The ubiquitin ligase XBAT32 regulates lateral root development in Arabidopsis


(fax +352 392 6532; e-mail


Ubiquitin-mediated protein modification plays a key role in many cellular signal transduction pathways. The Arabidopsis gene XBAT32 encodes a protein containing an ankyrin repeat domain at the N-terminal half and a RING finger motif. The XBAT32 protein is capable of ubiquitinating itself. Mutation in XBAT32 causes a number of phenotypes including severe defects in lateral root production and in the expression of the cell division marker CYCB1;1::GUS. The XBAT32 gene is expressed abundantly in the vascular system of the primary root, but not in newly formed lateral root primordia. Treatment with auxin increases the expression of XBAT32 in the primary root and partially rescues the lateral root defect in xbat32-1 mutant plants. Thus, XBAT32 is a novel ubiquitin ligase required for lateral root initiation.


Lateral roots originate from differentiated pericycle cells that are arrested in the G2 phase of the cell cycle (Beeckman et al., 2001). Upon initiation, the pericycle cells divide and develop lateral root apical meristems which in turn produce mature lateral roots (Malamy and Benfey, 1997). Several lines of evidence support that auxin plays a key role in the regulation of lateral root formation. Treatment of roots with exogenous indole-3-acetic acid (IAA) promotes lateral root production (Laskowski et al., 1995). Arabidopsis mutants with increased endogenous IAA levels display excess lateral root proliferation (Boerjan et al., 1995; King et al., 1995). In contrast, auxin-resistant mutants show a reduction in the number of lateral roots (Hobbie and Estelle, 1995). Auxin transport has also been implicated in the control of lateral root production (Reed et al., 1998). Blocking auxin transport by inhibitors arrests lateral root development (Casimiro et al., 2001; Ruegger et al., 1997).

Ubiquitin-mediated protein modification has been implicated in auxin-mediated lateral root formation. Ubiquitin is first activated by the ubiquitin-activating enzyme (E1), then transferred to the ubiquitin-conjugating enzyme (E2), and finally linked to a target substrate by the ubiquitin-protein ligase (E3) (Ciechanover, 1998). The specificity of the ubiquitin pathway is determined by E3, which often binds to the targeted substrate. Upon attachment to ubiquitin, the target protein can be degraded by the 26S proteasome or subjected to other proteolysis-independent processes (Ben-Neriah, 2002; Ciechanover, 1998).

The first E3 isolated in the auxin-mediated lateral root developmental pathway was the Arabidopsis SCFTIR1 protein complex. The key component, TIR1, was identified from an auxin-responsive mutant that is resistant to the growth-inhibiting properties of the auxin-transport inhibitors N-1-naphthylphthalamic acid (NPA) and 2-carboxyphenyl-3-phenylpropane-1,2-dione (CPD) (Ruegger et al., 1998). Mutation in TIR1 disrupts a number of auxin-regulated growth processes including hypocotyl elongation and lateral root formation. TIR1 encodes a protein that contains leucine-rich repeats and an F box motif. The TIR1 protein forms the SCFTIR1 complex by interacting with cullin AtCUL1, the RING-H2 protein RBX1, and an SKP1-like protein (ASK1 or ASK2) (Gray et al., 1999, 2002, 2003). The AUX/IAA proteins are targets of the SCFTIR1 complex. Auxin promotes the binding of SCFTIR1 with AXR2/IAA7 and AXR3/IAA17, and the subsequent degradation of the targets (Gray et al., 2001).

The Arabidopsis SINAT5 is a RING finger-containing E3 (Xie et al., 2002). Transgenic plants that overexpress SINAT5 show a reduction in the number of lateral roots, whereas plants that overexpress a dominant-negative mutant of SINAT5 produced more lateral roots. SINAT5 was identified as a binding protein of the transcription activator NAC1 in a yeast two-hybrid screening. NAC1 consists of an N-terminal-conserved NAC domain that binds to DNA and a C-terminal transcription activation domain (Xie et al., 2000). Overexpression of NAC1 resulted in more lateral roots. In contrast, reduction in NAC1 by antisense led to a decrease in the emergence of lateral roots. Interestingly, SINAT5 can both autoubiquitinate and ubiquitinate NAC1 (Xie et al., 2002). SINAT5 attenuates the auxin signal by targeting NAC1 for degradation.

