(UBX) domain-containing protein AtPUX7 in ...

Gene 526 (2013) 299–308

Contents lists available at SciVerse ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

Functional characterization of the plant ubiquitin regulatory X (UBX) domain-containing protein AtPUX7 in Arabidopsis thaliana Jean-Luc Gallois a,⁎, Jan Drouaud b, Alain Lécureuil b, Anouchka Guyon-Debast b, Sandrine Bonhomme b, Philippe Guerche b a b

INRA-UR 1052 Génétique et Amélioration des Fruits et Légumes (GAFL), Domaine St Maurice, CS 60094, F-84143 Montfavet Cedex, France Institut Jean-Pierre Bourgin, UMR1318 INRA-AgroParisTech, INRA Centre de Versailles-Grignon, Route de St-Cyr (RD10), 78026 Versailles Cedex, France

a r t i c l e

i n f o

Article history: Accepted 21 May 2013 Available online 4 June 2013 Keywords: Arabidopsis thaliana UBX-containing protein p97/VCP/CDC48 Gametophyte

a b s t r a c t p97/CDC48 is a major AAA-ATPase that acts in many cellular events such as ubiquitin-dependent degradation and membrane fusion. Its specificity depends on a set of adaptor proteins, most of them containing the ubiquitin regulatory X (UBX) domain. Using a differential hybridization system, we isolated a UBX-containing protein that is expressed during the early phase of male gametophyte development in the crop Brassica napus and isolated and characterized its closest Arabidopsis thaliana homolog, AtPUX7. The AtPUX7 gene is expressed broadly in both the sporophyte and gametophyte due to regulation inferred by its first intron. The subcellular localization of AtPUX7 was assigned mainly to the nucleus in both the sporophyte and in pollen, mirroring the AAA-ATPase AtCDC48A localization. Furthermore, AtPUX7 interacts specifically with AtCDC48A in yeast as well as in planta in the nucleus. This interaction was mediated through the AtPUX7 UBX domain, which is located at the protein C-terminus, while an N-terminal UBA domain mediated its interaction with ubiquitin. Consistent with those results, a yeast-three hybrid analysis showed that AtPUX7 can act as a bridge between AtCDC48A and ubiquitin, suggesting a role in targeted protein degradation. It is likely that AtPUX7 acts redundantly with other members of the Arabidopsis PUX family because a null Atpux7-1 mutant does not display obvious developmental defects. © 2013 Elsevier B.V. All rights reserved.

1. Introduction ATPases associated with diverse cellular activities (AAA-ATPases) participate in a variety of cellular mechanisms including proteolysis, membrane fusion, protein folding and cytoskeletal regulation (Vale, 2000). Among these AAA-ATPases, Cdc48/p97, a highly abundant and ubiquitously expressed eukaryotic protein, assembles in a homohexameric complex. The Cdc48/p97 ATPase activity induces conformational change to this structure, generating a mechanical force that can be used to dissociate proteins from other cellular structures or to disassemble protein complexes (Ye, 2006). The specificity of the Cdc48/p97-generated mechanical forces relies on a set of specific cofactors, the p97 adaptors. The majority of these adaptor proteins are characterized by a conserved binding motif, the

Abbreviations: AAA-ATPase, ATPase associated with diverse cellular activities; ER, endoplasmic reticulum; ERAD, endoplasmic reticulum-associated protein degradation; IME, intron-mediated enhancement; PUX, plant ubiquitin regulatory X domaincontaining protein; Ub, ubiquitin; UBA, ubiquitin-associated; UBX, ubiquitin regulatory X; UIM, ubiquitin-interacting motif; UPS, ubiquitin/26S proteasome system. ⁎ Corresponding author. Tel.: +33 4 32 72 27 96; fax: +33 4 32 72 27 02. E-mail addresses: [email protected] (J.-L. Gallois), [email protected] (J. Drouaud), [email protected] (A. Lécureuil), [email protected] (A. Guyon-Debast), [email protected] (S. Bonhomme), [email protected] (P. Guerche). 0378-1119/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gene.2013.05.056

ubiquitin-regulatory X (UBX) domain (Kloppsteck et al., 2012). The UBX domain is an 80 amino acid domain that is typically located at the carboxyl terminus of proteins and often is found in conjunction with N-terminal UBA or UBL domains. The three dimensional structure of the UBX domain from the human Fas-associated factor 1 (FAF1) revealed a structure that can be superimposed on the one of ubiquitin, suggesting a common origin. However, because of the absence of key amino acids, the UBX peptides are unlikely to be conjugated to proteins in the same way as ubiquitin (Buchberger et al., 2001). Instead, the hallmark of the UBX domain is to bind CDC48/p97 through a conserved loop (Schuberth and Buchberger, 2008). All 11 tested human UBX proteins, six Caenorhabditis elegans and several yeast UBX-containing proteins were all shown to interact with their respective CDC48 partners (Alexandru et al., 2008; Decottignies et al., 2004; Sasagawa et al., 2010). UBX-containing proteins have been classified in different families based on the various other domains they contain but two major groups have been distinguished (Alexandru et al., 2008): those that possess an N-terminal UBA domain and bind ubiquitin, such as the human Fas-associated factor 1 (FAF1; Song et al., 2005), and those that lack a UBA domain. Members of the former group of ubiquitin-binding UBX proteins have been shown to play a major role in endoplasmic reticulum (ER)-associated protein degradation (ERAD) by mediating the binding of CDC48 to ERAD ubiquitinated substrates (Neuber et al., 2005; Schuberth and Buchberger, 2005), as well as to the ubiquitinproteasome system (UPS) (Song et al., 2005). The role in the UPS is

