Ectopic Overexpression of an AUXIN/INDOLE-3-ACETIC ACID (Aux/IAA) Gene OsIAA4 in Rice Induces Morphological Changes and Reduces Responsiveness to Auxin

Auxin has pleiotropic effects on plant growth and development. AUXIN/INDOLE-3-ACETIC ACID (Aux/IAA) proteins are short-lived transcriptional regulators that mediate auxin responses through interaction with an auxin receptor, the F-box protein transport inhibitor response 1 (TIR1). Most functions of Aux/IAA proteins have been identified in Arabidopsis by studying the gain-of-function mutants in domain II. In this study, we isolated and identified an Aux/IAA protein gene from rice, OsIAA4, whose protein contains a dominant mutation-type domain II. OsIAA4 has very low expression in the entire life cycle of rice. OsIAA4-overexpressing rice plants show dwarfism, increased tiller angles, reduced gravity response, and are less sensitive to synthetic auxin 2,4-dichlorophenoxyacetic acid (2,4-D).


Introduction
It is known that auxin plays a very important role in a wide variety of plant developmental and physiological processes [1]. Several genes involved in the regulation of auxin dependent transcription have been identified, and of this Aux/IAA protein family members have been studied most thoroughly. Aux/IAAs are short-lived transcription factors comprising four highly conserved domains, known as OPEN ACCESS domain I, II, III, and IV. Domain I contains conserved motif "LxLxL" and acts as a strong transcriptional repressor [2]. Domain II contains a core sequence "GWPPV" responsible for the rapid degradation of Aux/IAA proteins by interacting with a component of the ubiquitin-proteasome protein degradation pathway, which is stimulated by auxin [3,4]. Domains III and IV, shared by ARF proteins and Aux/IAA proteins, mediate homo-and hetero-dimerization among members of these two protein families [5,6].
Several Aux/IAA proteins with a mutation in domain II have been identified in Arabidopsis, including IAA1/AXR5, IAA3/SHY2, IAA6/SHY1, IAA7/AXR2, IAA12/BDL, IAA14/SLR, IAA17/AXR3, IAA18, IAA19/MSG2, and IAA28 [7][8][9][10][11][12][13][14][15][16][17]. Several Aux/IAA genes were also identified in rice. mOsIAA3, a mutant with an amino acid changed in domain II, showed a gain-of-function phenotype like insensitive to auxin and gravitropic stimuli, exhibited short leaf blades, reduced crown root formation, and abnormal leaf formation [18]. OsIAA13 is involved in lateral root initiation [19]. OsIAA11 affected lateral root development [20]. OsIAA23 is involved in postembryonic maintenance of quiescent center [21]. Our previous analyses of the expression profile of the OsIAA gene family in the developmental process and responses to phytohormones and stress treatments showed that most OsIAAs are involved in multiple hormone and abiotic stress responses [22]. OsIAA1 is also involved in plant morphogenesis mediated by auxin and brassinosteroid (BR) hormone signaling pathway [23]. Unlike most OsIAA genes, which have high expression levels or organ/tissue expression patterns in the entire life cycle of rice, and are rapidly induced by auxin, the expression of OsIAA4 (LOC_Os01g18360) is relatively stable and has a very low expression level in the whole life cycle of rice [22]. In this study, we further characterized OsIAA4, which has a partially conserved domain II, and contains an amino acid substitution that could cause a dominant mutation in Aux/IAA genes. Transgenic rice plants of ectopically overexpressing OsIAA4 exhibited dwarfism, increased tiller angle, resistant to synthetic auxinic herbicide 2,4-D and decreased gravity response.

Identification of OsIAA4
The OsIAA gene family is involved in various developmental processes and responds to distinct phytohormones and abiotic stresses. Among them, OsIAA4 and OsIAA8 belong to one category with the expression level undetectable in almost all the organs/tissues [22]. Phylogenetic analysis of the OsIAA gene family shows OsIAA4, OsIAA8, and OsIAA10 belong to one subclade, of which no further investigation into their functions have been reported (Figure 1a). Expression data from the CREB database (http://crep.ncpgr.cn/) [24] indicated OsIAA4 has very low expression in the callus, stem, and endosperm, and no expression can be detected in any other organs/tissues. We were only able to obtain the full length cDNA of OsIAA4 from the callus by RT-PCR, which confirmed the CREB data (data not shown). Sequencing results show OsIAA4 encodes a 204-amino acid protein which contains four domains. In Arabidopsis, shy2-3 and iaa18-1 have a dominant mutation-type domain II that contains a G-to-E substitution; and AtIAA31 contains a G-to-D substitution in domain II that would cause a dominant mutation of Aux/IAA genes. In rice, OsIAA4 contains the same domain II as AtIAA31, whose core amino-acid motif is "DWPPV/I", while the core motif of domain II of the canonical Aux/IAA protein is "GWPPV", thus, OsIAA4 can be regarded as a dominant-mutation type protein (Figure 1b). Red letters indicate differences in domain II between dominant mutants (shy2-3 and iaa18-1) or non-canonical Aux/IAA (AtIAA31 and OsIAA4) and canonical Aux/IAA proteins.

