recent studies indicated that U-box-containing proteins such as CHIPs are a new class of E3 ubiquitin ligases (Demand et al., 2001; Hatakeyama et al., 2001; Jiang et al., 2001). We tested AtCHIP in an in vitro ubiquitylation experiment and demonstrated that AtCHIP is an E3 ubiquitin ligase.
Sequence comparison of AtCHIP with animal CHIP proteins. The sequences of fruitfly CHIP (dCHIP), human CHIP (hCHIP), and mouse CHIP (mCHIP) were from Ballinger et al. (1999). The three lines with arrows underline TPRs, and the fourth line underlines the U-box domain.
AtCHIP can use ubiquitin, E1, and E2 of animal sources to make ubiquitin polymers very efficiently (Fig. 2).
In vitro experiment demonstrating that AtCHIP is an E3 ligase. Ub, Ubiquitin; E1, ubiquitin-activating enzyme; E2, ubiquitinconjugating enzyme.
AtCHIP Is a Single-Copy Gene, and Its Transcript Is Up-Regulated by Stress Conditions
We conducted a genomic DNA-blot analysis on AtCHIP using a cDNA clone as a probe, and the result indicates that AtCHIP is a single-copy gene in Arabidopsis (Fig. 3). Sequence analysis showed no restriction site for enzymes BamHI and XhoI, one site for EcoRI, and two sites for HindIII in the AtCHIP gene. Restriction digestion of Arabidopsis genomic DNA with these enzymes gave rise to hybridizing bands exactly as expected: one band from BamHI and XhoI, two bands from EcoRI, and three bands from HindIII digestion (Fig. 3). Because many TPR-containing proteins are involved in stress response in organisms that range from yeast, plant, mouse, to human (Nicolet and Craig, 1989; Honore et al., 1992; Torres et al., 1995; Blatch et al., 1997; Dolinski et al., 1998; Marsh et al., 1998), we tested if expression of AtCHIP is affected by stress conditions. Northernblot analysis indicated that transcript levels of AtCHIP increased in response to treatments such as chilling temperature (7°C), heat (38°C), and salts (selenate and NaCl), a pattern similar to that of GST6, a gene that is induced by similar stress conditions in Arabidopsis (Chen et al., 1996; Fig. 4), suggesting that AtCHIP activity may increase under these stress conditions. However, it should be noted that the AtCHIP transcript in control plants was detected only after overnight exposure (>15 h), indicating that the steady-state level of the AtCHIP transcript is not high.
DNA-blot analysis of AtCHIP in Arabidopsis. B, BamHI; E, EcoRI; H, HindIII; X, XhoI. The HindIII digestion of λDNA was used as length markers (at left in kilobases).
RNA-blot analysis on AtCHIP transcript. The genes used as probes are listed on the right side: GST6, encoding glutathione S-transferase 6; UBQ3, a polyubiquitin gene used as an RNA-loading control.
Overexpression of AtCHIP Leads to Increased Sensitivity to Temperature Stress Conditions
To study the physiological functions of AtCHIP in Arabidopsis, we created transgenic plants that overexpress AtCHIP by fusing its full-length cDNA sequence to a strong promoter, the cauliflower mosaic virus 35S promoter in the pBI19-derived vector (Jefferson et al., 1987). Over 80 independent transgenic plants were obtained, and more than one-half of them are single T-DNA insertion lines based on the segregation ratio of kanamycin resistance versus susceptibility in the T1 progeny. The segregation data for 15 single T-DNA insertion lines are given in Table I. Northern-blot analyses were conducted on all single T-DNA insertion lines, and 20 of them were found to express high levels of the introduced transgene transcript. An example of northern blot is shown in Figure 5, in which transgenic lines 1, 2, 4, etc. were considered high-expression lines, whereas lines 3, 5, 13, etc. were considered low-expression lines. The endogenous AtCHIP transcript was not evident in this figure because the exposure time for the blot was only 3 h due to the strength of the cauliflower mosaic virus 35S promoter that drives transgene expression. An overnight exposure would reveal the endogenous AtCHIP transcript that is longer than the transgene transcript because of the 260 nucleotides in the 3′-untranslated region.