300

J.-L. Gallois et al. / Gene 526 (2013) 299–308

further reinforced in the human UBXD7 protein by the presence of another conserved domain, the ubiquitin interacting motif (UIM), which directly binds to the ubiquitin-like protein NEDD8 on the E3 cullin subunit, linking CDC48/p97 to multiple E3 ubiquitin ligases (Alexandru et al., 2008; Bandau et al., 2012; Besten et al., 2012). Therefore UBX proteins participate broadly to the regulation of protein degradation. In Arabidopsis thaliana, there are three genes encoding CDC48 isoforms (Rancour et al., 2002) and disruption of the more abundantly expressed gene, AtCDC48A, results in pleiotropic developmental defects in both seedlings and pollen tubes (Park et al., 2008). The Arabidopsis genome also encodes a family of 15 UBX-containing proteins among which the most characterized to date is the plant ubiquitin regulatory X domain-containing protein, PUX1 (Park et al., 2007; Rancour et al., 2004). PUX1 structure is unusual as its UBX domain is not located at the C-term of the protein but rather near the middle. Also, PUX1 promotes the dissociation of the hexameric AtCDC48 structure in vitro, suggesting a negative regulatory effect. Consistent with this observation, a null pux1 mutation triggers phenotypes that are opposite to those of a cdc48 loss-of-function mutant, mildly promoting cellular development and growth (Rancour et al., 2004). Further studies suggest that the negative effect of PUX1 on CDC48 relies on its unusual C-terminus (Park et al., 2007). Apart from this well-characterized gene, little is known about the other PUX genes in plants, although a report showing that AtPUX2 negatively regulates the powdery mildew–Arabidopsis interaction suggests that the role of plant PUX proteins may be diverse (Chandran et al., 2009). We previously set up a screen using cDNA subtraction and a differential hybridization strategy to isolate early expressed genes during male gametophyte development in rapeseed (Brassica napus), a widespread and important crop species (Fourgoux-Nicol et al., 1999). Male gametophyte development covers the haploid phase that, following meiosis in the anther, goes from microspore to tricellular pollen grain in which a large vegetative cell encloses the two sperm cells (Boavida et al., 2005). Through this screen, we characterized a subunit of the ubiquitin/26S proteasome system (UPS) E3 ligase, BnSKP1γ1 (Drouaud et al., 2000; Fourgoux-Nicol et al., 1999). BnSKP1γ1 is expressed early in male gametophyte development, consistent with the importance of targeted protein degradation during this development (Book et al., 2009; Gallois et al., 2009; Honys and Twell, 2004). Here, we isolate and characterize the PUX protein BnUBX1, a PUX protein expressed early in male gametophyte development, along with the homolog of BnUBX1 in the close relative A. thaliana. 2. Materials and methods 2.1. Plant materials and plant transformation B. napus L. cv. Brutor were grown under standard greenhouse conditions and were used for RNA extraction. A. thaliana Col (Columbia) and Ws (Wassilevskija) accessions were used as wild-type controls. The Atpux7-1 mutant allele was isolated from the Versailles collection of T-DNA insertion mutants (Ws accession) as FLAG-585C03. For growth on either plates or soil, seeds were stratified for 2 days at 4 °C and grown at 18 to 20 °C, with 16 h light (fluorescent light: 100 μmol photons m−2 s−1) and 8 h dark cycles. Immature flowers were emasculated and manually cross-pollinated for crosses. All binary vectors were transformed into A. thaliana using the floral dip method (Clough and Bent, 1998) in Col plants. All T1 plants were selected on Germination Media plates supplemented with 15 mg/L Hygromycin B. 2.2. Plasmid constructions All constructed plasmids and oligonucleotides used in this study are listed in Supplemental Tables 1 and 2, respectively.

Briefly, for the GUS reporter constructs an 819 bp B. napus upstream sequence to BnUBX1 cDNA was obtained by TAIL-PCR. This sequence includes 527 bp of promoter, 76 bp of 5′ UTR and 216 bp of the Exon I coding sequence. This fragment was translationally fused to the uidA gene in pJD81. The BnUBX1::uidA gene fusion was then introduced in the binary vector pEC2 (Cartea et al., 1998), resulting in the pJD131 plasmid. Similarly, a 1168 bp A. thaliana upstream sequence to AtPUX7 was transcriptionally fused to the uidA in pEC2, and a 1920 bp fragment, containing the 1168 bp promoter fragment of AtPUX7 plus the first exon (102 bp), the first intron (554 bp) and the beginning of the second exon (96 bp) was translationally fused to the uidA gene resulting in pPAG107 and pPAG113, respectively. To construct the new set of LAT52-gateway plasmids, a −492 bp LAT52 promoter was amplified on pLAT52::GUS (a gift from D. Twell) and cloned as a HindIII–KpnI fragment to replace the CaMV 35S promoter in the binary vectors pMDC32 and pMDC43 (Curtis and Grossniklaus, 2003), resulting in pJL231 and pJL233, respectively. Entry clones were prepared by RT-PCR amplification introducing the attB1/attB2 gateway recombination sequences, and BP clonase recombination into the pDONR207 vector (Invitrogen). All clones were sequence-checked. All further clones were obtained by LR clonase recombination reactions. Destination vectors used are as follows: pMDC43 for the CaMV 35S-driven GFP-C term fusion (Curtis and Grossniklaus, 2003); pJL233 for LAT52-driven GFP fusion (this study); pGADT7 gateway and pGBKT7 gateway for yeast two-hybrid interaction studies (Rossignol et al., 2007); pRED-NLS gateway for yeast threehybrid interaction studies (Ferrario et al., 2003) and pBIFP2 and pBIFP3 for bimolecular fluorescent complementation experiments (Azimzadeh et al., 2008). 2.3. Protein–protein interaction studies For yeast two-hybrid analyses, the Matchmaker GAL4 two-hybrid system 3 was used according to protocols described in the Clontech Yeast Protocol Handbook, except that the PJ69a and PJ69α yeast strains were used. pGADT7- and pGBKT7-derived vectors were transformed into PJ69a and PJ69α, respectively. After mating, yeast colonies transformed by both plasmids were selected on selective medium lacking Leucine and Tryptophane (SD-LW). For each combination, three large colonies were resuspended in 100 μL sterile water and 10 μL was spotted on SD-LW, as control, and on selective medium lacking Leucine, Tryptophane and Histidine (SD-LWH) on which growth only happened if both candidates interact. For yeast three-hybrid analyses, similar methods were followed except that PJ69α was first transformed with the pRED-NLS-derived vectors before being further re-transformed with pGBKT7-derived vectors. After mating, diploid yeasts transformed by pGADT7, pRED-NLS and pGBKT7 vectors were selected on media lacking Leucine, Tryptophane and Uracyl (SD-LWU) and interactions were assessed on media lacking Leucine, Tryptophane and Uracyl and Histidine (SD-LWUH). Bimolecular fluorescence complementation (biFC) assay was carried out as described (Azimzadeh et al., 2008). Co-expression of pBIFP2- and pBIFP3-derived vectors was performed by transient Agrobacterium tumefasciens transformation and YFP fluorescence was detected 2 to 3 days after infiltration. As a control, interaction between the MADS box transcription factors DEFICIENS and GLOBOSA was carried out using the clones pBIFP2-DEF and pBIFP3-GLO. 2.4. Histological analysis Histochemical GUS staining was performed as described (Marrocco et al., 2003). Flower stages were determined according to Smyth et al. (1990). Pollen viability was assessed using Alexander staining (Alexander, 1969) and observed with a PL Fluotar X25 dry objective on a Leitz Diaplan microscope.