In this report, we demonstrate that the Arabidopsis gene XBAT32 encodes a protein with ubiquitin ligase activity. Through characterization of an xbat32 mutant, we show that this gene is positively involved in lateral root development.


XBAT32 is a RING-finger-containing protein

The rice protein XB3 was identified as a binding protein of the resistance gene product XA21 using the yeast two-hybrid approach (L.-Y. Pi and W.-Y. Song, unpublished data; Song et al., 1995). XBAT32 (for XB3 ortholog 2 in Arabidopsis), is one of the five Arabidopsis gene products structurally related to XB3. Phylogentic and sequence analyses revealed that the five Arabidopsis products can be grouped into three classes with 23–73% similarity (Figure 1a, Table 1). The genomic sequence of XBAT32 is 2357 bp with 10 exons disrupted by nine introns. The sequencing of cDNA from XBAT32 indicates that the introns are processed as predicted. XBAT32 encodes a protein of 508 amino acids (aa) (Figure 1b). The N-terminus consists of two putative overlapping N-myristoylation sites (Resh, 1999), suggesting that the XBAT32 protein may localize to cellular membranes. Amino acids 15–274 and aa 321–371 carry eight imperfect copies of ankyrin repeats and a RING finger motif, respectively. Ankyrin repeats are implicated in protein–protein interactions (Sedgwick and Smerdon, 1999), whereas RING fingers often function as E3s (Lorick et al., 1999). The C-terminus of XBAT32 contains 137 aa with no homology to known functional domains.

Figure 1.

(a) Phylogenetic tree based on the predicted amino acid sequences of the five XBAT genes from Arabidopsis thaliana. The GenBank accession numbers are At2g28840 (XBAT31), At4g14365 (XBAT34), At3g23280 (XBAT35), At5g07270 (XBAT33), and At5g57740 (XBAT32). Phylogenetic tree was generated with the ClustalW program from EBI (
(b) Predicted amino acid sequence of the XBAT32 protein. The deduced protein domains are indicated as follows: (I) putative N-myristoylation sites (underlined); (II) ankyrin repeats (the highly conserved amino acids in the ankyrin domain are shown in bold); (III) unknown; (IV) RING finger (the eight conserved cysteine and histidine residues that are important for zinc-chelating in the RING finger are shown in italic); (V) C-terminus tail with no known function.

Table 1.  Amino acid sequence comparisons between the five XBAT proteins from Arabidopsis. Identity (left) and similarity (right) values (%) were calculated using the ‘GAP’ tool within the Genetics Computer Group (Madison, WI) program
XBAT3230, 35   
XBAT3332, 3951, 58  
XBAT3427, 3424, 3023, 29 
XBAT3530, 5023, 2320, 3068, 73

The XBAT32 protein is an active E3 capable of autoubiquitination

A number of RING finger-based E3s can catalyze autoubiquitination in an E2-dependent manner (Lorick et al., 1999). To determine whether XBAT32 has E3 activity, we examined its capacity to mediate autoubiquitination in vitro. The resin-bound MBP-XBAT32 was incubated with 32P-labeled ubiquitin and ATP in the presence or absence of the wheat E1 and the human E2 protein UbcH5B. The appearance of ubiquitinated products with higher molecular weight was observed only in the presence of E1 and E2 (Figure 2). The MBP control was not ubiquitinated. Furthermore, we examined whether an MBP-XBAT32 mutant in which the metal-binding residue Cys321 is replaced with Ala can support the ubiquitination with the above assay. As shown in Figure 2, the XBAT32 mutant protein was unable to autoubiquitinate. Thus, XBAT32 is capable of autoubiquitination in an E1 and E2-dependent manner.

Figure 2.

