DDB1A and DET1 Exhibit Complex Interactions During Arabidopsis … · 2007-07-25 · Wesam M. Al Khateeb and Dana F. Schroeder1 Department of Botany, University of Manitoba, Winnipeg,

Copyright � 2007 by the Genetics Society of AmericaDOI: 10.1534/genetics.107.070359

DDB2, DDB1A and DET1 Exhibit Complex Interactions DuringArabidopsis Development

Wesam M. Al Khateeb and Dana F. Schroeder1

Department of Botany, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada

Manuscript received January 2, 2007Accepted for publication February 28, 2007

ABSTRACT

Damaged DNA-binding proteins 1 and 2 (DDB1 and DDB2) are subunits of the damaged DNA-bindingprotein complex (DDB). DDB1 is also found in the same complex as DE-ETIOLATED 1 (DET1), a nega-tive regulator of light-mediated responses in plants. Arabidopsis has two DDB1 homologs, DDB1A andDDB1B. ddb1a single mutants have no visible phenotype while ddb1b mutants are lethal. We have identifieda partial loss-of-function allele of DDB2. To understand the genetic interaction among DDB2, DDB1A, andDET1 during Arabidopsis light signaling, we generated single, double, and triple mutants. det1 ddb2 partiallyenhances the short hypocotyl and suppresses the high anthocyanin content of dark-grown det1 andsuppresses the low chlorophyll content, early flowering time (days), and small rosette diameter of light-growndet1. No significant differences were observed between det1 ddb1a and det1 ddb1a ddb2 in rosette diameter,dark hypocotyl length, and anthocyanin content, suggesting that these are DDB1A-dependent phenotypes.In contrast, det1 ddb1a ddb2 showed higher chlorophyll content and later flowering time than det1 ddb1a,indicating that these are DDB1A-independent phenotypes. We propose that the DDB1A-dependent pheno-types indicate a competition between DDB2- and DET1-containing complexes for available DDB1A, while,for DDB1A-independent phenotypes, DDB1B is able to fulfill this role.

PLANT development is dependent on environmen-tal conditions. Because light is the energy source

for plant growth, plants have evolved highly sensitivemechanisms for perceiving light. This information isused to regulate development and to maximize lightutilization for photosynthesis. The transition from thevegetative to the reproductive stage is also regulatedby light. Seedlings implement different developmentalprograms when grown in light or darkness. Light-grownseedlings undergo photomorphogenesis, exhibiting shorthypocotyls, open and expanded cotyledons, and pho-tosynthetically active chloroplasts. In contrast, seedlingsgrownunderdarkconditionsareetiolated,havingclosedand unexpanded cotyledons, elongated hypocotyls, andundeveloped chloroplasts. This developmental pattern isknown as skotomorphogenesis (Chen et al. 2004).

This developmental switch (etiolation/de-etiolation)is under the control of at least 10 genes (COP/DET/FUS). Molecular genetic studies in Arabidopsis indicatethat these proteins function downstream of the photo-receptors to repress photomorphogenesis in the ab-sence of light. Mutation of these genes results in seedlingswith a de-etiolated (det) or constitutive photomorpho-genic (cop) phenotype when grown under dark con-ditions. The null mutations of these genes are seedling

lethal with high anthocyanin levels ( fus) (Wang andDeng 2002). COP1 is a WD-40 and RING finger proteinwith E3 ubiquitin (Ub) ligase activity, which targets pho-toreceptors and downstream transcription factors forubiquitination and subsequent degradation. The COP9signalosome (CSN) is a multiprotein complex composedof eight subunits, which associates with and supports theactivity of multiple cullin-containing E3 ubiquitin ligasecomplexes. In Arabidopsis, mutants in CSN componentsalso exhibit the constitutive photomorphogenic/de-etiolated/fusca(cop/det/fus)phenotype(Schwechheimer

et al. 2004).De-etiolated 1 (det1-1) partial loss-of-function mutants

exhibit short hypocotyls, open cotyledons, and high an-thocyanin levels in the dark (Chory et al. 1989). Underlight conditions, det1-1 seedlings are smaller and palerthan wild type. In addition, they show reduced apicaldominance, day-length-insensitive flowering (Pepper

and Chory 1997) and defects in germination, expres-sion of light-regulated genes, and chloroplast develop-ment (Chory and Peto 1990). Approximately 1000 genesare either up- or downregulated in det1-1 compared towild type (Schroeder et al. 2002).

DET1 is a nuclear protein (Pepper et al. 1994) presentin a complex with an approximate mass of 350 kDa. Intobacco BY2 cells, this complex includes the plant ho-molog of UV-damaged DNA-binding protein 1 (DDB1).In Arabidopsis, two homologs of DDB1 have beenfound: DDB1A and DDB1B, which are 91% identical

1Corresponding author: Department of Botany, University of Manitoba,Winnipeg, MB R3T 2N2, Canada. E-mail: [email protected]

Genetics 176: 231–242 (May 2007)

at the amino acid level. Arabidopsis DDB1A matchestobacco DDB1 more closely than Arabidopsis DDB1B(Schroeder et al. 2002). DDB1A is expressed at almosttwofold higher levels than DDB1B throughout the Ara-bidopsis life cycle in all organs studied (Figure 1A).ddb1a and ddb1b mutants in Arabidopsis were studiedusing T-DNA insertions. ddb1a mutants show no obviousphenotype in a wild-type background, but mutation ofDDB1A in the det1 background enhanced det1 pheno-types. In contrast to ddb1a, ddb1b mutants are lethal, sug-gesting a crucial role for DDB1B during Arabidopsisdevelopment. DDB1 is evolutionarily conserved as Ara-bidopsis DDB1A is 83 and 46% identical at the aminoacid level with rice and human DDB1, respectively(Schroeder et al. 2002).

The Arabidopsis DDB1 protein is homologous tohuman DDB1 (127 kDa), a component, along withDDB2 (48 kDa), of the damaged DNA-binding proteincomplex (DDB). DDB1 is present at higher levels thanDDB2 in human cells (Liu et al. 2000). DDB1 is presentin the cytoplasm, but upon UV irradiation, translocatesto the nucleus. Loss of DDB2 function prevents accu-mulation of DDB1 in the nucleus (Shiyanov et al.1999), whereas loss of DDB1 function has no effect onbinding activity of DDB2 to the damaged DNA ( J. Li

et al. 2006). The suggested role of the DDB complex is torecognize DNA lesions, initiating nucleotide excisionrepair. In humans, the rare inherited disease Xerodermapigmentosa group E (XPE) results from mutation of DDB2.XPE patients display an increased skin sensitivity to UVlight and are at a highly elevated risk of developing UV-induced skin cancer (Cleaver 2005).