Morphological Changes of OsIAA4-Overexpressed Rice Plants
To characterize the function of OsIAA4 in rice, the full-length cDNA of OsIAA4 was cloned into pCAMBIA1301U driven by a maize Ubiquitin promoter ( Figure 2a) and transformed into Zhonghua 11 using the Agrobacterium method. In total, 20 independent OsIAA4 transgenic plants were generated. Leaves of all the OsIAA4 transgenic plants were collected for detecting the expression level of OsIAA4. RNA-blot analyses indicated that 30% of OsIAA4 transgenic plants (6 lines) had higher expression levels compared to the WT (Figure 2b). To investigate the phenotypes of OsIAA4-overexpressing transgenic plants, four independent overexpression lines (O12, O14, O15, and O19) and one non-overexpressing line (O10) were grown in the field. At the seedling stage, no visible phenotype change was observed ( Figure 2d). However, at the vegetative stage, the overexpression lines of OsIAA4 showed significantly reduced plant height and increased tiller angle (Figure 2c,e). The height of OsIAA4-overexpressing plants is around 68 ± 5 cm while the control is around 85 cm ( Figure 2f). The maximum tiller angle (angle between the main culm and its side tillers) in OsIAA4-overexpressing plants is more than 30 degrees but only 10 degrees in control plants ( Figure 2g). The GUS gene-specific primers were used to check the positive transgenic plants and the PCR result indicated that the phenotypic change was co-segregated with GUS gene expression (data not shown). This result was obtained in T1 generation and similar result was repeated in T2 generation.

OsIAA4-Overexpressing Plants Show Less Sensitive to Auxin
To assess auxin responsiveness in OsIAA4-overexpressing plants, we performed auxin-responsive root elongation assays. Positive transgenic seeds were selected by germinating on hygromycin-containing Murashige and Skoog (MS) medium. Because of their faster growth rate, non-overexpressing transgenic and WT seeds were germinated on the MS medium one day later so that all germinated seeds were of a similar degree of vigor (with shoot length about 2 mm) for auxin treatment. Seedlings of similar height were transferred to MS medium containing 0.1 μM 2,4-D, and germinated seeds with the same vigor were transplanted onto the normal MS medium as a control. Two weeks after transplantation, there was no significant difference in plant morphology between the over-  We also compared the auxin sensitivity between OsIAA1-overexperssing and OsIAA4-overexpressing plants by comparison of the seminal root length. Result shows OsIAA4 transgenic plants displayed significantly stronger auxin insensitivity phenotype than OsIAA1 transgenic plants (Figure 4). It is known that auxin signaling transduction has a negative feedback loop and Aux/IAA proteins function as a primary negative feedback regulator to auxin [25,26]. To test the role of non-canonical OsIAA4 in this feedback pathway, the expression responses of five OsIAA genes, including OsIAA8 (belonging to the same subclade as OsIAA4, which lacks domain II as do AtIAA20, 30, 32, 33, and AtIAA34), OsIAA9 (showing a strong auxin up-regulation), and a pair of sister genes OsIAA6 and OsIAA18 which were significantly induced by drought [22] to auxin treatment were checked. We found that the expression responses of the OsIAA genes were not significantly changed in OsIAA4-overexpressing plants ( Figure 5), suggesting OsIAA4 is not the main auxin negative feedback regulator.