Autoubiquitination of the XBAT32 fusion protein. GST-ubiquitin was labeled with [γ-32P]ATP by protein kinase A, digested with thrombin to release the 32P-ubiquitin. Resin-bound MBP and its fusion proteins were incubated with 32P-ubiquitin and ATP at 30°C for 90 min in the presence or absence of the wheat E1 and/or the human E2 UbcH5B. C321A is a mutant of MBP-XBAT32 in which the conserved Cys321 was replaced with Ala. Samples were resolved by 8% SDS-PAGE, followed by autoradiography. MBP-XBAT32 migrates at approximately 97.7 kDa (indicated by an arrow).

Tissue-specific expression and auxin induction of the XBAT32 gene

Northern blot analysis indicated that XBAT32 is expressed in roots, flowers, stems and leaves (Figure 3a). A weak signal of the XBAT32 transcript was also detected from the stem, but not from siliques. To obtain a more detailed expression pattern of XBAT32, a 1830-bp sequence containing the 5′-genomic region of XBAT32 was fused to the β-glucuronidase (GUS) reporter gene and the construct was transformed into wild-type Arabidopsis plants. Consistent with the Northern blot results, high XBAT32::GUS activity was observed in the vascular system of primary roots, particularly in the zone where lateral root initiation occurs (Figure 3b). In contrast, no XBAT32::GUS expression was detected in the newly formed lateral roots. Strikingly, we did not find the XBAT32::GUS activity in the apical meristem of primary and lateral roots. In leaves, XBAT32::GUS staining was primarily detected in the vascular tissue through the leaf blade. In flowers, XBAT32::GUS expression was found in the anthers.

Figure 3.

Tissue-specific expression of the XBAT32 gene.
(a) Northern blot analysis showing XBAT32 expression in different tissues.
(b) Histochemical localization of XBAT32::GUS expression in transgenic plants. (I) whole seedling, (II) young lateral root, (III) primary root tip, (IV) mature leaves, (V) flower.
(c) XBAT32::GUS expression in root of transgenic seedlings with or without treatment of 2 μm naphthalene-1-acetic acid (NAA) for 24 h. Tissues were stained for GUS activity for 1 h at 37°C. Scale bar, 100 μm.

Auxin can induce the expression of several genes that are involved in root development (Reed, 2001; Xie et al., 2000, 2002). To test whether auxin regulates the expression of XBAT32, we transferred 3-day-old XBAT32::GUS seedlings, grown on half-strength MS media, to the same media supplemented with the auxin naphthalene-1-acetic acid (NAA) or IAA. Twenty-four hours after the treatment, the seedlings were subjected to the GUS assay. As shown in Figure 3(c), the expression of XBAT32::GUS is increased in the vascular system of the primary root in the presence of NAA. Similar results were obtained from the XBAT32::GUS seedlings treated with IAA (data not shown).

The mutant allele xbat32-1 is deficient in lateral root formation

A search of the Salk insertion mutant database (Alonso et al., 2003) led to the identification of a mutant line carrying a T-DNA insertion in the XBAT32 gene in the Col-0 background. The insertion is located in the third exon, between +273 and +274 relative to the start codon (Figure 4a). This was confirmed by sequencing the T-DNA border flanking regions amplified from the mutant line. A Southern blot analysis identified a homozygous mutant line, referred to as xbat32-1 (Figure 4b). Northern blot analysis further revealed that no detectable XBAT32 RNA transcripts were present in the root tissue of the mutant line (Figure 4c), therefore xbat32-1 is a null allele.

Figure 4.

The xbat32-1 mutant is deficient in lateral root development.
(a) Scheme of the XBAT32 gene structure. Two inverted repeats of T-DNA were inserted after amino acid 91. The ATG and TGA codons and the location of the T-DNA insertion in the xbat32-1 allele are marked. Closed boxes represent exons and lines between boxes represent introns. A bar indicates scale.
(b) Identification of the homozygous mutant line xbat32-1 by Southern blot analysis. Genomic DNA from wild-type Col-0 and three T-DNA insertion-containing plants were hybridized with an XBAT32-specific probe.
(c) RNA blot analysis shows that the xbat32-1 mutant line expresses no detectable XBAT32 transcripts. Total RNA from wild-type Col-0 and xbat32-1 plants was probed with an XBAT32-specific sequence. Both autorad (upper) and agarose gel (lower) are shown.
(d–f) Wild-type Col-0 and xbat32-1 plants were grown in soil. Photographs were taken from 27-day-old (d), 35-day-old (e), and 56-day-old plants (f).
(g, h) Wild-type Col-0 and xbat32-1 plants were grown on half-strength MS media. Photographs were taken from 11-day-old (g) and 35-day-old plants (h). Scale bars, 1 cm.

