Characterization of MORE AXILLARY GROWTH Genes in Populus

Background Strigolactones are a new class of plant hormones that play a key role in regulating shoot branching. Studies of branching mutants in Arabidopsis, pea, rice and petunia have identified several key genes involved in strigolactone biosynthesis or signaling pathway. In the model plant Arabidopsis, MORE AXILLARY GROWTH1 (MAX1), MAX2, MAX3 and MAX4 are four founding members of strigolactone pathway genes. However, little is known about the strigolactone pathway genes in the woody perennial plants. Methodology/Principal Finding Here we report the identification of MAX homologues in the woody model plant Populus trichocarpa. We identified the sequence homologues for each MAX protein in P. trichocarpa. Gene expression analysis revealed that Populus MAX paralogous genes are differentially expressed across various tissues and organs. Furthermore, we showed that Populus MAX genes could complement or partially complement the shoot branching phenotypes of the corresponding Arabidopsis max mutants. Conclusion/Significance This study provides genetic evidence that strigolactone pathway genes are likely conserved in the woody perennial plants and lays a foundation for further characterization of strigolactone pathway and its functions in the woody perennial plants.


Introduction
Plant architecture plays a major role in determining photosynthetic light use efficiency and biomass yield. Optimal plant architecture is critical for achieving maximum carbon capture per unit land area especially when plants are grown in high-density stands. One recent breakthrough in the field of plant biology was the discovery of strigolactones (SLs) as a new class of plant hormones controlling shoot branching [1,2]. More importantly, the synthesis of SLs in plants is regulated by the nutrient availability, particularly Pi deficiency, in soil and SLs exuded by roots serve as host recognition signals for symbiotic fungi [2][3][4][5][6][7][8][9][10][11][12][13][14]. Therefore, SLs are viewed as integrative signaling molecules that can couple nutrient availability and microbial symbiosis to the control of plant architecture and productivity. This breakthrough offers a new opportunity to dissect the interaction between plants and their environmental systems and to link this interaction to the control of plant architecture and productivity. Substantial progress has been made in the last decade to understand the biosynthesis, signal transduction and physiological functions of SLs (reviewed by [12,[15][16][17][18][19][20][21][22][23][24][25][26][27][28]). MORE AXILLARY GROWTH1 (MAX1), MAX2, MAX3 and MAX4 are four founding members of strigolactone pathway genes in the model plant Arabidopsis [29][30][31][32]. While MAX1, MAX3 and MAX4 are key genes involved in SL biosynthesis, MAX2 is a key gene involved in SL signaling. MAX1 encodes a cytochrome P450 monooxygenase [30]. MAX3 and MAX4 encode two carotenoid cleavage dioxygenases (CCD), CCD7 and CCD8, respectively [29,31]. MAX2 encodes an F-box leucine-rich protein [32]. Lossof-function mutations in each of these four MAX genes resulted in increased shoot branching [29][30][31][32]. Orthologues for MAX genes have been identified and corresponding mutants have been characterized in pea (ramosus, rms), rice (dwarf, d; high tillering dwarf, htd) and petunia (decreased apical dominance, dad) (reviewed by [15][16][17][19][20][21]23]). In addition to MAX genes, several other genes involving in SL biosynthesis or signaling have been identified. For example, D27 encoding a novel ironcontaining protein is involved in SL biosynthesis [33,34]. PLEIOTROPIC DRUG RESISTANCE1 (PDR1) encoding an ATP-binding cassette transporter functions as a cellular SL exporter [35]. D14 encoding a protein of the a/b-fold hydrolase superfamily is involved in SL signaling and likely functions as a receptor for SLs [36][37][38][39][40]. D53, a member of class I Clp ATPase protein family, is a substrate of the SCF D3 ubiquitination complex and the degradation of D53 protein is promoted by SL and is dependent on D14 and D3 (a rice ortholog of MAX2) [41,42]. The degradation of BES1, a positive regulator in brassinosteroid signaling pathway, is also dependent on MAX2 [43].
However, despite this important discovery, little is known about this new class of plant hormones in perennial woody plants. So far, among perennial woody plants, strigolactone pathway genes have only been recently studied in willow [44,45]. Because differences have been observed between monocots and dicots, between herbaceous and woody plants, and between annual and perennial plants, it has remained elusive whether the SL pathways are conserved in the perennial woody plants, such as Populus. For example, studies in herbaceous plants (such as Arabidopsis and pea) suggested that SLs act as long-distance signaling molecules that can be transported from roots to shoots to exert functional control on shoot branching. However, the woody perennial plants, such as Populus, are typically several orders of magnitude taller than annual herbaceous plants. This raises the question of whether SL can function similarly in Populus.
As a first step towards exploring SL pathways in the model woody plant Populus, we conducted a genome-wide search of sequence homologues of strigolactone pathway genes in the sequenced genome of Populus trichocarpa [46]. By using Arabidopsis strigolactone pathway genes as templates, we identified Populus sequence homologues for each of those four founding members of strigolactone pathway genes, namely MAX1, MAX2, MAX3 and MAX4. We found that Populus MAX paralogous genes are differentially expressed across various tissues and organs. We showed that Populus MAX genes could complement or partially complement the shoot branching phenotypes of corresponding Arabidopsis max mutants. These findings provide genetic evidence that SL pathways are likely conserved in the woody perennial plants.