Segregation data of selected AtCHIP transgenic plants of the T1 generation on kanamycin plates
kanr, Kanamycin resistant; kans, kanamycin sensitive; P, probability for 3:1 ratio.
RNA-blot analysis of wild-type plants and AtCHIP transgenic plants. The 18S rRNA gene was used as an RNA-loading control. WT, Wild type; lanes 1 to 15, 15 independent transgenic lines.
We analyzed several high and low-expression lines by western blot and found that all high-expression lines contained increased levels of AtCHIP protein, as compared with the low-expression lines and wild-type plants. An example of our western-blot analysis is shown in Figure 6. Plants that overexpress AtCHIP have at least 4- to 5-fold more AtCHIP protein based on the densitometry analysis. All high-expression lines and several low-expression lines selected from the northern-blot analyses were studied further for their growth behaviors under normal, low, and high temperatures. These plants looked healthy with normal seed yield under normal temperature conditions (between 22°C–26°C). However, after they were moved to a growth chamber with a temperature set at 12°C for 42 d or 7°C for 32 d, all high-expression lines demonstrated sensitivity to chilling temperatures: very little growth, whereas the low-expression lines and unrelated transgenic plants demonstrated similar phenotypes as wild-type plants (Figs. 7 and 8). In a higher temperature treatment (34°C for 2 h and 24°C for 22 h for 24 d), these high-expression plants produced much less seeds than wild-type plants (Fig. 9). In fact, most of them did not produce siliques. Even if a few siliques were produced, seed yield was significantly reduced. These experiments were repeated two more times in the next two generations, and similar results were obtained for each of the 20 independent high-expression lines. Based on the northern- and western-blot data, there is a direct causal relationship between the expression of AtCHIP transgene and the cold- or heat-sensitive phenotypes.
Western-blot analysis of wild-type and AtCHIP transgenic plants. A, Wild-type plant; B to E, four independent AtCHIP high-expression lines (nos. 1, 2, 8, and 12 on the RNA blot shown in Fig. 5); F, AtCHIP low-expression line (no. 3 on the RNA blot). GapC, Cytosolic glyceraldehyde-3-phosphate-dehydrogenase as a loading control. The intensities of AtCHIP bands are normalized with the GapC band in each lane in densitometry analysis.
Phenotypes of wild-type and AtCHIP transgenic plants after chilling temperature treatment. Eighteen-day-old Arabidopsis plants were treated with chilling temperature (12°C) for 42 d. Wild-type plants (B) continued to grow and flower, whereas the growth of AtCHIP high-expression plants (A and C, two independent transgenic lines) was greatly inhibited.
Phenotypes of wild-type plants, unrelated transgenic plants, and AtCHIP transgenic plants after chilling temperature treatment. Eighteen-day-old Arabidopsis plants were treated with chilling temperature (7°C) for 32 d. Wild-type plants (A), AtCHIP low-expression plants (E), and unrelated transgenic plants (F) continued to grow, whereas AtCHIP high-expression plants (B–D, three independent lines) grew very little during this time.
Phenotypes of wild-type and AtCHIP high-expression plants after high-temperature treatment. Eighteen-day-old Arabidopsis plants were moved to a growth chamber with a temperature cycle set at 34°C for 2 h and 24°C for 22 h each day for 24 d. The wild-type plant (left) continued to grow and produced normal amount of seeds, whereas the AtCHIP high-expression plant (right) grew but produced few seeds.
The Membrane Damage Is More Evident in AtCHIP Overexpression Plants than Wild-Type Plants after Low-Temperature Treatment
The temperature-sensitive phenotype observed in AtCHIP overexpression plants indicates that there might be cellular damages that adversely affected plant growth and development under temperature stress conditions. To test this possibility, we measured the electrolyte leakage of leaves from both AtCHIP transgenic and wild-type plants after chilling temperature treatment for 2 weeks and found that the electrolyte leakage in AtCHIP high-expression plants is significantly higher than that of wild-type and low-expression plants (Fig. 10), suggesting that AtCHIP overexpression might lead to membrane damage under chilling temperature conditions. However, this damage was not severe enough to cause plants to die under such conditions.