Compared with wild-type Col-0, the xbat32-1 plants appeared to be smaller in overall size during most of the developmental stages. For instance, at the 27th day, the mutant plants were compact with relatively small, round leaves having short petioles (Figure 4d). The mutant is also about 7 days delayed in flowering relative to wild-type plants grown under similar conditions (Figure 4e). Unlike many dwarf mutants, xbat32-1 was able to eventually achieve a plant size comparable with Col-0 (Figure 4f), indicating that the developmental process in the mutant is delayed.

The most notable phenotype of the xbat32-1 mutant is the deficiency in lateral root production. Relative to Col-0, the mutant plants had significantly fewer lateral roots when grown on half-strength MS media for 11 days (Figure 4g). Due to the reduction in lateral roots, the xbat32-1 plants established a poor root system at the 35th day and beyond (Figure 4h). The primary roots and root hairs showed no noteworthy difference between xbat32-1 and Col-0. These results suggest that XBAT32 is involved in lateral root development.

The xbat32-1 mutant is severely impaired in the expression of the cell division marker CYCB1;1::GUS

As described above, the XBAT32 gene is highly expressed in the lateral root initiation zone. This suggests that XBAT32 may be involved in the initiation process of lateral root development. The mitotic cyclin gene CYCB1;1 is specifically expressed in actively dividing cells during the G2–M phase of the cell cycle (Ferreira et al., 1994). This gene has also been used as a marker for lateral root initiation (DiDonato et al., 2004). We transformed xbat32-1 mutant plants with CYCB1;1::GUS to address whether XBAT32 is involved in lateral root initiation. The roots of wild-type Col-0 and xbat32-1 plants expressing CYCB1;1::GUS were assayed for GUS activity. The expression of CYCB1;1::GUS was observed in the root tip of both wild-type Col-0 and xbat32-1 plants, indicating that the reporter gene is expressed in these two lines (Figure 5a,b). However, there is a severe reduction in the number of GUS spots distal to the primary root tip in the xbat32-1 mutant when compared with wild-type Col-0 (Figure 5a). Moreover, this GUS activity observed in the mutant plant is also lower than that of wild-type Col-0 (Figure 5c). These results suggest that XBAT32 is involved in lateral root initiation.

Figure 5.

The expression of the mitotic cyclin CYCB1;1::GUS is severely impaired during lateral root initiation in xbat32-1 plants.
(a) Wild-type Col-0 and xbat32-1 seedlings carrying CYCB1;1::GUS were grown for 14 days on half-strength MS media and then assayed for GUS activity for 12 h at 37°C. Arrows indicate the GUS-positive zones. A higher magnification of the primary root tips of wild-type Col-0 and xbat32-1 plants are shown as insets.
(b) A close-up of the indicated areas in (a).
(c) Higher magnification of circled areas in (b). Scale bars, 100 μm.

Exogenous auxin can partially restore the lateral root deficiency in xbat32-1 mutant plants

Lateral root production is increased by exogenous NAA or IAA treatment (Laskowski et al., 1995). We grew xbat32-1 plants on media containing various concentrations of NAA or IAA to address whether auxin can rescue the lateral root defect in the mutant. Five-day-old seedlings were transferred from half-strength MS media to the same media supplemented with NAA or IAA. Five days later, we observed that the number of lateral roots in both mutant and Col-0 was increased (Figure 6). However, the levels of lateral root induction in xbat32-1 seedlings were higher than those in Col-0 plants. For instance, 2 μm of NAA caused a 16-fold increase in lateral roots in the mutant line, whereas the same treatment only increased the number of lateral roots by eightfold in the Col-0 control. Similar results were obtained when the plants were treated with IAA (data not shown). These results indicate that auxin can partially restore the lateral root deficiency caused by mutation in XBAT32.

Figure 6.