A

T

M

BCP TCP

S

P

301

B

C

1.4 kb

Fig. 1. Characterization of BnUBX1, a male gametophyte-specific rapeseed gene. A. Northern blot of rapeseed tissues carried out with the M32 probe (top panel) and ethidium bromide stain of the agarose gel (bottom panel). Samples are labelled as follow: T, tetrad; M, microspore; BCP, bi-cellular pollen; TCP, tri-cellular pollen; S, sepal; P, plantlets. B–C. GUS staining of an Arabidopsis flower transgenic for the uidA coding sequence expressed under the control of the BnUBX1 promoter. B. Staining of a stage 6 flower according to Smyth et al. (1990). Petals and sepals were removed C. Stamens from a stage 12 flower.

GFP and FM4-64 imaging were obtained using a Leica SP2 AOBS confocal laser scanning microscope. For GFP imaging in pollen, DAPI staining observations (Park et al., 1998) were carried out using a Leica DMRXA2 microscope; photographs were taken using a CoolSNAP HQ (Roper) camera driven by Open LAB 4.0.4 software. 3. Results and discussion 3.1. BnUBX1 is expressed during early male gametophyte development in Brassicaceae We previously reported a differential hybridization strategy to isolate cDNAs expressed during early male gametophyte development in B. napus (Fourgoux-Nicol et al., 1999). One of the isolated clones, M32 allowed the detection of a 1.4 kb mRNA. The early expression of this

cDNA during B. napus male gametophyte development was confirmed by RNA blot analysis on mRNA extracted from all stages of male gametophyte development (Fig. 1A). M32 expression was found to be highly specific to the tetrad and microspore stages and could not be detected in mRNA extracted from any other tested tissues, including plantlets (i.e. sporophytic tissues). The isolated full-length M32 cDNA contains a 1158 base open reading frame, encoding a putative 386 amino acid (AA) protein (deposited in GenBank under the accession number JX566882). Sequence analyses revealed three conserved protein domains: an N-terminal UBA-like domain, a UAS domain and a C-terminal UBX domain (InterPro accession numbers IPR009060, IPR006577 and IPR001012, respectively). Because this structure is similar to that of human UBXD7 and UBXN-3 (Alexandru et al., 2008; Sasagawa et al., 2010), we named the gene encoding this protein BnUBX1.

A

B

UBA-like

BnUBX1

UBX

UAS UBA-like

AtPUX7

(386 AA)

UIM UAS

UBX

(468 AA)

UBA-like CeUBXN-3

UAS UBA

Hh-UBXD7

UBX

(632 AA)

UIM UAS

UBX

(489 AA)

Fig. 2. Protein sequences deduced from the BnUBX1 and AtPUX7 genes. A. Sequence alignment using ClustalW and boxshade software of the BnUBX1 and AtPUX7 proteins. The UBA-like (black: InterPro accession number IPR009060), UAS (red: InterPro accession number IPR006577), UIM (blue: InterPro accession number IPR003903) and UBX (green: IPR001012) domains are boxed. B. Schematic representation of BnUBX1 and AtPUX7 as well as their Caenorhabditis elegans UBXN-3 (F48A11.5) and human UBXD7 (NP_056377) homologs.

302


3.2. Characterization of an Arabidopsis protein highly homologous to BnUBX1

To analyse the pattern of BnUBX1 expression in planta, an 819 bp BnUBX1 genomic DNA fragment, including 520 bp of sequence upstream of the 5′ UTR, 84 bp of 5′ UTR and 215 bp of coding sequence, was fused to the uidA coding sequence and expressed in the related Brassicaceae A. thaliana. Transgenic plants of the related Brassicaceae A. thaliana were obtained for this construct and the expression pattern of the reporter gene was studied: as for the BnUBX1 mRNA expression in B. napus, no GUS expression could be detected during the sporophytic phase, and a strong expression was observed in the early male gametophyte (Figs. 1B, C). In A. thaliana lines harbouring a single-locuscopy of the BnUBX1-uidA transgene in the hemizygous state, about 50% of the pollen grains were GUS-positive, whereas GUS was not detected in the remaining 50%, confirming the gametophytic expression driven by the BnUBX1 promoter (data not shown). Taken together, these expression data indicate that the 604 bp B. napus promoter is sufficient to drive the highly specific gametophytic expression of BnUBX1 in A. thaliana, and suggest that the cis-elements contained in this region also drive the gametophytic expression profile in the native B. napus.