Analysis of membrane damage of wild-type and AtCHIP transgenic plants after cold (7°C) treatment. WT, Wild type; A and B, AtCHIP high-expression lines; C, low-expression line. Data represent mean ± sd, n = 6.
DISCUSSION
Based on the above data, AtCHIP clearly plays an important role in temperature stress tolerance in Arabidopsis. Although AtCHIP transcript is upregulated by several stress conditions, there is no correlation between increased AtCHIP expression and increased temperature stress tolerance. In fact, overexpression of AtCHIP renders plants more sensitive to both low- and high-temperature stresses. An increase in AtCHIP expression alone did not increase stress tolerance, which is not unexpected because both low and high temperatures induce expression of many genes (Thomashow, 1999; Forreiter and Nover, 1998), and the collective effects of those induced genes may be required for higher stress tolerance. However, the decreased stress tolerance in AtCHIP overexpression plants is somewhat surprising because, to our knowledge, we are not aware of a similar example in literature that overexpression of a gene leads to sensitivity to both low and high temperatures. Our data hint a novel mechanism that operates under stress conditions, and AtCHIP may play a key role in this mechanism.
The mode of action of AtCHIP in plant cells may involve other proteins, like what happens with animal CHIP proteins. The function of the mouse CHIP is associated with chaperones Hsp70 and Hsp90 that participate in refolding nonnative proteins or mediating degradation of their substrates through the ubiquitin-proteasome machinery (Höhfeld et al., 2001; Cyr et al., 2002). The folding and degradation are mutually exclusive choices that chaperones would choose in determining the fate of their substrates. CHIP links Hsp70/Hsp90 to the 26S proteasome by interacting with components of the three systems (Ballinger et al., 1999; Connell et al., 2001; Meacham et al., 2001) and plays a key role in the selection of nonnative proteins for degradation (Höhfeld et al., 2001; Cyr et al., 2002). CHIP inhibits the ATPase activity of Hsp70 and its substrate binding and prevents other cochaperones from binding to Hsp90, in turn inhibiting the protein folding activities of both chaperone systems (Ballinger et al., 1999; Connell et al., 2001). CHIP also induces ubiquitylation of the substrates of Hsp70 and Hsp90 and stimulates their degradation through the proteasome (Connell et al., 2001; Demand et al., 2001; Meacham et al., 2001). Therefore, CHIP functions as a degradation factor in protein turnover metabolism.
Under normal conditions, overexpression of AtCHIP may not affect plant growth and development because few damaged proteins are produced in plant cells. However, under temperature stress conditions, overexpression of AtCHIP may lead to quick degradation of some denatured proteins that may be otherwise refolded back to functional proteins by Hsp70 and Hsp90. Repair or removal of damaged proteins by chaperones or proteasome under stress conditions appear to be regulated by cochaperones such as AtCHIP, and both protein refolding and degradation are essential for plant cells to cope with stress conditions. There is apparently a mechanism that maintains the optimal level of AtCHIP expression, and any disturbances in the delicate regulation of chaperone functions by AtCHIP would likely lead to a compromised defense against temperature stress conditions in plants. In humans, the cystic fibrosis transmembrane conductance regulator (CFTR), a plasma-membrane chloride ion channel protein, is regulated by Hsc70 and CHIP (Meacham et al., 2001). If CFTR is not folded correctly, it will be sent to the proteasomal degradation pathway by Hsc70 and CHIP. In this process, CHIP promotes ubiquitylation of CFTR. If AtCHIP regulates some membrane channel proteins like that in animal systems, then overexpression of AtCHIP may remove those damaged but still refoldable proteins under temperature stress conditions, which might affect membrane functions and lead to increased electrolyte leakage in leaf cells.