Auxin can partially rescue lateral root defects in xbat32-1 mutant plants. Five-day-old wild-type Col-0 (closed bars) and xbat32-1 (open bars) plants were grown on half-strength MS media supplemented with the indicated concentrations of naphthalene-1-acetic acid (NAA) for an additional 5 days. Lateral roots were scored under a microscope. Each data point represents the average number of lateral roots counted from 20 primary roots. Error bars indicate standard deviations.

xbat32-1 is sensitive to the auxin-transport inhibitor CPD

Auxin-transport inhibitors (NPA and CPD) can block polar auxin transport. Wild-type Col-0 seedlings grown on media containing these inhibitors exhibit inhibition of lateral root formation and primary root elongation, and loss of root gravitropism (Ruegger et al., 1997). The tir1-1 mutant was found in a screening for resistance to the inhibition effects of NPA and CPD (Ruegger et al., 1998). To determine whether xbat32-1 is resistant to an auxin-transport inhibitor, we grew plants in the presence of 5 μm CPD. Similar to Col-0, the xbat32-1 mutant seedlings were sensitive to the CPD-mediated root inhibition and a swelling was observed in the area close to the root tip (Figure 7). As a control, the tir1-1 mutant displayed a significant increase in the growth of the primary root, and no swelling in the root tip when grown under similar conditions. These results indicate that the mutation in xbat32-1 does not affect the inhibition of primary root elongation mediated by the auxin-transport inhibitor CPD.

Figure 7.

The xbat32-1 mutant is sensitive to the auxin-transport inhibitor 2-carboxyphenyl-3-phenylpropane-1,2-dione (CPD). Seedlings were grown on half-strength MS media in the presence or absence of 5 μm CPD for 8 days. The length of primary roots was measured (n = 30) (upper). Photograph showing swelling of root tips were taken 14 days after exposure to 5 μm CPD (lower).

The xbat32-1tir1-1 double mutant displays a defect in lateral root production similar to xbat32-1

To determine the relationship between XBAT32 and TIR1, a double mutant between tir1-1 and xbat32-1 was created by crossing homozygous lines of these two mutants. tir1-1 is resistant to the inhibition of the CPD-mediated primary root elongation, whereas xbat32-1 or wild-type Col-0 seedlings are sensitive to CPD. The F2 progeny produced from this cross were subjected to the CPD assay to identify the homozygous tir1-1 mutant. F3 progeny of CPD-resistant F2 lines were analyzed by PCR to identify the xbat32-1tir1-1 double mutant. As shown in Figure 8, the lateral root production in the double mutant is less than that in tir1-1 or wild-type Col-0, but similar to that in xbat32-1.

Figure 8.

The xbat32-1tir1-1 double mutant exhibits a similar defect in lateral root production as xbat32-1. Seedlings from the indicated lines were grown on half-strength MS media. The number of lateral roots and the length of primary root were measured 10 days after germination.

The wild-type XBAT32 gene can restore lateral root production in xbat32-1 plants

To confirm that the deficiency of lateral roots is due to the mutation in XBAT32, we cloned the entire genomic region of this gene from Col-0. A construct encompassing 1830 bp of upstream sequence, 2357 bp of coding and intron regions, and 919 bp of 3′ flanking sequence was transformed into the xbat32-1 mutant plants. Two independent transgenic lines, T14 and T23, were further characterized. The T2 generation of these lines was grown on half-strength MS media and lateral root production was determined. The results indicate that both of these two lines were complemented for the lateral root defect in xbat32-1 (Figure 9). In addition, other shoot phenotypes observed in the xbat32-1 mutant were also complemented in the transgenic lines (data not shown). These results confirm that the wild-type XBAT32 gene is required for lateral root development.

Figure 9.

Complementation of xbat32-1 in lateral root production by the cloned XBAT32 gene. T2 seeds of the transgenic line T23 were germinated on half-strength MS media. The lateral roots were scored 11 days after germination. The number of lateral roots per centimeter of primary root was determined for 18 seedlings: Col-0, 3.23 + 1.00; T23, 3.24 + 0.86; xbat32-1, 0.16 + 0.32. Similar results were obtained from the independent transgenic line T14.