Recently, human DDB1 and DDB2 were found to becomponents of an E3 ubiquitin ligase. DDB1, along withCUL4 and ROC1, is a component of several types of E3ligases, including one with DDB2 and another with thetranscriptional coupled repair factor CSA. Both theseE3 ligase complexes are regulated by the COP9 signal-osome (Groisman et al. 2003). Subsequently, DDB1-CUL4complexes were found to interact with many WD40-repeatproteins and use them as adaptors in recognizing sub-strates for proteolysis (Angers et al. 2006; He et al. 2006;Higa et al. 2006; Jin et al. 2006). Human DDB1 is alsofound in a complex with CUL4, ROC1, DET1, and COP1(Wertz et al. 2004). Arabidopsis DDB1A and DET1 copurifywith the E2 Ub conjugase variant COP10 (Yanagawa

et al. 2004) and these proteins have recently been foundto form a complex with AtCUL4 and RBX1 (ROC1)(Bernhardt et al. 2006; Chen et al. 2006). Thus DDB1appears to be a central component of CUL4 E3 ligases.

The DDB complex is present not only in humans, butalso in other organisms such as rice (Oryza sativa cv.Nipponbare) (Ishibashi et al. 2003). Arabidopsis DDB2shows 59 and 30% identity with rice and human DDB2,respectively. In this study we examined the role ofArabidopsis DDB2 and its interaction with DDB1A andDET1 in light signaling. All combinations of double and

triple mutants were generated to understand the geneticinteraction between these genes. Plants were grown andanalyzed at different developmental stages. Comparisonbetween the mutants revealed complex interactions be-tween these genes. In some cases, the modulation ofdet1 phenotypes by ddb2 was DDB1A dependent; in othercases, it was DDB1A independent. We interpret theseresults as consistent with a model whereby separateDET1- and DDB2-containing complexes compete forDDB1A, in the case of the dependent phenotypes, andwhere DDB1B is able to fill this role, in the case ofDDB1A-independent phenotypes.

MATERIALS AND METHODS

Plant materials and growth conditions: All mutations usedin this experiment were in the Col background of Arabidopsisthaliana. The DET1 partial loss-of-function allele det1-1 andthe ddb1a T-DNA mutant were described previously (Pepper

et al. 1994; Schroeder et al. 2002), and the ddb2 allele (SALK_040408) was obtained from the Arabidopsis Stock Center(http://www.arabidopsis.org/). cop1-4 was kindly provided by X.W. Deng of Yale University. Seedlings were grown in a growthchamber at 20� and 50% relative humidity. Light was providedby fluorescent bulbs (100 mmol photons m�2 sec�1). Plants weregrown in Sunshine mix number 1 (SunGro, Bellevue, WA). Short-day conditions corresponded to 10 hr light and 14 hr dark;long-day conditions consisted of 16 hr light and 8 hr dark.

Construction of double and triple mutants: Double mutants:The det1 ddb1a mutant was generated as described inSchroeder et al. (2002); cop1-4 ddb2, det1 ddb2, and ddb1addb2 double mutants were derived from genetic crosses oftheir corresponding single-parental mutants. Because allmutations analyzed were recessive, double homozygous plantswere identified in the F2 generation, where they occur in aratio close to 1:15. Putative double mutants in the F2 gen-eration were selected on the basis of mutant phenotypes andPCR genotyping. For example, for the ddb2 3 det1 cross,�100F2 seeds were plated—segregating det1 homozygotes identifiedby their dwarf stature—transplanted, and genotyped for ddb2.In the F3 generation, several independent det1 ddb2 double-mutant lines consistently exhibited shorter hypocotyls anddecreased anthocyanin in dark-grown seedlings and increasedchlorophyll in light-grown seedlings relative to their det1DDB2/� siblings (data not shown).

Triple mutant: Pollen from plants homozygous for ddb2 wasused to fertilize flowers of det1 ddb1a plants. As expected, all F1

plants showed wild-type phenotype. PCR was used to confirmthe presence of the ddb2 insertion. Plants heterozygous forddb2 and det1 ddb1a were selfed to produced F2 [DET1 andDDB1A are linked�10 cM apart (Schroeder et al. 2002)]. det1ddb1a homozygotes were identified as extreme dwarfs andwere then genotyped to identify ddb2 homozygotes. Due to theinfertility of this triple mutant, stocks are maintained as a seg-regating population homozygous for ddb2 and heterozygousfor det1 ddb1a.

PCR reactions were conducted to confirm the ddb1a andddb2 insertions. DNA was extracted according to Weigel andGlazebrook (2002). ddb1a insertions were detected usingLB2 (59-TTG GGT GAT GGT TCA CGTAGT GGG CCATCG-39)and UV1.17 (59-ACT GGG CTC AAC TAG AAA ATA TGG AACAA-39) while UV1.17 and UV1.1 (59-GTC TTG ACT GTG CATTTC AGA GTG CTT AT-39) were used to detect the wild-typeDDB1A. For ddb2, LB2 and DDB2.1 (59-TTG GGT GAT GGTTCA CGT AGT GGG CCA TCG-39) were used, while DDB2.1

232 W. M. Al Khateeb and D. F. Schroeder

and DDB2.3 (59-ACG ACG TGT TTT GTC GGT GTG GAAGAA-39) were used for wild-type DDB2.

RNA extraction and RT–PCR: Total RNA was extractedfrom 7-day-old seedlings using the RNeasy plant mini kit(QIAGEN, Valencia, CA) according to the manufacturer’sinstructions. Quality and quantity of isolated RNA was checkedby denaturing gel electrophoresis and by spectrophotometricanalysis. cDNA synthesis and PCR amplification were per-formed in the same reaction using Access RT–PCR kit (Promega,Madison, WI) according to the manufacturer’s instructions.DDB2-specific primers in exon 2 (59-ACAGCCTGGCCATGAAGCTGGA-39) and in exon 6 (59-CCTGCCATCCATCAGGGTTGAG-39) were used. cDNA synthesis was performed at 45� for45 min, followed by PCR [5 min at 94�, 30 times (30 sec at 94�,50 sec at 67�, 2 min at 72�), 2 min at 72�]. To detect relativedifferences in transcript levels, amplification was performedwhen the PCR product was accumulating exponentially withrespect to cycle number (30 cycles). UBQ10 was used as aninternal control (59-GATCTTTGCCGGAAAACAATTGGAGGATGGT-39 and 59-CGACTTGTCATTAGAAAGAAAGAGATAACAGG-39) [5 min at 94�, 22 times (1 min at 94�, 1 min at 64�,1 min at 72�), 2 min at 72�]. PCR products were separatedon 1% (w/v) agarose gels, and the intensities of ethidium-bromide-stained bands were determined using ImageJ soft-ware (1.36b National Institutes of Health).

Hypocotyl elongation assay: Seeds were plated on Mura-shige and Skoog (MS) media (13 MS salts, 0.8% phytoagar,2% sucrose) after sterilization and stratified at 4� for 2 days.Plates were transferred to a growth chamber and grown undereither light (16 hr light 80 mmol m�2 sec�1) or dark conditions(after 6 hr light as germination enhancer). Seven days later,hypocotyl length was measured for at least 10 seedlings usingNational Institutes of Health image software.