Based on the above reasoning, we hypothesize that AtCHIP functions with certain chaperone proteins in protein quality control, and AtCHIP's substrates may include membrane channel proteins. However, the reasoning is logical considering the fact that AtCHIP is highly homologous to animal CHIPs, but other possibilities do exist. For example, AtCHIP may interact with proteins other than molecular chaperones or channel proteins, and the increased expression of AtCHIP may disrupt the stoichiometry of AtCHIP and those interacting proteins, which affects certain aspects of cellular metabolism under temperature stress conditions, leading to altered membrane permeability. Identification of AtCHIP's interacting proteins or substrate proteins may reveal how AtCHIP is involved in temperature stress tolerance or how AtCHIP influences protein folding and degradation metabolisms in plant cells. Because AtCHIP is the first plant U-box E3 ubiquitin ligase characterized, further understanding of the molecular mechanism of AtCHIP's involvement in protein ubiquitylation and stress response will contribute significantly to our understanding of cellular metabolism that is central to plant growth and development under normal and stress conditions.
MATERIALS AND METHODS
Isolation of AtCHIP
AtCHIP was identified as one of the GF14λ-interacting proteins in a yeast two-hybrid screening in which the LexA/GF14λ.33-194 (residues 33–194 of GF14λ) was used as the bait (the LexA portion of the bait is only the DNA-binding domain of the original LexA protein; Golemis et al., 1996). Because AtCHIP interacts weakly with GF14λ in yeast cells, their interaction was not studied further at this time. The gene coding for AtCHIP was also found independently by others on a bacterial artificial chromosome clone from chromosome III (Arabidopsis Genome Initiative, 2000). The GenBank accession number for this gene is AAK68747, and the MIPS number is At3g07370.
Expression of AtCHIP in Bacterial Cells and Analysis of AtCHIP Activity
AtCHIP was expressed in Escherichia coli by using the pET system (Novagen, Madison, WI), and then AtCHIP was purified according to the manufacturer's protocol. Polyclonal antibodies were raised against purified AtCHIP at Animal Pharm Services, Inc. (Healdsburg, CA). In vitro ubiquitylation reaction with purified AtCHIP as the E3 ligase was carried out according to the condition described by Jiang et al. (2001). The reaction mixture included rabbit E1 (Calbiochem-Novabiochem, San Diego), human (Homo sapiens) E2 (UbcH5a), and bovine ubiquitin (Sigma, St. Louis). Antiubiquitin antibodies (Sigma) were used to detect ubiquitin polymers in the western blot of the in vitro ubiquitylation assay (Fig. 2).
Construction of Transforming Vector and Arabidopsis Transformation
A full-length AtCHIP fragment was amplified from a cDNA library with PCR using primers 5′-ATCGGATTCATGGTTACAGGCGTGGCTTCC-3′ (forward) and 5′-CATGAGCTCTCAACAACCCATCTTGTAAGCC-3′ (reverse) and subcloned into the vector pBI121 (Jefferson et al., 1987) by replacing the β-glucuronidase gene with restriction enzymes BamHI and SstI to form the transforming vector. The vector was then introduced into the Agrobacterium tumefaciens GV3101, which was used to transform wild-type Arabidopsis (ecotype C24) according to the protocol of Clough and Bent (1998).
DNA and RNA Manipulation
Genomic DNA of wild-type Arabidopsis (ecotype C24) was prepared according to the method of Dellaporta et al. (1983), digested with restriction enzymes, separated by electrophoresis (5 μg lane–1), blotted to a nylon membrane, and hybridized with an AtCHIP cDNA probe. Total RNAs were isolated from Arabidopsis plants using the TRIzol reagent (Invitrogen, Carlsbad, CA), separated by electrophoresis (10 μg lane–1), blotted to a nylon membrane, and hybridized with various probes. Hybridization was carried out according to the method of Church and Gilbert (1984) using probes labeled by random priming. The washing conditions were as follows: two times (10 min each) in 0.5% (w/v) bovine serum albumin, 1 mm EDTA, 40 mm Na2HPO4 (pH 7.2), and 5.0% (w/v) SDS at 63°C; then four times (5 min each) in 1 mm EDTA, 40 mm Na2HPO4 (pH 7.2), and 1% (w/v) SDS at 63°C. The same filter in RNA blot was used for hybridizations with probes AtCHIP (cDNA clone), GST6 (full length coding sequence), and UBQ3 (cDNA clone) or 18S rRNA (cDNA clone), consecutively. The condition for stripping filter was as follows: twice (15 min each) in 2 mm Tris (pH 8.2), 2 mm EDTA (pH 8.0), and 0.1% (w/v) SDS.