The XBAT32 gene is involved in lateral root development in Arabidopsis. An XBAT32 null mutant exhibits severely impaired lateral root emergence. By using the cell division marker CYCB1;1::GUS, we showed that the expression of this gene is largely reduced in the pericycle cells of the mutant plants, indicating that the xylem-adjacent pericycle cells of the xbat32-1 plants are defective in cell divisions that are required for lateral root initiation. In addition, mutation of XBAT32 is also responsible for the delayed development of several above-ground organs. These deficiencies could be due to the poor root system produced in the mutant plants. Alternatively, XBAT32 may have other functions in regulating the developmental process of these organs. Consistent with this idea, the expression of the XBAT32 gene was found in roots, leaves, stem, and anthers.

Auxin plays an important role in lateral root production. Previous evidence supports that the auxin required for lateral root initiation is produced by light-dependent hydrolysis of auxin conjugates in the root tip and by biosynthesis of auxin in the developing leaves (Bhalerao et al., 2002). Auxin transport from the source tissues to roots becomes critical for lateral root development. The auxin-transport inhibitor NPA can arrest lateral root development (Casimiro et al., 2001). Co-cultivation of NPA-treated roots with NAA partially rescues lateral root formation (Casimiro et al., 2001). Moreover, NAA treatment can completely rescue the lateral root defects in the auxin transport mutant aux1 (Marchant et al., 2002). We found that auxin can induce the expression of the XBAT32 gene. Treatment of xbat32-1 mutant plants with exogenous auxin also partially rescues the lateral root defect, with restoration occurring along the entire primary root. In addition, we observed that the auxin transport inhibitor CPD causes a similar inhibition of primary root elongation in both xbat32-1 and wild-type Col-0 plants. Such a inhibition is thought to be the consequence of IAA accumulation in root tip when auxin transport is blocked (Ruegger et al., 1997). Thus, it is unlikely that xbat32-1 is deficient in auxin production. We propose that XBAT32 functions in auxin transport and mutation in the XBAT32 gene causes auxin to be suboptimal for lateral root initiation. This hypothesis is consistent with the observation that the XBAT32 gene is highly expressed in the vascular system of primary roots.

The bacterially expressed XBAT32 fusion protein is capable of autoubiquitination, indicating that XBAT32 is an active E3. Although the substrate for XBAT32 remains to be determined, the presence of ankyrin repeats at the N-terminal half suggests that XBAT32 may directly interact with its substrate through the ankyrin domain. Another potential region for binding a substrate is the C-terminus of XBAT32 consisting of 137 aa. The large C-terminal tail may contain a novel protein–protein interaction domain or small motifs that can be recognized by the substrate. Alternatively, autoubiquitination of XBAT32 may play a role in transducing auxin signals. Autoubiquitination of the RING finger-containing protein TRAF6 activates downstream kinases in the animal immune response (Wang et al., 2001). The third possibility is that XBAT32 may be involved in a ubiquitin ligase protein complex. The Arabidopsis RING finger-containing protein COP1 can autoubiquitinate in vitro (Saijo et al., 2003). In addition to the RING finger motif, COP1 consists of a coiled-coil domain and seven WD-40 repeats (Deng et al., 1992). Interestingly, the human counterpart of COP1 was identified in a CUL4A-containing ubiquitin ligase complex (Wertz et al., 2004).

Three E3s with a distinct structure have been identified as being involved in Arabidopsis root development. The RING finger-containing E3 protein SINAT5 acts as a negative regulator in the auxin-mediated production of lateral roots (Xie et al., 2002). Both TIR1 and XBAT32 are positively involved in lateral root development, but differ from one another in several aspects. First, the xbat32-1 mutant produces significantly fewer lateral roots than the tir1-1 plants (data not shown). Secondly, unlike TIR1, an F-box protein functioning as part of the SCFTIR1 complex (Gray et al., 2001), XBAT32 is a RING finger-containing enzyme capable of autoubiquitination. Thirdly, XBAT32 is mainly expressed in the vascular system of primary roots with accumulation in the areas where lateral roots are initiated, but there is no observable expression in newly formed lateral roots or apical meristem of the primary root. In contrast, high levels of TIR1 expression are observed in the primary root apical meristem and in developing lateral root primordia (Ruegger et al., 1998). Fourthly, xbat32-1 exhibits the normal growth-inhibiting effects induced by the auxin-transport inhibitor CPD on primary root elongation, but the tir1-1 mutant is resistant to CPD (Ruegger et al., 1998). Finally, TIR1 is involved in auxin response, whereas XBAT32 might function in auxin transport. Thus, XBAT32 represents a novel E3 involved in lateral root initiation.