Pigment analysis: For chlorophyll measurement, 7-day-oldseedlings were extracted with 80% acetone overnight, the A645

and A663 were determined in a spectrophotometer (model2100 pro, Ultrospec), and chlorophyll content was calculatedaccording to the method of Mackinney (1941). Anthocyaninwas determined using standard technique (Fankhauser andCasal 2004). Experiments were repeated at least three timeswith four replicates per line in each experiment. Each rep-licate contained at least 50 mg of plant tissue.

Growth parameter measurements: Seedlings were trans-planted from plates to soil at 14 days old. For each line, 18plants were used for growth parameter analysis. Floweringtime was scored for each plant as the number of days until thefirst bud became visible. Also, the total number of rosette andcauline leaves on the main inflorescence was counted. Plantswere moved regularly to random positions in the growthchamber to prevent any positional effects on plant growth.The following traits were also recorded: shoot fresh weight (4weeks old), rosette diameter (4 weeks old), and total numberof inflorescences and height (6 weeks old).

Statistical analysis: All experiments were repeated at leastthree times. Results presented are means with 95% confidenceintervals of 10–18 replicates. Means were compared byStudent’s t-test. Probabilities of 0.05 or less were consideredto be statistically significant.

RESULTS

To examine the role of DDB2 in light signaling andinteraction with DDB1A and DET1, we studied the physi-ological role of these proteins at different stages of Ara-bidopsis development. Gene expression data available atthe Genevestigator website (https://www.genevestigator.

ethz.ch) show that DDB2 is expressed in all Arabidopsisorgans, as are DDB1A, DDB1B, and DET1 (Figure 1A).A T-DNA insertion in DDB2 (SALK_040408) was ob-tained from the Arabidopsis Stock Center (http://www.arabidopsis.org/). This insertion is located 180 bp up-stream from the DDB2 ATG (Figure 1B). PCR genotyp-ing was performed to confirm this insertion (data notshown). RT–PCR analysis showed that DDB2 expressionwas reduced by approximately twofold in ddb2 mutantscompared to the wild type (Figure 1, C and D). Loss offunction of DDB2 in other mutant backgrounds resultsin similar relative levels of expression (Figure 1, C andD). These data suggest that this T-DNA insertion resultsin partial loss of function of DDB2. This result is incontrast to Koga et al. (2006) who observed completeabsence of DDB2 expression in the same T-DNA in-sertion line (SALK_040408). This variation in transcriptlevel may be due to cosuppression.

For all the growth parameters measured in the follow-ing sections, no significant differences were observedamong wild type, the single mutants ddb2 and ddb1a, andthe double mutant ddb1a ddb2. Since we had previouslyshown that ddb1a exhibited no phenotype on its own,yet enhanced the phenotypes of the DET1 partial loss-of-function allele det1-1, we generated the det1-1 ddb2double mutant. The triple mutant was generated todetermine if det1-1 ddb2 phenotypes were DDB1A de-pendent or independent.

Dark-grown seedlings: Hypocotyl elongation: In theabsence of light, wild-type plants show increased hypo-cotyl elongation, but det1 mutants lack this response andhave short hypocotyls. Lines carrying mutations in DET1,DDB1A, and DDB2 singly or in combination were grownunder dark conditions for 7 days to study the role ofthese proteins in de-etiolation and light signaling (Fig-ure 2). Plants with partial loss of function of DDB2 inthe wild-type or ddb1a background did not differ sig-nificantly from wild type. In contrast, mutation of DDB2in the det1 background resulted in enhancement of thedet1 short hypocotyl phenotype (P # 0.001), resulting in31% shorter hypocotyls. However, the triple mutant det1ddb1a ddb2 did not differ significantly from the doublemutant det1 ddb1a. Thus enhancement of the det1 shorthypocotyl phenotype by ddb2 is a DDB1A-dependentphenotype.

Anthocyanin content: We studied anthocyanin accu-mulation in seedlings, which ordinarily does not accu-mulate when plants are grown under dark conditions.Very low levels of anthocyanin were detected in Col-0,ddb2, ddb1a, and ddb1a ddb2 seedlings (Figure 2C). det1showed higher levels of anthocyanin than the wild type(18-fold higher). The ddb2 mutation in the det1 back-ground suppressed this increase in anthocyanin con-tent, showing only 28% of the levels observed in the det1single mutant. In contrast to ddb2 partial loss of functionin the det1 background, ddb1a loss of function in thesame background showed an increase in anthocyanin

DDB2, DDB1A and DET1 Interactions 233

content, with 6.5-fold enhancement. The anthocyanincontent of the triple mutant det1 ddb1a ddb2 exhibitedno significant difference from the double mutant det1ddb1a. Thus, ddb2 exhibits DDB1A-dependent suppres-sion of det1 dark anthocyanin accumulation.

Light-grown seedlings: Hypocotyl elongation: In light,det1 seedlings are also shorter than wild type. Growth of wildtype and ddb2, ddb1a, and ddb1a ddb2 mutants under long-day conditions for 7 days showed no significant differences(Figure 3, A and B). Loss of function of DDB2 in the det1background did not affect the det1 short hypocotyl. Incontrast, mutation of DDB1A in the same background

reduced hypocotyl elongation (P # 0.05). No significantdifferences were observed between det1 ddb1a and det1ddb1a ddb2. Thus ddb2 has no effect on this phenotype.

Anthocyanin content: When grown under light con-ditions, seedlings accumulate anthocyanin. Col-0, ddb2,ddb1a, and ddb1a ddb2 had similar levels while det1showed more than a 60% increase compared to wildtype (Figure 3, A and C). While no significant differencewas observed between det1 and det1 ddb2, the increasedanthocyanin content of det1 was dramatically enhancedin det1 ddb1a. However, ddb2 significantly suppressed(P # 0.03) anthocyanin accumulation in the det1 ddb1a

Figure 1.—(A) Gene expression data as obtained from Genevestigator database (https://www.genevestigator.ethz.ch). (B) Sche-matic of 59-end of the DDB2 gene (At5g58760). Transcription start sites (11) are based on the GenBank accession nos. BX832566and AK175124. Exons are shown as boxes and introns and upstream sequences as lines. Position of the T-DNA insertion(SALK_040408) is indicated, along with primers used for genotyping. (C) RT–PCR products for DDB2 and UBQ10 of 7-day-old seedlings grown under long-day conditions. (D) Quantification of DDB2 transcript level (DDB2 values normalized toUBQ10 levels, relative to Col-0). Data are shown as the means 6 SE of three different technical repeats. Numbers above errorbars indicate expression relative to DDB2 wild-type control.


double mutant as levels in the triple mutant were only76% that of the double mutant. Thus, ddb2 partiallysuppresses anthocyanin accumulation in light-growndet1 ddb1a seedlings but not in the det1 single mutant.

Chlorophyll content: In light, det1 mutants accumulateless chlorophyll than wild type. Wild type, ddb2, ddb1a,and the double mutant ddb1a ddb2 all showed the simi-lar levels of chlorophyll accumulation (Figure 3D). det1showed a decrease compared to the previous genotypeswith only 34% of wild-type chlorophyll content. Lossof function of DDB2 in the det1 background partiallysuppressed the pale color of det1 and resulted in 60%more chlorophyll in the double mutant. In contrast, lossof function of DDB1A in the det1 background showed achlorophyll content similar to that of det1 alone. Similarto the effect of the ddb2 mutation on the det1 back-

ground, loss of function of DDB2 in the det1 ddb1abackground resulted in twofold more chlorophyll accu-mulation in the triple mutant det1 ddb1a ddb2. Thissuggests that suppression of the det1 pale phenotype byddb2 is DDB1A independent.