Overexpression of OsIAA4 Impaired Gravity Response in Shoot
Gravitropic response is a typical auxin-related phenotype. Several Aux/IAA mutants display agravitropism in the hypocotyls and/or roots in Arabidopsis. In rice lazy1 and lpa1 mutant exhibits a tiller-spreading phenotype resulting from reduced shoot gravitropism [27][28][29]. Since OsIAA4 transgenic plant shows typical auxin-related response and tiller-spreading phenotype, we assessed the gravitropic response of OsIAA4-overexpressing seedlings. We grew transgenic and wild-type seedlings vertically for one day, and then reoriented them by 90 degrees (horizontally) for an additional two days and measured the angle of shoots. We found that OsIAA4 transgenic plants displayed impaired gravitropic responses compared to the WT (Figure 6a). The shoot angles of transgenic plants (25 degrees) were significantly smaller than those of the WT plants (35 degrees) on the second day of reorienting (Figure 6b).

Discussion
It is well known that rapid turnover of Aux/IAA protein is required for normal auxin response and domain II contains a "qvVGWPPvrsyRkN" conserved motif; in particular, the conserved "GWPP" has been linked to rapid degradation in canonical Aux/IAA family members. Mutation in domain II results in relative stabilization of the mutated proteins. Several gain-of-function mutants with altered auxin response or morphology have been identified in Aux/IAA genes in Arabidopsis. These auxin responses and morphologies include phototropism/gravitropism, root formation, apical dominance, stem/hypocotyl elongation, and leaf expansion. In an earlier OsIAA family analysis, OsIAA4 was found to belong to one small subgroup that has very low expression levels in the whole life cycle of rice [22]. Phylogenetic tree analysis indicated OsIAA4, OsIAA8, and OsIAA10 form one small subclade. Predicted protein sequence showed that the conserved G in domain II is replaced to D in OsIAA4. Ectopic overexpression of OsIAA4 shows dwarfism, increased tiller angle, insensitivity to the inhibitory effect of auxin on root elongation, and impaired gravity response in shoots. IAA31 has the same domain II as OsIAA4 which cause a dominant mutation of Aux/IAA gene in Arabidopsis.
Overexpression of IAA31 in Arabidopsis by using cauliflower mosaic virus 35S promoter exhibited auxin-related aberrant phenotypes including smaller plant size, longer lateral stems than primary stems, and defects in gravitropic response in hypocotyls [30]. All these results suggested that ectopic overexpression of dominant mutation-type Aux/IAA proteins cause auxin-insensitive related phenotypes.
Auxin plays a crucial role in root development including regulation of the main root growth and the number of lateral, adventitious roots in Arabidopsis [8,12,13,17]. Auxin-insensitive phenotypes such as dwarfism, decreased gravitropic response, and suppression of root growth were observed by studying mutants in Arabidopsis. These abnormal phenotypes were also found in OsIAA1-overexpressing and mOsIAA3 plants [18,23]. In OsIAA4-overexpressing plants the growth of the seminal root was insensitive to 2,4-D and the number of crown root was decreased under 2,4-D treatment, suggesting auxin contributes to the root development in rice similar to its effect in Arabidopsis.
Aux/IAA is a large gene family with more than 30 members in rice. In earlier work we identified several T-DNA insertion mutants of OsIAAs, and no obvious changes in morphology or hormone responses were observed, implying functional redundancy [22]. The protein sequence of OsIAA4 shows it has a dominant mutation-type domain II, which contains "DWPPI" instead of "GWPPV/I". As no T-DNA insertion mutant of OsIAA4 was identified, we made OsIAA4-RNAi transgenic plants, but found no visible phenotypic change (data not shown). Over-expression of OsIAA4 shows a similar auxin response to over-expression of OsIAA1. Both OsIAA4 and OsIAA1 exhibited decreased auxin response, but the OsIAA4 transgenic plants displayed a stronger auxin insensitivity phenotype, possibly associated with the longer half-life of OsIAA4; OsIAA1 contains intact domain II while OsIAA4 contains a dominant mutation-type domain II. Instead of an increased lamina joint angle, as in the OsIAA1 transgenic plant, OsIAA4 exhibits increased tiller angle, suggesting each OsIAA may have a specific function in its developmental process. These results together suggest OsIAA1 and OsIAA4 are functionally redundant and specific.
Rice is one of the most important food crops and is a model monocot. Rice yield is mainly modulated by its architecture [31,32], which is defined by tiller number and angle, internodes elongation, panicle morphology and leaf angle [33,34]. Tiller angle is defined as the angle between the main culm and its side tillers, and is critical for dense planting because a wide tiller angle will increase leaf shade and decrease photosynthesis efficiency, whereas a narrow tiller angle is more efficient in plant architecture [35]. Leaf angle is the inclination between the leaf blade and vertical culm which is mainly controlled by the leaf lamina joint [36]. Over-expression of OsIAA1 changed the leaf angle and over-expression of OsIAA4 changed the tiller angle, which may be useful for molecular breeders to achieve ideal plant architecture to improve grain yield. Rice plants develop tillers from the axillary buds. The roles of auxin in the development of axillary buds in Arabidopsis imply that auxin and its signaling may be the key regulator in the development of tillers. Transgenic plants with a reduction of OsPIN1 gene expression display an increase in tiller number as well as tiller angle [37]. It is known that auxin transport is mainly controlled by auxin influx (AUX1) and the PIN-FORMED (PIN) efflux carriers and the MULTIDRUG RESISTANCE/P-GLYCOPROTEIN (PGP) class of ATP-binding cassette auxin transporters. Aux/IAAs are early auxin responsive genes. It is not surprising that OsIAA4-overexpressing plants have a similar phenotype to OsPIN1 down-regulated plants. Although two different types of auxin related genes were found to be involved in rice tiller angle (OsPIN1 and OsIAA4), the underlying molecular mechanism of how these two types of genes control rice tiller development is still unknown. Further work should focus on deciphering the precise molecular mechanism on how OsPIN1 down-regulated plants show similar phenotypes to OsIAA4 overexpression plants.