In order to gain insight into BnUBX1 function, we looked for putative orthologs in A. thaliana. To this end, we blasted the BnUBX1 protein sequence against the translated Arabidopsis genomic DNA sequences and isolated several genes encoding putative UBX proteins. The three closest homologs were found to encode intron-less pseudogenes as suggested by the presence of stop codons and/or frameshifts within their coding sequences (Sup. Figs. 1 and 2). Indeed, analysis of the closest related At4g14250 gene suggested that it encoded two tandemly arranged pseudogenes. As such, the gene annotation was split into two intron-less pseudogenes, At4g14245 and At4g14250, in the TAIR database (Lamesch et al., 2012). Details on these pseudogenes as well as on the related pseudogene At1g59550 will be reported elsewhere. Apart from these loci, EST analyses supported the expression of a closely related protein, encoded by At1g14570, which is homologous to BnUBX1 throughout its sequence (Identities = 140/303 (46%), Positives = 192/ 303 (63%), Gaps = 23/303 (7%)) (Fig. 2A).

A

B 1000

Root

RL

CL

Stem

Infl

AtPUX7 500

APT1 UNM

C

BCP

TCP

MPG

1168 pb promoter

GUS

D

1168 pb promoter

E1

I1

E2

GUS

Fig. 3. AtPUX7 expression pattern. A. RT-PCR expression of AtPUX7 in the indicated Arabidopsis tissues. APT1 is amplified as a control. RL, rosette leaves; CL, cauline leaves; Infl, inflorescence. B. AtPUX7 microarray relative expression data (gathered from Honys and Twell, 2004) during male gametophyte development. The development stages are UNM, uninucleate microspores; BCP, bicellular pollen; TCP, tricellular pollen; and MPG, mature pollen grains. C and D. AtPUX7 expression pattern in Arabidopsis plants of promoter–GUS fusions (C) and GUS translational fusions with AtUBX4 exon 2 (D). GUS staining is shown in a 6-day old sporophyte (left panel) and in all flowers from an inflorescence (right panel), presented from the youngest to the oldest flower, as well as siliques. Most of the petals and sepals were removed.


Analysis of the deduced 468-amino acid sequence of the NP_973827 protein encoded by At1g14570 indicated that the UBA-like domain (AA 6–52), the UAS domain (AA 170–293) and the UBX domain (AA 387–466) also present in BnUBX1 were conserved, and revealed an additional ubiquitin interacting motif (UIM), spanning AA 328–347. In line with previous reports on plant UBX domain containing proteins (Rancour et al., 2004), the NP_973827 protein was named PUX7 (for plant UBX domain-containing protein 7). With its additional UIM domain, AtPUX7 presents the same overall organization as the mammalian protein UBXD7 (Alexandru et al., 2008) (Fig. 2B). 3.3. The AtPUX7 promoter drives specific early male gametophyte expression whereas intron mediated enhancement drives AtPUX7 sporophytic expression To characterize the expression pattern of AtPUX7, RT-PCR was carried out on mRNA extracted from different A. thaliana organs

A

FM4-64

303

(Fig. 3A). AtPUX7 was found ubiquitously expressed in sporophytic tissues. Moreover, data collected from microarrays on different stages of male gametophyte development (Honys and Twell, 2004) showed that AtPUX7 also was expressed throughout male gametophyte development (Fig. 3B). To further characterize AtPUX7 protein expression, a 1.2 kb AtPUX7 promoter sequence was amplified, fused to the uidA reporter gene and expressed in A. thaliana. In contrast to ubiquitous expression of AtPUX7 mRNA, the reporter protein expression was detected uniquely in early gametophyte development (Fig. 3C), whereas a promoterless control construct showed no GUS staining in the anther. Enhancing introns that lay within 1 kb of the transcription start site have been shown to increase gene transcription (Le Hir et al., 2003) and can affect the expression pattern of the gene (Jeong et al., 2007). AtPUX7 contains a large first intron located 102 bp downstream of the transcription start site, thus, we hypothesised that AtPUX7 expression might be influenced by intron mediated enhancement (IME). To test

GFP

overlap

p35S:GFP:AtPUX7

B s s v

pLAT52:GUS

pLAT52:GFP::AtPUX7

pLAT52:GFP

pLAT52:GFP::AtCDC48A

Fig. 4. AtPUX7 is localized to the nuclei in both sporophytic and gametophytic cells. Confocal microscopy of cells from transgenic Arabidopsis plants expressing the GFP:AtPUX7 protein fusion expressed under the control of the CaMV 35S (A) or the pollen specific LAT52 (B) promoters. A. Confocal microscopy of root tip cells. From left to right, panels represent the FM4-64 channel—that stains root cell walls—the GFP channel, and the overlay of both. Scale bars represent 40 μm. B. Confocal microscopy of pollen grains. For each construct DAPI staining (left panel) and the GFP image (right panel) are shown. The large vegetative nucleus (V) and two dense spermatic nuclei (S) are indicated. pLAT52:GUS and pLAT52:GFP lines are also shown, as well as the GFP:AtCDC48A line.

304


this possibility, we used the IMeter web-based software (http://korflab. ucdavis.edu/cgi-bin/web-imeter.pl), which predicts the ability of an intron to enhance expression (Rose, 2008), to analyse AtPUX7. A high IMeter score (32.289) was obtained for the 554 bp AtPUX7 first intron, suggesting that AtPUX7 expression could be influenced by IME. As such, a new GUS reporter construct was obtained that contained 1168 bp of AtPUX7 promoter sequence and 752 bp of gene sequence, included the first intron and the second exon that was translationally fused to the GUS coding sequence. An Arabidopsis plant expressing this construct displayed strong ubiquitous GUS expression throughout the sporophyte (Fig. 3D). Although more experiments will be necessary to dissect the exact role of AtPUX7 intron 1, our data suggest a role for intron 1 and, thus, IME in driving strong AtPUX7 expression in the sporophyte. 3.4. AtPUX7 is localized in the nucleus of both sporophytic and gametophytic cells To assess a putative role for AtPUX7 during development, we analysed its subcellular localization of the AtPUX7 protein. To this end, the AtPUX7 coding sequence was cloned as a C-terminal fusion to the GFP6 fluorescent marker, under the control of the strong promoter 35S CaMV (Curtis and Grossniklaus, 2003), and its subcellular localization was assessed using confocal microscopy in the T2 progeny of four independent lines. The GFP6:AtPUX7 fusion was localized to the nucleus and, to a weaker extent, to the cytosol (Fig. 4A). Because of its significant expression during male gametophyte development, we investigated AtPUX7 subcellular localization in pollen. Due to the fact that the CaMV 35S promoter is poorly expressed during pollen development (Wilkinson et al., 1997), we drove the expression of AtPUX7:GFP fusion under the control of the strong pollen specific promoter LAT52 (Twell et al., 1989), which allows a specific and strong expression in the pollen vegetative cell (Eady et al., 1995). The AtPUX7:GFP6 fusion showed a strong fluorescent signal in nuclei as well as a uniform cytosolic fluorescence, similar to its localization pattern in sporophytic cells (Figs. 4A and B).