Adult plants: Rosette diameter: Adult det1 are alsosmaller than wild type, so after 1 month of growth undereither long-day or short-day conditions, rosette diame-ter was measured for all genotypes. Analysis of rosettediameter for different genotypes showed similar trendswhen grown under either long- or short-day conditions.In general, plants grown under long-day conditionsshowed larger rosette diameter than those grown undershort-day conditions. ddb2, ddb1a, and ddb1a ddb2 loss-of-function mutants did not differ significantly from wildtype (Figure 4, A and B). This result is in contrast toKoga et al. (2006), who found that DDB2 mutationresulted in reduction in leaf length and width. Ourweaker phenotype is consistent with the fact that we stillobserve�50% of wild-type DDB2 expression in our ddb2mutants while Koga et al. (2006) used an RNA null. det1showed smaller rosette diameter than wild type (Figure4, A and B). The ddb2 mutation partially suppressed thedet1 small rosette diameter in both long- and short-dayconditions (P # 0.01). In contrast, the ddb1a mutationsignificantly enhanced this phenotype in the det1 back-ground, resulting in smaller rosette diameter. Mutationof ddb2 in the det1 ddb1a background has no significanteffect under both conditions. Therefore, the partial sup-pression of the det1 small rosette diameter by ddb2 isDDB1A dependent.

Flowering time: To examine the role and interaction ofthese proteins in controlling Arabidopsis flowering time,we compared wild type, single mutants, double mutants,and the triple mutant grown in short and long days.Arabidopsis is a facultative long-day plant, so floweringin wild type is accelerated in long day. We found thatwild type, ddb2, ddb1a, and ddb1a ddb2 start floweringafter 26 and 62 days on average in long- and short-dayconditions, respectively (Figure 4C). Loss of functionof DET1 resulted in early flowering time compared tothe wild type (18 or 22 days, respectively). The doublemutant det1 ddb2 showed significant suppression (P #

0.01) of det1 early flowering time (20 and 27 days, re-spectively). No significant difference was observed be-tween det1 ddb2 and det1 ddb1a in both photoperiods.The triple mutant det1 ddb1a ddb2 showed significant(P # 0.01) suppression of early det1 ddb1a floweringtime, flowering at 24 and 32 days, respectively. Thussuppression of det1 early flowering time (in days) by ddb2is a DDB1A-independent phenotype.

In addition, flowering time was measured by countingthe number of rosette leaves plus cauline leaves on themain inflorescence. ddb2, ddb1a, and ddb1a ddb2 had thesame number of leaves as wild type at flowering time.ddb2 suppressed significantly (P # 0.01) the earlyflowering time of det1 in terms of number of leaves at

Figure 2.—Phenotypic analysis of 7-day-old dark-grownseedlings. (A) From left to right: Col-0, ddb2, ddb1a, ddb1addb2, det1, det1 ddb2, det1 ddb1a, and det1 ddb1a ddb2. (B) Hy-pocotyl length. (C) Anthocyanin content (A530–A657/g freshweight). Error bars indicate 95% C.I. **P # 0.01 relative toDDB2 wild-type control.


flowering under long-day conditions but not undershort-day conditions (data not shown). No significantdifferences were observed between the double mutantdet1 ddb1a and the triple mutant det1 ddb1a ddb2 in eithercondition. Therefore, suppression of det1 early flower-ing time (as measured by number of leaves) by ddb2 isa DDB1A-dependent phenotype under long-day condi-tions but has no effect under short-day conditions.

Height: After full adult height had been achievedunder long- or short-day conditions, plant height wasmeasured from the soil surface to the last flower on theinflorescence. ddb2, ddb1a, and ddb1a ddb2 did not differfrom wild type (Figure 4, A and D). det1 mutants are 35%shorter than wild type. No significant difference wasobserved between det1 and det1 ddb2. In contrast, det1ddb1a significantly enhanced the short det1 phenotypeunder both photoperiods. Loss of function of DDB2 inthe det1 ddb1a background resulted in further enhance-ment of the short phenotype (P # 0.01). det1 ddb1a ddb2plants showing only 50% of det1 ddb1a height underlong-day and 21% under short-day conditions. Thus

ddb2 enhances the dwarf phenotype of the doublemutant det1 ddb1a.

Fertility assessment: The observation that the doublemutant det1 ddb1a produced few and very small seed-containing siliques encouraged us to look thoroughlyat floral development in the different genotypes. Long-day-grown wild type, ddb2, ddb1a, and ddb1a ddb2 did nothave significantly different silique lengths (Figure 5).det1 exhibited shorter silique length, but mutation ofddb2 in this background partially suppressed the shortsilique phenotype (P # 0.01). The double mutant det1ddb1a showed very short siliques (4.75 mm). Loss offunction of DDB2 in the det1 ddb1a background en-hanced this short silique phenotype and resulted in1.63-mm siliques on average. After dissecting at least sixsiliques from each genotype, we counted the numberof seeds in each half. No significant differenceswere observed among wild type, ddb2, ddb1a, and ddb1addb2, with 29–33 seeds in each half-silique (Figure 5B).The det1 mutant showed lower seed numbers comparedto the previous genotypes. Again, the ddb2 mutation

Figure 3.—Phenotypic analysis of 7-day-old long-day-grown seedlings. (A) From left to right: Col-0, ddb2, ddb1a, ddb1a ddb2, det1,det1 ddb2, and det1 ddb1a, det1 ddb1a ddb2. (B) Hypocotyl length. (C) Anthocyanin content. (D) Chlorophyll content (microgramsof chlorophyll/milligram fresh weight). Error bars indicate 95% C.I. *P # 0.05 or **P # 0.01, respectively, relative to DDB2 wild-type control.


in the det1 background suppressed this phenotypeand resulted in more seeds. The det1 ddb1a siliquesshowed a lower seed number (9.17 seeds on averageper half-silique). On the other hand, det1 ddb1a ddb2siliques were found to be pale in color and dry; dis-secting these siliques showed that none of them haddeveloped seeds.

These observations led us to examine flowers of thesegenotypes. Early flowers of wild type, ddb2, ddb1a, andddb1a ddb2 showed long petals and long stamens that arevery close to the stigma in terms of height (Figure 5C).det1 flowers showed slightly shorter stamens than the wildtype or the other single mutants. det1 ddb2 flowers werefound to have longer stamens than det1, relatively similarto wild type. So ddb2 mutation in the det1 backgroundsuppressed the short stamens and subsequently facili-tated the fertilization process. In contrast to det1 ddb2,mutation of DDB1A in the det1 background reduced

stamen elongation. Dissecting det1 ddb1a flowers showedthat the stamens exhibit approximately one-third thestigma length, as well as short petals. In later flowers, det1ddb1a stamens were found to be longer and closer to thestigma, resulting in partial fertility. det1 ddb1a ddb2 flowersshowed reduced size and petal length and short stamensthroughout their life cycle. Thus, ddb2 suppresses therelated det1 phenotypes of short silique length, reducednumber of seeds, and short stamens. This suppression isnot observed in the absence of DDB1A; in fact, thesephenotypes are enhanced in the triple mutant.