Construct and Rice Transformation
The full-length cDNA of OsIAA4 was amplified from Minghui 63 (Oryza sativa L. ssp. indica) callus using primer: 5'-ATAGGTACCTATGGAGGAGTGCAAGGG-3' and 5'-TATGGATCCTTCTG CTATACGGTAGGTAG-3'. Sequence was obtained by RT-PCR and confirmed by sequencing, and digested by KpnI and BamHI and then inserted into pCAMBIA1301U under the control of a maize ubiquitin promoter, and transferred into Zhonghua 11, japonica rice that can be easily transformed by Agrobacterium-mediated transformation method.

Northern Blot and Real-Time PCR
Total RNA was isolated from T0 transgenic plants with Trizol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Total RNA was separated on a 1.2% agarose gel containing 2% formaldehyde and then transferred onto a nylon membrane by capillary method. RNA gel blots were hybridized with 32 P-dCTP-labeled gene-specific probes over-night using Perfect-HYB Plus buffer (Sigma, St. Louis, MO, USA) at 65 °C. Blots were washed three times (twice with 2× SSC/0.1% SDS for 10 min and once with 0.5× SSC/0.1% SDS for 5 min) at 65 °C, and then subjected to radiophotography.
For real-time PCR analysis, first-strand cDNAs were synthesized from DNaseI-treated total RNA using Superscript II reverse transcriptase (Invitrogen) according to the manufacturer's instructions. Real-time PCR was performed in an optical 96-well plate with an ABI PRISM 7500 real-time PCR system (Applied Biosystems, Foster City, CA, USA). Each reaction contained 12.5 μL of SYBR Green Master Mix reagent (Applied Biosystems), 4.0 μL of cDNA samples, and 200 nM gene-specific primers in a final volume of 25 μL. The thermal cycle used was as follows: 95 °C for 3 min; 45 cycles at 95 °C for 30 s, 60 °C for 30 s, and 72 °C for 1 min. The Osactin gene was amplified as an internal control to quantify the relative amounts of cDNA. The primers used for real-time PCR are listed in Table 1.

Plant Growth and Treatments
Transgenic seeds were germinated on a Murashige and Skoog (MS) medium with 50 mg/L hygromycin and wild-type (WT) seeds were germinated on a MS medium (usually one day later) without hygromycin at 26 °C. Plantlets with same vigor (shoot length 2-3 mm) were transferred onto the MS medium containing different concentrations of 2,4-D and transferred to a growth chamber at 26 °C with a 14-h-light/10-h-dark cycle. Plants were measured for root length and crown root numbers were counted 12 days after transplanting. For the gravitropic response experiment: transgenic and wild-type seeds were germinated on a MS medium, three-day-old plants were grown vertically for one day and reoriented by 90° and grown for another two days. The shoot angle was measured. To check the expression levels of auxin responsive genes in the OsIAA4 overexpression plants, 8-day-old transgenic and wild type (WT) seedlings were treated with 20 μM 2,4-D and sampled at 3 h.

Conclusions
In conclusion, our results indicate OsIAA4 is a low expression gene, encodes an Aux/IAA protein containing a dominant mutation-type domain II, and OsIAA4 overexpression lines have typical auxin related characterization. Overexpression of OsIAA4 resulted in dwarfism, increased tiller angle and decreased gravity response.