XBAT32 was identified as one of the five Arabidopsis proteins structurally similar to the rice E3 enzyme XB3, which can interact with the kinase domain of the receptor-like kinase XA21 involved in rice innate immunity (L.-Y. Pi and W.-Y. Song, unpublished data). In this study, we showed that XBAT32 is involved in auxin-mediated lateral root production. Interestingly, the Arabidopsis defense gene SGT1b, essential for the action of multiple disease resistance genes (Austin et al., 2002; Azevedo et al., 2002; Tor et al., 2002), is required for SCFTIR1-mediated auxin response (Gray et al., 2003). Mutation of SGT1b results in a number of defects including lateral root development. Therefore, the plant signal transduction pathways, used to perceive hormones to control developmental processes, share some common elements with the disease resistance pathways. Another example supporting this notion is the tomato resistance gene I-2. I-2 is expressed in the vascular system with accumulation at the base of lateral root primordia and is involved in the development of vascular tissue (Mes et al., 2000). Thus, it would be interesting to determine whether XBAT32 or its homologs also play a role in Arabidopsis disease resistance.

Experimental procedures

Plant materials

The xbat32-1 mutant line was ordered from the Arabidopsis Biological Resource Center. The accession number for the mutant line is Salk_015002. Insertion mutant information was obtained from the Salk Institute Genomic Analysis Laboratory (SIGnAL) website at The tir1-1 mutant seeds were a gift from M. Estelle.

Autoubiquitination assays

The coding region of XBAT32 was PCR amplified with the primers: 5′ GGATCCATGAGGTTTCTAAGCCTCGTCGGA 3′ and 5′ GGATCCTGAGTTAGCAAGCACTTCCACCGG 3′, cloned into the BamHI site of pMAL-c2X (New England Biolabs, Beverly, MA, USA), and verified by sequencing. Site-directed mutagenesis was carried out to create MBP-XBAT32 (C321A) using the primers 5′ CCATGTCGTGACCCTGCAGCCGTTTGTTTGGAAAG 3′ and 5′ CCATGTCGTGACCCTGCAGCCGTTTGTTTGGAAAG 3′.

Autoubiquitination of MBP-XBAT32 was essentially as described (Lorick et al., 1999) except for the following modifications: GST-ubiquitin was 32P-labeled by protein kinase A as described (Scheffner et al., 1994). After cleavage with thrombin and depletion with benzamidine Sepharose, the labeled ubiquitin (105 cpm) was incubated with MBP-XBAT32 or MBP or MBP-XBAT32(C321A), 4 mm ATP, and 20 ng each of bacterially expressed wheat E1 and human E2 UbcH5B lysates in the ubiquitination buffer [50 mm Tris–HCl (pH 7.5), 2.5 mm MgCl2, 0.5 mm DTT] for 90 min at 30°C. The samples were mixed with β-mercaptoethanol-containing loading buffer and separated by 8% SDS-PAGE, stained, and exposed.

PCR, Southern and Northern analyses of the xbat32-1 mutant line

Genomic DNA was isolated from the mature leaves of the xbat32-1 mutant plants for PCR with the primer sets: pROK2 LBb1 5′GCGTGGACCGCTTGCTGCAACT 3′/XBAT32U-1 5′ GACCTAGAATCGAATCCAGCAACTG 3′ and pROK2 LBb1/XBAT32D-1 5′ CAGAGACGGTGATTTACAGGAAGCT 3′. The amplified products were cloned into the pGEM-T vector (Promega, Madison, WI, USA) for sequencing.