3.5. AtPUX7 interacts with both ubiquitin and AtCDC48 through its UBA-like and UBX domains, respectively It has been shown in different species that UBX domains bind to the AAA-ATPase CDC48/p97 (Schuberth and Buchberger, 2008), and that UBX proteins are CDC48 cofactors. In Arabidopsis, four UBX proteins have been shown to interact with AtCDC48 (PUX1–PUX4, Rancour et al., 2004). Further analysis of the domain required for PUX1–AtCDC48 interaction pointed to both the UBX domain and the C-terminal domain, making PUX1 an exception to the general rule (Rancour et al., 2004). There are three CDC48 genes in A. thaliana CDC48A, B and C (Rancour et al., 2002). Atcdc48a mutant plants show impaired male gametophyte development (Park et al., 2008), suggesting it plays a role at that stage. Indeed, AtCDC48A is expressed throughout the male gametophyte development (Honys and Twell, 2004), and YFP:CDC48A fusions are localized to the vegetative nucleus and around (Park et al., 2008). We cloned the more broadly expressed AtCDC48A gene (Feiler et al., 1995) using the same LAT52 driven expression as for AtPUX7, and found a similar strong expression of AtCDC48A in nucleus and in cytosol (Fig. 4B), mirroring the AtPUX7 subcellular localization. Given the presence of both a UBA-like and a UBX domain in AtPUX7, we tested by yeast two-hybrid experiments the AtPUX7 interaction potential with both ubiquitin and AtCDC48A. Because of the self-activation property of the UAS domain (data not shown), the AtPUX7 was fused to the GAL AD domain. Interactions between the UBA-like domain and ubiquitin and between UBX and AtCDC48, respectively, have been shown to be accurately monitored by this method (Bertolaet et al., 2001; Decottignies et al., 2004). We found the complete AtPUX7 to interact with both ubiquitin and AtCDC48 (Fig. 5A). To ensure that this binding was mediated by the UBA-like and the UBX domains, both cDNA fragments encompassing each of these two domains were used in a further yeast two-hybrid test (Figs. 5B and C). The interaction between the AtPUX7 UBA-like domain and ubiquitin as well as the interaction between the AtPUX7 UBX domain and CDC48A were confirmed in both orientations of bait and target. We note that the AtPUX7 sequence fragments used were not strictly limited to those

A AD -

BD

Ub

AtCDC48A

-

Ub

AtCDC48A

UBA

UAS

UBX SD-LW

B

SD-LWH

C AD -

BD

Ub

Ub

UBX

UBX AtCDC48A

SD-LWH

UBA

-

AtCDC48A UBX

AtCDC48A

SD-LWH

-

-

BD

SD-LW

UBA

AD UBA SD-LW

-

Ub

Fig. 5. AtPUX7 interacts with both ubiquitin and AtCDC48A in yeast two-hybrid system. Yeast transformed with both bait and prey vectors were spotted on selective dropout without Leucine or Tryptophane (SD-LW) as controls and on selective dropout without Leucine, Tryptophane or Histidine (SD-LWH) to check for interaction between both partners. In each case, a control with an empty vector (—) was tested for self-activation of the constructs. A. The complete AtPUX7 protein interacts with ubiquitin (Ub) and with AtCDC48A. B–C. Interactions carried on with AtPUX7 UBA-like (B) and UBX (C) domains, respectively.


305

3.6. AtPUX7 can bridge ubiquitin to AtCDC48A

YFPN DEF YFPC GLO

YFPN AtPUX7 YFPC AtCDC48A

YFPN AtPUX7 ΔUBX YFPC AtCDC48A

YFPN AtPUX7 YFPC GLO

Fig. 6. AtPUX7 interacts with AtCDC48A through its UBX domain. Protein–protein interactions between AtPUX7 deletions and AtCDC48 in Nicotiana benthamiana using bimolecular fluorescent complementation (BiFC). All tested proteins are linked as C-term fusions to the YFP N-term or C-term. Strong interaction between MADS box transcription factors Deficiens (DEF) and Globosa (GLO) is shown as control. All confocal microscopy pictures have been taken using the same settings. Bars represent 50 μm.

encoding the predicted UBA-like and UBX domains, leaving the question open that the interaction domain could also implicate sequences outside of these domains. Although a second putative ubiquitin binding domain in the central part of AtPUX7 (UIM, ubiquitin interacting motif, spanning AtPUX7 AA 328–347) was predicted, no interaction could be detected between this domain and ubiquitin (data not shown). It is possible that, as it has been shown for the human UBXD7, the UIM domain of AtPUX7 interacts directly with other ubiquitin-like proteins such as NEDD8 (Bandau et al., 2012; Besten et al., 2012). The existence of such interactions will require further testing. The interaction between AtPUX7 and AtCDC48A was also confirmed in vivo, by bimolecular fluorescence complementation (biFC, Fig. 6). Indeed, a strong interaction between AtPUX7 and AtCDC48A could be visualised both in the nucleus and in the cytosol in agrobacteriuminoculated Nicotiana benthamiana. This interaction is consistent with the subcellular localization of both AtPUX7 and AtCDC48A (Fig. 4). Deleting the C-terminal UBX domain of AtPUX7 strongly diminished the fluorescence signal to a level similar to that observed for the negative control (AtPUX7 + Globosa transcription factor).