Other growth traits were also assessed, such asnumber of inflorescences produced by each plant andfresh weight at 1 month of age under long- or short-dayconditions. For these traits, loss of function of DDB2had no effect in any background (data not shown).

Interactions with other de-etiolated mutants: Todetermine if ddb2 interacts with det1 specifically or

Figure 4.—Adult phenotypes (A) From left to right: Col-0, ddb2, ddb1a, ddb1a ddb2, det1, det1 ddb2, det1 ddb1a, and det1 ddb1addb2 grown in long-day conditions. (Top) Rosette diameter at 4 weeks. (Bottom) Height at 6 weeks. (B) Rosette diameter of 4-week-old plants. (C) Flowering time (in days). (D) Height (in centimeters) of adult plants. Error bars indicate 95% C.I. *P # 0.05or **P # 0.01, respectively, relative to DDB2 wild-type control.


nonspecifically modifies the de-etiolated phenotype, wegenerated the cop1-4 ddb2 line. cop1-4 is a partial loss-of-function allele that is phenotypically similar to det1-1,exhibiting short hypocotyls when grown in the dark.However, loss of function of DDB2 in the cop1-4background resulted in no significant effect on hypo-cotyl length under dark conditions (Figure 6, A and B)or on anthocyanin content at the seedling stage undereither light or dark conditions (data not shown).Analysis of adult plants showed no significant differ-ences between cop1-4 and cop1-4 ddb2 in flowering time,rosette diameter, height, number of inflorescences, andsilique length (data not shown). Thus ddb2 appears tospecifically modify det1, rather than the de-etiolatedphenotype in general.

DISCUSSION

Arabidopsis DDB1A has been found in a complex withDET1 (Schroeder et al. 2002) and COP10 (Yanagawa

et al. 2004). Also, DDB1 has been found in anothercomplex with DDB2 involved in DNA repair (Cleaver

2005). Because DDB1 is involved in the formation of bothcomplexes, we were interested in studying the interac-tion between DET1 and DDB2 through DDB1. To testthis hypothesis, we generated the double and triple mu-tants of these genes. Our analysis indicates several modesof interaction among DDB2, DDB1A, and DET1.

Suggested models of interaction: DDB1A dependent:For some phenotypes, loss of function of DDB2 in thedet1 background resulted in significant changes, whileloss of function of DDB2 in the det1 ddb1a backgroundshowed no effect. This suggests that modulation of thesephenotypes by ddb2 is DDB1A dependent. These phe-notypes include dark hypocotyl elongation, dark antho-cyanin content, and rosette diameter. Our hypothesisfor the basis of this behavior is as follows. Due to the factthat the det1-1 mutation is not a null mutation, thereare small amounts of this protein still active in the cell.This small portion of DET1 may compete with DDB2for binding with DDB1A. Mutation of DDB2 may in-crease the availability of DDB1A inside the cell to inter-act with DET1. Since loss of DDB1A resulted in reducedDET1 activity, more DDB1A liberated from the DDB2complex may increase DET1 activity. This model ex-plains the basis of the suppression of det1 by ddb2 thatwe observed. However, for dark-grown hypocotyls, lossof function of DDB2 was found to enhance the det1phenotype. In this case, perhaps the increase in DDB1Aresults in an increase in inactive complexes or increaseddegradation of ligase components. There is evidencethat overexpression of E3 ligase components can re-sult in loss-of-function phenotypes (e.g., Gray et al.2002). Alternatively, both DET1-dependent and DDB2-dependent pathways may be required for optimal darkhypocotyl development.

Figure 5.—Floral and fruit morphology. (A) Silique length (in millimeters). (B) Number of seeds per half-silique (n ¼ 6). (C)(Top) Flower characteristics. From left to right: Col-0, ddb2, ddb1a, ddb1a ddb2, det1, det1 ddb2, det1 ddb1a, and det1 ddb1a ddb2.(Bottom) Dissected flower. From left to right: of Col-0, det1, det1 ddb2, det1 ddb1a, and det1 ddb1a ddb2. Error bars indicate95% C.I. *P # 0.05 or **P # 0.01, respectively, relative to DDB2 wild-type control.


DDB1A independent: For some phenotypes, loss offunction of DDB2 resulted in significant changes in boththe det1 and det1 ddb1a backgrounds. This pattern ofinteraction suggests that regulation of these phenotypesby ddb2 is DDB1A independent. These phenotypes in-clude light chlorophyll and anthocyanin content, flow-ering time (days), and height. Our hypothesis for thebasis of this behavior is as follows. Arabidopsis is the onlyorganism with two versions of the DDB1 protein (DDB1Aand DDB1B). DDB1B is expressed at lower levels thanDDB1A (Figure 1A). While DDB1A loss of function re-sults in no obvious phenotype in wild-type background,DDB1B loss of function is lethal (Schroeder et al. 2002);thus little is known about the role of DDB1B in lightsignaling. Perhaps DDB2 interacts with DDB1B in theabsence of DDB1A. Thus the DDB1A-independent phe-notypes may be acting via DDB1B. If the redundantaction of DDB1B results in DDB1A-independent phe-notypes, this implies that DDB1B is unable to compen-sate for DDB1A-dependent phenotypes. For example,light-grown seedlings exhibit independent phenotypes(chlorophyll, anthocyanin) while dark-grown seedlingsexhibit dependent phenotypes (hypocotyl, anthocya-nin). Genevestigator data show that expression of DDB1Aand DDB1B does not vary significantly between lightand dark and that DDB1A is expressed at twice the levelof DDB1B in both cases. Perhaps DDB1B is inactivated

post-transcriptionally in dark-grown seedlings, resultingin DDB1A-dependent phenotypes.