Immunoblot Analysis
Leaf proteins were extracted by grinding three mature leaves in a mortar in extraction buffer (50 mm Na2HPO4 [pH 7.0] and 1 mm EDTA). The crude extracts were centrifuged in a microfuge at 14,000 rpm for 10 min, and the supernatants, which contain mainly cytosolic proteins, were added to an equal volume of 2× SDS loading buffer (125 mm Tris-Cl, 2% [w/v] SDS, 20% [v/v] glycerol, 200 mm dithiothreitol, and 0.01% [w/v] bromphenol blue [pH 6.8]). Protein concentration in the extraction buffer was determined by the Bradford (1976) method using bovine serum albumin as a standard. Proteins from AtCHIP transgenic plants and wild-type plants were subjected to electrophoresis in a 12% (w/v) SDS polyacrylamide gel. The conditions for blotting and color development were the same as described previously (Yan et al., 2002). The densitometry analysis of the western-blot data was conducted by using the NIH Image software (version 1.57, U.S. National Institutes of Health, Bethesda, MD).
Cold, Heat, and Salt Treatments
Arabidopsis seeds of wild-type plants, unrelated transgenic plants, and homozygous AtCHIP transgenic plants of T4 generation were planted into soil mix and cold treated at 4°C for 6 d, then moved to room temperature under continuous white fluorescent light. For cold treatment, 18-d-old plants were moved to refrigerators that were set at 12°C and 7°C, respectively. Plants in the refrigerators were provided with white fluorescent light (50 μE m2 s–1) and allowed to grow for 5 to 6 weeks. For heat treatment, 18-d-old plants were moved to a growth chamber with a temperature cycle set at 34°C for 2 h and 24°C for 22 h each day for 3 to 4 weeks. The light intensity in the growth chamber was around 60 μE m2 s–1. For northern-blot analysis shown in Figure 4, total RNAs were isolated from 19-d-old wild-type plants that were treated with 200 mm NaCl, 100 μm Na2SeO4, 7°C (cold), and 38°C (heat) for 24 h, respectively.
Electrolyte Leakage Assay
Four-week-old wide-type and AtCHIP transgenic plants were moved to a refrigerator that was set at 7°C and provided with white fluorescent light (50 μE m2 s–1). After 2 weeks in the refrigerator, two leaves were cut from each plant and rinsed briefly in distilled water, then placed immediately in a tube of 4 mL of distilled water. The electrolyte content was measured after the tubes were agitated gently for 3 h by using an Orion 120 conductivity meter (Expotech USA Incorporated, Houston, TX). After the solutions and leaves were autoclaved, the electrolyte content was measured again. The ratio of electrolyte content before and after autoclave was used as an indicator for membrane damage after cold treatment. Six repetitions for each treatment were conducted.
Acknowledgments
We thank Dr. Karam Singh for providing the genomic GST6 clone, Dr. Ming-che Shih for providing the antibodies against cytosolic glyceraldehyde-3-phosphate-dehydrogenase, and the Arabidopsis Biological Resource Center (Ohio State University, Columbus) for proving the polyubiquitin gene UBQ3 (ATTS0348). We thank Cixin He for reading the manuscript.
Footnotes
Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.103.020800.
↵1 This work was supported by grants from the Texas Advanced Technology Program.
↵2 Present address: Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, TX 77030.
- Received January 21, 2003.
- Revised March 4, 2003.
- Accepted March 24, 2003.
- Published May 1, 2003.
LITERATURE CITED