Southern and Northern blot analyses were carried out using standard procedures. The XBAT32-specific probe was prepared by PCR using the primers: 5′GGATCCTCTCTGTCGCAATGGCA3′ and 5′GAGCTCGCAAGCACTTCCACCGG3′.

Arabidopsis transformation

For generation of transgenic lines expressing XBAT32::GUS, a 1830-bp sequence containing the 5′-genomic region, relative to the +1 site, was PCR-amplified using the primers 5′GAGCTCCTTTTGGGACATAGGCTTATGCAAG 3′ and 5′GAGCTCACTAGTCGTTTTACGATACAAAATTGAAATTTTG 3′ and cloned into the pGEM-T vector for sequencing. The fragment was then fused to the GUS reporter gene and subcloned into the binary vector pMON11054. The resulting construct was transformed into the Agrobacterium strain ABI by electroporation. Arabidopsis transformation was performed using standard vacuum infiltration. Transformants were selected on MS media supplemented with 6.3 μg ml−1 of glyphosate (Sigma, St Louis, MO, USA). GUS staining was carried out at 37°C using the buffer [Na2HPO4 (pH 7.0) (25 mm), NaH2PO4 (pH 7.0) (25 mm), K3Fe(CN)6 (0.5 mm), K4Fe(CN)6 (0.5 mm), Triton X-100 (0.1%), Methanol (20%), X-Gluc (0.5 mg ml−1) (Gold Bio Technology Inc., St. Louis, MO, USA)]. The stained tissues were cleared in 70% ethanol for 24 h and photographed under dissecting scope and microscopes using SPOT Insight v 3.5 (Diagnostic Instruments Inc., Sterling Heights, MI, USA).

For lateral root initiation, the pBI101.3 plasmid carrying cyc1At::CDBGUS (CYCB1;1::GUS) was transformed into xbat32-1 plants using the procedure described above. Transformants were selected on half-strength MS media containing 50 μg ml−1 kanamycin (Sigma).

For complementation analysis, a 5.1-kb genomic fragment containing the XBAT32 gene was PCR-amplified from Col-0 using the primers 5′GAGCTCCTTTTGGGACATAGGCTTATGCAAG3′ and 5′GAGCTCTTTTGTTTTACGATTTCTATGGCTTCCA3′, verified by sequencing, and cloned into the SacI site of the pFMVnos vector to create pFMV-XBAT32G. The XBAT32G was then subcloned into pMON11054 to create pMON11054-XBAT32G. The resulting plasmid was transformed into the xbat32-1 mutant plants using the procedure described above.

NAA and CPD experiments

For the NAA induction of XBAT32::GUS, 3-day-old seedlings were transferred to half-strength MS media supplemented with NAA for 24 h and then assayed for GUS activity for 1 h at 37°C. Photographs were taken under a microscope using SPOT Insight v 3.5 (Diagnostic Instruments Inc.).

For induction of lateral roots, 5-day-old seedlings were transferred to half-strength MS media supplemented with NAA. The number of lateral roots was scored 5 days later with a microscope.

For CPD treatment, seeds were germinated on half-strength MS media containing CPD for 8 days. The length of the primary roots was measured on day 8 and photographed on day 14.


We thank Drs Zhiyong Wang and Harry Klee, and T.A. Davoli for critical reading of the manuscript and invaluable comments on the work; M. Manak and P. Moussatche for technical assistance, ABRC for the seeds of Salk_015002; M. Estelle for seeds of the tir1-1 mutant; H. Klee for the vectors pFMVnos and pMON11054, and the Agrobacterium tumefaciens strain ABI; A.M. Weissman for UbcH5B; R.D. Vierstra for the Wheat E1; P.M. Howley for GST-ubiquitin; J. L. Celenza for the pBI101.3 vector carrying cyc1At::CDBGUS; the Salk Institute Genomic Analysis Laboratory for providing the sequence-indexed Arabidopsis T-DNA insertion mutant. This research was supported by the Florida Agricultural Experiment Station and grants from NSF (MCB-0080155) and USDA to W.-Y. S. Funding for the SIGnAL-indexed insertion mutant collection was provided by the NSF. This work was approved for publication as Journal Series No. R-10441.