These models suggest that both DET1 and DDB2 areable to interact with DDB1A and DDB1B. Support forthis hypothesis is as follows. Arabidopsis DET1 andDDB1A interact genetically (Schroeder et al. 2002) aswell as in a yeast two-hybrid assay (Bernhardt et al.2006). In addition, DET1, DDB1A, COP10, CUL4, andRBX1 are able to form an active E3 Ub ligase complex(Chen et al. 2006). While no direct interaction betweenDET1 and DDB1B has been demonstrated, Bernhardt

et al. (2006) showed that both DDB1A and DDB1Binteract with At CUL4 in an in vitro pull-down assay, andmyc-tagged DET1 immunoprecipitates with CUL4 fromplant extracts. Residues required for interaction be-tween human DET1 and DDB1 have recently beenidentified: DDB1 910-913 MALY in b-propeller C ( Jinet al. 2006) and Arabidopsis DDB1A and DDB1B andrice DDB1 all contain the same variation of thissequence—LALY (Schroeder et al. 2002). With regardto DDB2/DDB1 interaction, Bernhardt et al. (2006)also showed that At DDB2 can interact with DDB1A in ayeast two-hybrid assay. No direct interaction betweenDDB2 and DDB1B has been demonstrated, nor have thehuman DDB1 residues required for interaction withDDB2 been identified, but competition experimentshave shown that DDB2 competes with the viral proteinSV5 for DDB1 interaction (Leupin et al. 2003) and SV5has been shown to interact with the b-propeller C ofDDB1 (T. Li et al. 2006). Several recent studies, however,have identified a WDXR motif at approximately residue273 in human DDB2 and other DDB1-interacting WD40proteins (DCAFs) that is required for interaction withDDB1 (Angers et al. 2006; He et al. 2006; Higa et al.2006; Jin et al. 2006). This motif is also conserved in riceand Arabidopsis DDB2 (Ishibashi et al. 2003). ThusDDB1 appears to act as a scaffold to recruit specificfactors, including DET1 and DDB2, to CUL4 E3 ligasecomplexes. Interaction with these specific factors ap-pears to primarily occur via the b-propeller C domain ofDDB1. In Arabidopsis, DDB1A appears to be expressedat nearly twice the level of DDB1B throughout develop-ment (Figure 1A), and both DET1 and DDB2 caninteract with DDB1A in vitro (Bernhardt et al. 2006).Our genetic analysis suggests that this interaction isfunctionally important.

Modulation of dark hypocotyl elongation: det1 seed-lings show short hypocotyls when grown under darkconditions, and this phenotype is similar to light-grownwild-type seedlings (Chory et al. 1989). Photoreceptorsmutations, which alone show long hypocotyls, in the det1background exhibit the det1 phenotype. This suggeststhat DET1 is functioning downstream of the photo-receptors (Wang and Deng 2002).

Enhancers of the det1 dark hypocotyl phenotypeinclude ddb1a (Schroeder et al. 2002) and a weak allele(cop1-6) of the WD-40 protein COP1. The double

Figure 6.—(A) 7-day-old seedlings of wild type, ddb2, cop1-4,and cop1-4 ddb2 grown under dark conditions. Error bars in-dicate 95% C.I. (B) Hypocotyl length of 7-day-old dark-grownseedlings.


mutant cop1-6 det1-1 exhibits dark purple cotyledons,very short hypocotyls, and adult lethality (Ang andDeng 1994). Similarly, we have shown that loss offunction of DDB2, another member of the WD-40protein family, enhances the short hypocotyl phenotypeof det1. All these enhancers appear to be acting nearDET1 in the light-signaling pathway.

Suppressors of the det1 dark hypocotyl phenotypeinclude CONSTANS-LIKE3 (col3). Dark-grown det1-1 col3exhibits 55% longer hypocotyls than det1-1 alone (Datta

et al. 2006). Loss-of-function alleles of the transcriptionfactor HY5 also suppress the det1 dark phenotype(Chory 1992; Pepper and Chory 1997). Mutation ofTED3 in the det1 background also completely suppressesthe short det1 hypocotyl (Pepper and Chory 1997).TED3 encodes a peroxisomal protein (Pex2p) involvedin Arabidopsis development (Hu et al. 2002). Thesesuppressors act downstream of DET1 in light signaling.

Regulation of chlorophyll content: Light-grown det1seedlings exhibit a pale phenotype compared to wildtype (Chory et al. 1989). This correlates with a decreasein CAB mRNA expression (Chory and Peto 1990). Theregion of the CAB2 promoter involved in DET1 upre-gulation of light expression has been mapped to theCAB upstream factor-1 element (CUF-1). The CUF-1element is bounded by the transcription factor HY5,and hy5 mutants also underexpress CAB2 in the light(Maxwell et al. 2003). Other studies indicate that circa-dian regulation of CAB transcription by DET1 mightbe post-transcriptional. Degradation of the LHY (lateelongated hypocotyl) factor is accelerated in det1-1mutants (Song and Carre 2005).

Our results show that mutation of DDB2 in det1 or det1ddb1a backgrounds significantly enhanced chlorophyllcontent of these seedlings (i.e., suppressed the det1 palephenotype). It will be interesting to determine the basisof this suppression.

Regulation of anthocyanin content: The anthocyaninbiosynthetic pathway is well studied. Several transcrip-tion factors that regulate anthocyanin biosynthesis havebeen identified. In addition, many environmental con-ditions regulate this pathway. Light is an importantfactor that regulates anthocyanin accumulation in plantcells (Koes et al. 2005). As shown previously, det1 seed-lings have more anthocyanin than wild type. Chory andPeto (1990) reported that the anthocyanin biosyntheticgene Chalcone synthase (CHS) is ectopically expressed inleaf mesophyll cells in det1 seedlings. Chory et al. (1989)studied CHS gene expression of dark-grown seedlingsand found that det1 seedlings have more (20- to 50-fold)CHS mRNA than the wild type.

In light-grown seedlings, we observed no significanteffect of ddb2 on det1 anthocyanin levels, but ddb2 sup-pressed anthocyanin accumulation in the det1 ddb1adouble mutant. In contrast, in dark-grown seedlings,ddb2 suppressed anthocyanin accumulation in a DDB1A-dependent fashion. Interestingly, while ddb2 regulation

of both anthocyanin levels and hypocotyl elongationwas found to be DDB1A dependent in dark-grown seed-lings, opposing effects were observed. Loss of functionof DDB2 suppressed the high anthocyanin content ofdet1 but enhanced the det1 short hypocotyl phenotype.This opposing trend was also observed for anothermutation. Loss of function of COL3 in the det1 back-ground enhanced the high anthocyanin content of det1but suppressed the short hypocotyl length of dark-grown seedlings (Datta et al. 2006).

Regulation of fertility parameters: Mutation ofDDB1A in the det1 background results in reduced floraldevelopment and seed production (Schroeder et al.2002). Mutation of DDB2 suppressed the short stamensand reduced dehiscence, number of seeds, and siliquelength in det1 single mutants, but enhanced thosephenotypes in the det1 ddb1a double mutant. Thesephenotypes were similar to those observed for loss-of-function alleles of jasmonic acid ( JA) biosynthesis orsignaling genes. JA, a lipid-derived signaling compoundpresent in most plant species, has been found to play amajor role in anther development. One model suggeststhat JA regulates water flow in the stamens and petals,which further regulates flower opening and stamenmaturation. On the other hand, regulation of pro-grammed cell death in the anther by JA as dehiscenceproceeds has also been proposed (Scott et al. 2004).Similar phenotypes were also observed in an over-expression line of RBX1 (Gray et al. 2002). RBX1 (alsoknown as ROC1) is a RING-domain protein and a com-ponent of cullin-containing E3 ubiquitin ligases, in-cluding the SCF-type complex COI1 involved in JAsignaling (Schwechheimer et al. 2004) and the humanDET1, COP1, DDB1, and CUL4A complex (Wertz et al.2004).