AD

NLS

Ub

UBA

UAS

Ub

UBA

UAS

Ub

UAS

3.7. An Atpux7-1 mutant does not display obvious developmental or pollen transmission defects To gain insight into the gene function, we searched for Atpux7 mutant alleles in the public Arabidopsis mutant collections and identified the FLAG-585C03 line from the Versailles T-DNA collection. In this line, the T-DNA is inserted into the first AtPUX7 intron, making it a likely complete loss-of-function Atpux7 allele that we called Atpux7-1. RT-PCR analysis of plants homozygous for this mutation showed that AtPUX7 mRNA expression was undetectable, indicating that Atpux7-1 is a loss-of-function mutant (Figs. 8A–C). However, homozygous Atpux7-1 mutant plants did not display obvious sporophytic phenotypes compared to wild-type plants (Figs. 8D, E), and pollen viability as assessed by Alexander staining was found to be unaffected (Figs. 8F, G). As we suspected that the AtPUX7 loss-of-function might hinder the male gametophyte development, male transmission rates of the Atpux7-1 mutation was assessed by crossing Atpux7-1 heterozygous plants to wild-type plants and scoring by genotyping the percentage of Atpux7-1-bearing plants in the progeny. The rationale for this experiment is that, as male gametophytes are haploid, if the AtPUX7 gene is essential for male gametophyte development, the pollen containing the Atpux7-1 allele should degenerate and not be transmitted to the progeny. As a result, the presence of Atpux7 heterozygous plants in the progeny of these crosses should be much lower than the expected Mendelian 50% segregation (Bonhomme et al., 1998). When the heterozygous Atpux7-1 plants were used as the male parent, the segregation of

BD

-

Ub

Several UBX-containing proteins have been shown to be able to bridge ubiquitin with AtCDC48 in yeast (Neuber et al., 2005; Schuberth and Buchberger, 2005) and in mammals (Song et al., 2005). Because of the presence of both a UBA-like and a UBX domain in AtPUX7, we reasoned that AtPUX7 could act similarly. Therefore, to test this interaction, we used a yeast three-hybrid assay that is based on a modified yeast two-hybrid system where a third protein partner is expressed as a nuclear protein via the pRED-NLS gateway vector (Egea-Cortines et al., 1999; Ferrario et al., 2003). Briefly, we expressed in yeast the ubiquitin and AtCDC48A, both fused as a translational C-term fusion to the GAL4 AD and BD, respectively, and added a third partner by expressing various forms of AtPUX7 fused to an NLS signal. No direct interaction was found between ubiquitin and AtCDC48A when they were co-expressed with an empty pRED-NLS vector, which is consistent with the weak affinity of AtCDC48 for ubiquitin (Meyer et al., 2002; Ye et al., 2003). In contrast, the expression of the fulllength AtPUX7 protein allowed yeast growth on selective SD-LWUH (-Leu-Trp-Ura-His) medium, therefore showing that it is sufficient to bridge ubiquitin with AtCDC48A (Fig. 7). Consistent with the yeast two-hybrid experiments, deletion of either the UBA-like or the UBX domain was sufficient to suppress this interaction.

AtCDC48A

UBX

AtCDC48A AtCDC48A

UBX

AtCDC48A SD-LWU

SD-LWUH

Fig. 7. AtPUX7 acts as a bridge between ubiquitin and AtCDC48A, as shown by yeast three-hybrid interactions. Yeast transformed with both prey (AD) and bait (BD) vectors and a third pRED-NLS vector were spotted on selective dropout without Leucine, Tryptophane nor Uracyl (SD-LWU) as controls and on selective dropout without Leucine, Tryptophane, Uracyl nor Histidine (SD-LWUH) to check for interactions between the three partners.

306


A

TAG6 RP

LP

R

F

B

C

-

LP + RP

AtPUX7

TAG6 + RP

APT1

D

Ws

E

pux7 pux7-1

F

Ws

G

pux7-1

Fig. 8. The Atpux7-1 loss-of-function mutant does not affect sporophytic development or pollen viability. A. Schematic representation of the AtPUX7 gene. Black boxes represent exons while the T-DNA insertion is represented by an open triangle. Oligonucleotides used for genotyping or RT-PCR (B and C) are shown as arrowheads. B. DNA genomic and T-DNA left border PCR amplifications on DNA extracted from Wassilevskija (Ws) wild-type and homozygous pux7-1 plants. C. RT-PCR on RNA isolated from wild-type and homozygous mutant leaves. D–E. rosettes from Ws wild-type (D) and homozygous pux7-1 (E) 4-week old plants. F–G. Mature pollen stained with Alexander's stain in anther locules from Ws wild-type (F) and homozygous pux7-1 (G) plants. Bars = 1 cm in D and E and 100 μm in F and G.

Atpux7-1 in the progeny was found to be 40% (n = 82), suggesting that the male gametophyte transmission is not significantly affected by the AtPUX7 knock-out. We also carried out the reverse experiment by using the heterozygous Atpux7-1 plants as the female parent and found that, similarly, the Atpux7-1 mutation did not affect the female gametophyte transmission (transmission to progeny 44%, n = 86). Combined mutant analysis or multiple gene knockdowns show that several UBX-containing proteins act redundantly during sporulation in Saccharomyces cerevisiae as well as during spermatogenesis in C. elegans, respectively (Decottignies et al., 2004; Sasagawa et al., 2010). Interestingly in C. elegans, the three redundant genes encode UBX proteins that show considerable structural differences: UBXN-3 is

the closest to AtPUX7 (Fig. 2B), UBXN-1 contains both a UBA and a Zinc-finger domain (InterPro accession number IPR015880) and UBXN-2 contains a SEP domain (IPR012989). This data indicates that UBX proteins can act redundantly despite having different overall domain compositions, suggesting that additional AtPUX genes may be involved in both gametophyte and sporophyte development. 4. Conclusion In this work, we describe two new plant ubiquitin regulatory X domain-containing (PUX) genes in the two related Brassicaceae B. napus and A. thaliana. Both BnUBX1 and AtPUX7 are expressed