DDB1 and DDB2 loss-of-function mutants in otherorganisms: DDB1 and DDB2 were originally identifieddue to their role in nucleotide excision repair. HumanXPE patients have loss-of-function alleles of DDB2(Cleaver 2005). DDB2 knockout mice, although viable,also have increased tumor formation (Itoh 2004; Yoon

et al. 2005). Decrease in leaf width and length andincrease in UV sensitivity were observed in null ddb2mutants in Arabidopsis (Koga et al. 2006). While goodDDB2 homologs have been identified only in verte-brates and higher plants, DDB1 homologs also exist inDrosophila, Caenorhabditis elegans, and Schizosaccharomy-ces pombe. Recently, loss of function of DDB1 in humancells has been found to result in defects in UV-damagerepair (J. Li et al. 2006) and increased DNA double-strand break accumulation throughout the genome(Lovejoy et al. 2006). DDB1 knockout flies are lethal,suggesting a crucial role for DDB1 in Drosophiladevelopment (Takata et al. 2004). RNAi screens in C.elegans have shown that loss of DDB1 results in embry-onic and larval lethality (Kim and Kipreos 2006). Yeast(S. pombe) DDB1 was found to have a significant role in


preventing mutation and genome stability (Holmberg

et al. 2005), as well as a role in cell division andreplication control (Zolezzi et al. 2002; Bondar et al.2003). In tomato (Lycopersicon esculentum), High Pigment1 (HP-1) is homologous to Arabidopsis and humanDDB1. hp-1 mutants exhibit highly pigmented fruit andshort hypocotyls like mutants in the tomato DET1 gene,HP-2 (Lieberman et al. 2004; Liu et al. 2004). Thus, therole of DDB2 appears to be specific to DNA repair, whileDDB1 appears to have additional roles during develop-ment. This is consistent with biochemical evidence thatDDB2 specifically recruits the DDB1/CUL4 E3 ligase toDNA damage (Kapetanaki et al. 2006), while DDB1/CUL4 can form complexes with multiple targeting factors,such as CSA (Groisman et al. 2003) and DET1 (Wertz

et al. 2004; Bernhardt et al. 2006; Chen et al. 2006).Here we show that the ddb2 and ddb1a single and

double mutants exhibit no developmental phenotypeson their own, but in the det1-1 background, strong (ddb1a)or more subtle (ddb2) developmental effects can be ob-served. We propose that these effects are due to changes inthe activity of the DET1 complex, either directly (ddb1a)or indirectly (ddb2) via modulations of the DDB1A/DDB1B pool.

We thank Mehdi Sefidgar, Chris Kuusselka, and Christine Yurkowskifor technical assistance and Paul Mains and Jennifer Nemhauser forcritical suggestions. The work was supported by a grant from theNatural Sciences and Engineering Research Council of Canada.

LITERATURE CITED

Ang, H., and X. Deng, 1994 Regulatory hierarchy of photomorpho-genic loci: allele-specific and light-dependent interaction be-tween the HY5 and COP1 loci. Plant Cell 6: 613–628.

Angers, S., T. Li, X. Yi, M. Maccoss, R. Moon et al., 2006 Moleculararchitecture and assembly of the DDB1–CUL4A ubiquitin ligasemachinery. Nature 443: 590–593.

Bernhardt, A., E. Lechner, P. Hano, V. Schade, M. Dieterle et al.,2006 CUL4 associates with DDB1 and DET1 and its downregu-lation affects diverse aspects of development in Arabidopsis thali-ana. Plant J. 47: 591–603.

Bondar, T., V. Ekaterina, D. Ucker, W. Walden, S. Mirkin et al.,2003 Schizosaccharomyces pombe Ddb1 is functionally linked tothe replication checkpoint pathway. J. Biol. Chem. 278: 37006–37014.

Chen, H., Y. Shen, X. Tang, L. Yu, J. Wang et al., 2006 ArabidopsisCULLIN4 forms an E3 ubiquitin ligase with RBX1 and the CDDcomplex in mediating light control of development. Plant Cell18: 1991–2004.

Chen, M., J. Chory and C. Fankhauser, 2004 Light signal transduc-tion in higher plants. Annu. Rev. Genet. 38: 87–117.

Chory, J., 1992 A genetic model for light-regulated seedling Arabi-dopsis. Development 115: 337–354.

Chory, J., and C. Peto, 1990 Mutations in the DET1 gene affect cell-type-specific expression of light regulated genes and chloroplastdevelopment in Arabidopsis. Proc. Natl. Acad. Sci. USA 87: 8776–8780.

Chory, J., C. Peto, R. Feinbaum, L. Pratt and F. Ausubel,1989 Arabidopsis thaliana mutant that develops as a light-grownplant in the absence of light. Cell 58: 991–999.

Cleaver, J., 2005 Cancer in Xeroderma pigmentosum and related dis-orders of DNA repair. Nat. Rev. Cancer 5: 564–573.

Datta, S., G. Hettiarachchi, X. Deng and M. Holm,2006 Arabidopsis CONSTANS-LIKE3 is a positive regulator ofred light signaling and root growth. Plant Cell 18: 70–84.

Fankhauser, C., and J. Casal, 2004 Phenotypic characterization ofa photomorphogenic mutant. Plant J. 39: 747–760.

Gray, W. H., D. Sunethra and M. Estelle, 2002 Role of the Ara-bidopsis RING-H2 protein RBX1 in RUB modification and SCFfunction. Plant Cell 14: 2137–2144.

Groisman, R., J. Polanowska, J. Kuraoka, J. Sawada, M. Saijo et al.,2003 The ubiquitin ligase activity in the DDB2 and CSA com-plexes is differentially regulated by the COP9 signalosome in re-sponse to DNA damage. Cell 113: 357–367.

He, Y., C. McCall, J. Hu, Y. Zeng and Y. Xiong, 2006 DDB1 func-tions as a linker to recruit receptor WD40 proteins to CUL4–ROC1 ubiquitin ligases. Genes Dev. 20: 2949–2954.

Higa, L., M. Wu, T. Ye, R. Kobayashi, H. Sun et al., 2006 CUL4–DDB1 ubiquitin ligase interacts with multiple WD40-repeatproteins and regulates histone methylation. Nat. Cell Biol. 8:1277–1283.

Holmberg, C., O. Fleck, H. Hansen, C. Liu, R. Slaaby et al.,2005 Ddb1 controls genome stability and meiosis in fissionyeast. Genes Dev. 19: 853–862.

Hu, J., M. Aguirre, C. Peto, J. Alonso, J. Ecker et al., 2002 A rolefor peroxisomes in photomorphogenesis and development ofArabidopsis. Science 297: 405–409.

Ishibashi, T., K. Seisuke, Y. Taichi, F. Tomoyuki, K. Takata et al.,2003 Rice UV-damaged DNA binding protein homologuesare most abundant in proliferating tissues. Gene 308: 79–87.

Itoh, T., D. Cado, R. Kamide and S. Linn, 2004 DDB2 gene disrup-tion leads to skin tumors and resistance to apoptosis after expo-sure to ultraviolet light but not a chemical carcinogen. Proc. Natl.Acad. Sci. USA 101: 2052–2057.

Jin, J., E. E. Arias, J. Chen, J. W. Harper and J. C. Walker, 2006 Afamily of diverse Cul4-Ddb1-interacting proteins includes Cdt2,which is required for S phase destruction of the replication factorCdt1. Mol. Cell 23: 709–721.

Kapetanaki, M. G., J. Guerrero-Santoro, D. C. Bisi, C. L. Hsieh,V. Rapic-Otrin et al., 2006 The DDB1–CUL4ADDB2 ubiquitinligase is deficient in Xeroderma pigmentosum group E and tar-gets histone H2A at UV-damaged DNA sites. Proc. Natl. Acad. Sci.USA 103: 2588–2593.

Kim, Y., and E. T. Kipreos, 2006 The Caenorhabditis elegans replica-tion licensing factor CDT-1 is targeted for degradation by theCUL-4/DDB-1 complex. Mol. Cell. Biol. 27: 1394–1406.

Koes, R., W. Verweij and F. Quattrocchio, 2005 Flavonoids: a col-orful model for the regulation and evolution of biochemicalpathways. Trends Plant Sci. 10: 236–242.

Koga, A., T. Ishibashi, S. Kimura, Y. Uchiyama and K. Sakaguchi,2006 Characterization of T-DNA insertion mutants and RNAisilenced plants of Arabidopsis thaliana UV-damaged DNA bind-ing protein 2 (AtUV-DDB2). Plant Mol. Biol. 61: 227–240.

Leupin, O., S. Bontron and M. Strubin, 2003 Hepatitis B virus Xprotein and simian virus 5 V protein exhibit similar UV-DDB1binding properties to mediate distinct activities. J. Virol. 77:6274–6283.

Li, J., Q. Wang, Q. Zhu, M. El-Mahdy, G. Wani et al., 2006 DNAdamage binding protein component DDB1 participates in nucle-otide excision repair through DDB2 DNA-binding and cullin 4Aubiquitin ligase activity. Cancer Res. 66: 8590–8597.

Li, T., X. Chen, K. Garbutt, P. Zhou and N. Zheng, 2006 Struc-ture of DDB1 in complex with a paramyxovirus V protein: viralhijack of a propeller cluster in ubiquitin ligase. Cell 124: 105–117.

Lieberman, M., O. Segev, N. Gilboa, A. Lalazar and I. Levin,2004 The tomato homolog of the gene encoding UV-damagedDNA binding protein 1 (DDB1) underlined as the gene thatcauses the high pigment-1 mutant phenotype. Theor. Appl.Genet. 108: 1574–1581.

Liu, W., A. Nichols, J. Graham, R. Dualan, A. Abbas et al.,2000 Nuclear transport of human DDB protein induced by ul-traviolet light. J. Biol. Chem. 275: 21429–21434.

Liu, Y., S. Roof, Z. Ye, C. Barry and A. Van Tuinen, 2004 Manip-ulation of light signal transduction as a means of modifying fruitnutritional quality in tomato. Proc. Natl. Acad. Sci. USA 101:9897–9902.

Lovejoy, C., K. Lock, A. Yenamandra and D. Cortez, 2006 DDB1maintains genome integrity through regulation of Cdt1. Mol.Cell. Biol. 26: 7977–7990.


Mackinney, G., 1941 Absorption of light by chlorophyll solutions.J. Biol. Chem. 140: 315–322.

Maxwell, B. B., C. R. Andersson, D. S. Poole, S. A. Kay andJ. Chory, 2003 HY5, circadian clock-associated 1, and a cis-element, DET1 dark response element, mediate DET1 regula-tion of Chlorophyll a/b-Binding Protein 2 expression. Plant Physiol.133: 1565–1577.

Pepper, A., and J. Chory, 1997 Extragenic suppressors of the Arabi-dopsis detl mutant identify elements of flowering-time and light-response regulatory pathways. Genetics 145: 1125–1137.

Pepper, A., T. Delaney, T. Washburn, D. Poole and J. Chory,1994 DET1, a negative regulator of light-mediated develop-ment and gene expression in Arabidopsis, encodes a novelnuclear-localized protein. Cell 78: 109–116.

Schroeder, D., M. Gahrtz, B. Maxwell, K. Cook, M. Kan et al.,2002 De-etiolated 1 (DET1) and damaged DNA binding pro-tein 1 (DDB1) interact to regulate Arabidopsis photomorphogen-esis. Curr. Biol. 12: 1462–1472.

Schwechheimer, C., L. Irina and A. Villalobos, 2004 Cullin-containing E3 ubiquitin ligases in plant development. Curr.Opin. Plant Biol. 7: 677–686.

Scott, R., M. Spielman and H. Dickinson, 2004 Stamen structureand function. Plant Cell 16(Suppl): S46–S60.

Shiyanov, P., S. Hayes, M. Donepudi, A. Nichols, S. Linn et al.,1999 The naturally occurring mutants of DDB are impairedin stimulating nuclear import of the p125 subunit and E2F1-activated transcription. Mol. Cell. Biol. 19: 4935–4943.

Song, H. R., and I. A. Carre, 2005 DET1 regulates the proteasomaldegradation of LHY, a component of the Arabidopsis circadianclock. Plant Mol. Biol. 57: 761–771.

Takata, K., Y. Hideki, Y. Masamitsu and K. Sakaguchi, 2004 Dro-sophila damaged DNA-binding protein 1 is an essential factor fordevelopment. Genetics 168: 855–865.

Wang, H., and X. W. Deng, 2002 Phytochrome signaling mechanism,pp. 1–28 in The Arabidopsis Book, edited by C. Somerville and E.Meyerowitz. American Society of Plant Biologists, Rockville, MD.

Weigel, D., and J. Glazebrook, 2002 Arabidopsis: A Laboratory Manual.Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Wertz, I., K. O’Rourke, Z. Zhang, D. Dornan, D. Arnott et al.,2004 Human de-etiolated-1 regulates c-Jun by assembling aCUL4A ubiquitin ligase. Science 303: 1371–1374.

Yanagawa, Y., J. Sullivan, S. Komatsu, G. Gusmaroli, G. Suzuki

et al., 2004 Arabidopsis COP10 forms a complex with DDB1and DET1 in vivo and enhances the activity of ubiquitin conjugat-ing enzymes. Genes Dev. 18: 2172–2181.

Yoon, T., A. Chakrabortty, R. Franks, T. Valli, H. Kiyokawa et al.,2005 Tumor-prone phenotype of the DDB2-deficient mice. On-cogene 24: 469–478.

Zolezzi, F., J. Fuss, S. Uzawa and S. Linn, 2002 Characterization ofa Schizosaccharomyces pombe strain deleted for a sequence homo-logue of the human damaged DNA binding 1 (DDB1) gene. J.Biol. Chem. 277: 41183–41191.

Communicating editor: B. Bartel


Documents

DDB1A and DET1 Exhibit Complex Interactions During Arabidopsis … · 2007-07-25 · Wesam M. Al Khateeb and Dana F. Schroeder1 Department of Botany, University of Manitoba, Winnipeg,