early during male gametophyte development, and we show that this expression is driven by their respective promoters. Interestingly, AtPUX7 shows a much broader and ubiquitous expression pattern in both the gametophyte and sporophyte, and this expression is controlled by the first intron of AtPUX7 likely through intron-mediated enhancement. Our functional characterization of AtPUX7 showed that it displays all the hallmarks of a functional UBX-containing protein. The protein is mainly localized to the nucleus were it interacts with the AAA-ATPase CDC48A through its UBX domain. It was shown also to interact in yeast two-hybrid experiments with ubiquitin through its UBA-like domain, making it a likely bridge between the AtCDC48 segregase and ubiquitinated substrate. In this way, AtPUX7 is a likely CDC48/p97 adaptor that could help to adjust the plant CDC48/p97 function to targeted protein degradation (Halawani and Latterich, 2006). However, in line with this hypothesis, and given the importance of the UPS machinery during gametophyte development (Gallois et al., 2009), one might expect that the Atpux7-1 mutation would affect male gametophyte development, as Atcdc48a does (Park et al., 2008). However, this is not the case. A possible explanation for the lack of male gametophytic defects in the Atpux7-1 mutant is that there are 15 putative AtPUX genes in A. thaliana, so AtPUX7 might act redundantly with other AtPUX genes, as has been shown for S. cerevisiae and C. elegans UBX proteins. It is very likely that PUX function in plant gametophyte and sporophyte development can be unravelled by both determining their expression patterns and by systematically analysing combinatorial mutants. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.gene.2013.05.056. Conflict of interest statement The authors declare no conflict of interest. Acknowledgments We thank David Twell, Pascale Rossignol, François Parcy, Pieter Ouwerkerk and Richard Immink for plasmid gifts and Donghui Li (TAIR) for helping with the gene annotation. Laurie Grandont is acknowledged for providing the pLG001 and pLG006 plasmids. Finally, we thank Allison Mallory for editing the manuscript. References Alexander, M.P., 1969. Differential staining of aborted and nonaborted pollen. Stain. Technol. 44, 117–122. Alexandru, G., Graumann, J., Smith, G.T., Kolawa, N.J., Fang, R., Deshaies, R.J., 2008. UBXD7 binds multiple ubiquitin ligases and implicates p97 in HIF1alpha turnover. Cell 134, 804–816. Azimzadeh, J., et al., 2008. Arabidopsis TONNEAU1 proteins are essential for preprophase band formation and interact with centrin. Plant Cell 20, 2146–2159. Bandau, S., Knebel, A., Gage, Z.O., Wood, N.T., Alexandru, G., 2012. UBXN7 docks on neddylated cullin complexes using its UIM motif and causes HIF1alpha accumulation. BMC Biol. 10, 36. Bertolaet, B.L., et al., 2001. UBA domains of DNA damage-inducible proteins interact with ubiquitin. Nat. Struct. Biol. 8, 417–422. Besten, W., Verma, R., Kleiger, G., Oania, R.S., Deshaies, R.J., 2012. NEDD8 links cullin-RING ubiquitin ligase function to the p97 pathway. Nat. Struct. Mol. Biol. 19, 511–516. Boavida, L.C., Becker, J.D., Feijo, J.A., 2005. The making of gametes in higher plants. Int. J. Dev. Biol. 49, 595–614. Bonhomme, S., et al., 1998. T-DNA mediated disruption of essential gametophytic genes in Arabidopsis is unexpectedly rare and cannot be inferred from segregation distortion alone. Mol. Gen. Genet. 260, 444–452. Book, A.J., et al., 2009. The RPN5 subunit of the 26S proteasome is essential for gametogenesis, sporophyte development, and complex assembly in Arabidopsis. Plant Cell 21, 460–478. Buchberger, A., Howard, M.J., Proctor, M., Bycroft, M., 2001. The UBX domain: a widespread ubiquitin-like module. J. Mol. Biol. 307, 17–24. Cartea, M.E., Migdal, M., Galle, A.M., Pelletier, G., Guerche, P., 1998. Comparison of sense and antisense methodologies for modifying the fatty acid composition of Arabidopsis thaliana oilseed. Plant Sci. 136, 181–194.

307

Chandran, D., et al., 2009. Temporal global expression data reveal known and novel salicylate-impacted processes and regulators mediating powdery mildew growth and reproduction on Arabidopsis. Plant Physiol. 149, 1435–1451. Clough, S.J., Bent, A.F., 1998. Floral dip: a simplified method for Agrobacteriummediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. Curtis, M.D., Grossniklaus, U., 2003. A gateway cloning vector set for high-throughput functional analysis of genes in planta. Plant Physiol. 133, 462–469. Decottignies, A., Evain, A., Ghislain, M., 2004. Binding of Cdc48p to a ubiquitin-related UBX domain from novel yeast proteins involved in intracellular proteolysis and sporulation. Yeast 21, 127–139. Drouaud, J., Marrocco, K., Ridel, C., Pelletier, G., Guerche, P., 2000. A Brassica napus skp1-like gene promoter drives GUS expression in Arabidopsis thaliana male and female gametophytes. Sex. Plant Reprod. 13, 29–35. Eady, C., Lindsey, K., Twell, D., 1995. The significance of microspore division and division symmetry for vegetative cell-specific transcription and generative cell differentiation. Plant Cell 7, 65–74. Egea-Cortines, M., Saedler, H., Sommer, H., 1999. Ternary complex formation between the MADS-box proteins SQUAMOSA, DEFICIENS and GLOBOSA is involved in the control of floral architecture in Antirrhinum majus. Embo J. 18, 5370–5379. Feiler, H.S., Desprez, T., Santoni, V., Kronenberger, J., Caboche, M., Traas, J., 1995. The higher plant Arabidopsis thaliana encodes a functional CDC48 homologue which is highly expressed in dividing and expanding cells. EMBO J. 14, 5626–5637. Ferrario, S., Immink, R.G., Shchennikova, A., Busscher-Lange, J., Angenent, G.C., 2003. The MADS box gene FBP2 is required for SEPALLATA function in petunia. Plant Cell 15, 914–925. Fourgoux-Nicol, A., Drouaud, J., Haouazine, N., Pelletier, G., Guerche, P., 1999. Isolation of rapeseed genes expressed early and specifically during development of the male gametophyte. Plant Mol. Biol. 40, 857–872. Gallois, J.L., et al., 2009. The Arabidopsis proteasome RPT5 subunits are essential for gametophyte development and show accession-dependent redundancy. Plant Cell 21, 442–459. Halawani, D., Latterich, M., 2006. p97: the cell's molecular purgatory? Mol. Cell 22, 713–717. Honys, D., Twell, D., 2004. Transcriptome analysis of haploid male gametophyte development in Arabidopsis. Genome Biol. 5, R85. Jeong, Y.M., Mun, J.H., Kim, H., Lee, S.Y., Kim, S.G., 2007. An upstream region in the first intron of petunia actin-depolymerizing factor 1 affects tissue-specific expression in transgenic Arabidopsis (Arabidopsis thaliana). Plant J. 50, 230–239. Kloppsteck, P., Ewens, C.A., Forster, A., Zhang, X., Freemont, P.S., 2012. Regulation of p97 in the ubiquitin–proteasome system by the UBX protein-family. Biochim. Biophys. Acta 1823, 125–129. Lamesch, P., et al., 2012. The Arabidopsis Information Resource (TAIR): improved gene annotation and new tools. Nucleic Acids Res. 40, D1202–D1210. Le Hir, H., Nott, A., Moore, M.J., 2003. How introns influence and enhance eukaryotic gene expression. Trends Biochem. Sci. 28, 215–220. Marrocco, K., Lecureuil, A., Nicolas, P., Guerche, P., 2003. The Arabidopsis SKP1-like genes present a spectrum of expression profiles. Plant Mol. Biol. 52, 715–727. Meyer, H.H., Wang, Y., Warren, G., 2002. Direct binding of ubiquitin conjugates by the mammalian p97 adaptor complexes, p47 and Ufd1–Npl4. EMBO J. 21, 5645–5652. Neuber, O., Jarosch, E., Volkwein, C., Walter, J., Sommer, T., 2005. Ubx2 links the Cdc48 complex to ER-associated protein degradation. Nat. Cell Biol. 7, 993–998. Park, S.K., Howden, R., Twell, D., 1998. The Arabidopsis thaliana gametophytic mutation gemini pollen1 disrupts microspore polarity, division asymmetry and pollen cell fate. Development 125, 3789–3799. Park, S., Rancour, D.M., Bednarek, S.Y., 2007. Protein domain–domain interactions and requirements for the negative regulation of Arabidopsis CDC48/p97 by the plant ubiquitin regulatory X (UBX) domain-containing protein, PUX1. J. Biol. Chem. 282, 5217–5224. Park, S., Rancour, D.M., Bednarek, S.Y., 2008. In planta analysis of the cell cycledependent localization of AtCDC48A and its critical roles in cell division, expansion, and differentiation. Plant Physiol. 148, 246–258. Rancour, D.M., Dickey, C.E., Park, S., Bednarek, S.Y., 2002. Characterization of AtCDC48. Evidence for multiple membrane fusion mechanisms at the plane of cell division in plants. Plant Physiol. 130, 1241–1253. Rancour, D.M., Park, S., Knight, S.D., Bednarek, S.Y., 2004. Plant UBX domain-containing protein 1, PUX1, regulates the oligomeric structure and activity of Arabidopsis CDC48. J. Biol. Chem. 279, 54264–54274. Rose, A.B., 2008. Intron-mediated regulation of gene expression. Curr. Top. Microbiol. Immunol. 326, 277–290. Rossignol, P., Collier, S., Bush, M., Shaw, P., Doonan, J.H., 2007. Arabidopsis POT1A interacts with TERT-V(I8), an N-terminal splicing variant of telomerase. J. Cell Sci. 120, 3678–3687. Sasagawa, Y., Yamanaka, K., Saito-Sasagawa, Y., Ogura, T., 2010. Caenorhabditis elegans UBX cofactors for CDC-48/p97 control spermatogenesis. Genes Cells 15 (12), 1201–1215. Schuberth, C., Buchberger, A., 2005. Membrane-bound Ubx2 recruits Cdc48 to ubiquitin ligases and their substrates to ensure efficient ER-associated protein degradation. Nat. Cell Biol. 7, 999–1006. Schuberth, C., Buchberger, A., 2008. UBX domain proteins: major regulators of the AAA ATPase Cdc48/p97. Cell. Mol. Life Sci. 65, 2360–2371. Smyth, D.R., Bowman, J.L., Meyerowitz, E.M., 1990. Early flower development in Arabidopsis. Plant Cell 2, 755–767. Song, E.J., Yim, S.H., Kim, E., Kim, N.S., Lee, K.J., 2005. Human Fas-associated factor 1, interacting with ubiquitinated proteins and valosin-containing protein, is involved in the ubiquitin–proteasome pathway. Mol. Cell. Biol. 25, 2511–2524.

308


Twell, D., Wing, R., Yamaguchi, J., McCormick, S., 1989. Isolation and expression of an anther-specific gene from tomato. Mol. Gen. Genet. 217, 240–245. Vale, R.D., 2000. AAA proteins. Lords of the ring. J Cell Biol 150, F13–F19. Wilkinson, J.E., Twell, D., Lindsey, K., 1997. Activities of CaMV 35S and nos promoters in pollen: implications for field release of transgenic plants. J. Exp. Bot. 48, 265–275.

Ye, Y., 2006. Diverse functions with a common regulator: ubiquitin takes command of an AAA ATPase. J. Struct. Biol. 156, 29–40. Ye, Y., Meyer, H.H., Rapoport, T.A., 2003. Function of the p97–Ufd1–Npl4 complex in retrotranslocation from the ER to the cytosol: dual recognition of nonubiquitinated polypeptide segments and polyubiquitin chains. J. Cell Biol. 162, 71–84.