Cloning of Populus MAX Homologous Genes
The full-length open reading frame (ORF) of each PtrMAX gene was determined according to the sequence information available at Phytozome [46,49]. Gene-specific primers were designed to amplify the full-length ORF of each PtrMAX gene from cDNA library derived from RNA isolated from leaves and roots of Populus trichocarpa plants. Subsequently, the full-length ORF of each PtrMAX gene was introduced into the pENTR vector by using pENTR/D-TOPO Cloning Kit (Life Technologies). The cloned RT-PCR products were validated by sequencing and then transferred into plant Gateway destination vector pGWB502V (26CaMV35SV) [50] using LR clonase (Life Technologies). All primers used for cloning are listed in Table S1.
Genetic Complementation pGWB502V (26CaMV35SV) binary vectors containing 35S:PtrMAX plasmid were transformed into Arabidopsis max mutants via Agrobacterium tumefaciens strain GV3101 mediated flower dipping transformation [51,52]. T1 transformants were selected using 20 mg/L hygromycin B. A minimum of 20 independent transgenic lines were selected for each transgene. Two independent transgenic lines were used for further studies.
When plants reached maturity, the number of primary rosetteleaf branches was counted. A minimum of 10 individual plants per genotype were used.

RT-PCR
To examine the absence or presence of Arabidopsis MAX transcript in the max mutant, total RNA was extracted from 7-dayold seedlings using the Invisorb Spin Plant Mini Kit (Stratec Molecular). Two mg of total RNA were reversely transcribed to cDNA using Fermentas RevertAid Reverse Transcriptase (Thermo Scientific). For semi-quantitative RT-PCR, Arabidopsis MAX-specific primers spanning the full-length ORF of MAX gene were used. PCR amplification of AtACTIN8 (AtACT8) served as a control.
For the examination of expression of Populus MAX genes in Arabidopsis transgenic lines, PtrMAX gene-specific primers were used. All primers used for semi-quantitative RT-PCR are listed in Table S1.

Quantitative RT-PCR (qRT)
qRT was conducted to examine the transcript level of each PtrMAX genes across various tissues and organs using a StepOnePlus (Applied Biosystems), Maxima SYBR Green/ROX qPCR Master Mix (Thermo Scientific) and cDNA corresponding to 80 ng RNA in a total volume of 25 ml. Three biological replicates were used for qRT analysis. The following cycling conditions were used for PCR: 10 min at 95uC, 35 cycles of 15 s at 95uC and 60 s at 60uC. The transcript level was normalized against that of PtrACTIN5 (PtrACT5). Gene-specific qRT primers were designed using QuantPrime [53]. All primers used for qRT analysis are listed in Table S1.

Sequence Homologues of MAX Proteins in Populus
MAX1, MAX2, MAX3 and MAX4 are four founding members of SL pathway genes in the model plant Arabidopsis. In order to identify sequence homologs of MAX genes in Populus trichocarpa (hereafter referred to as Populus), we used amino acid sequences of Arabidopsis MAX proteins as templates to perform protein homology search in the fully-sequenced Populus genome.
Based on the current annotation by TAIR (http://www. arabidopsis.org) [54] and Phytozome (www.phytozome.net) [49], Arabidopsis MAX1 (AtMAX1) has two transcript variants. The primary variant encoded by gene locus At2g26170.1 was used as a template for searching protein sequence homologues encoded by the Populus genome using the ''Protein Homologs'' search tool at Phytozome. The search identified two close sequence homologues encoded by loci Potri.006G226700 and Potri.018G062100, designated as PtrMAX1a and PtrMAX1b, respectively ( Figure 1). Based on the current annotation by Phytozome, PtrMAX1a has three transcript variants and PtrMAX1b has four transcript variants. The primary variants (e.g., Potri.006G226700.1 and Potri.018G062100.1) were used for subsequent analysis. PtrMAX1a showed approximately 84% similarity and 69% identity with AtMAX1 at the amino acid level. PtrMAX1b showed approximately 85% similarity and 69% identity with AtMAX1 at the amino acid level. PtrMAX1a and PtrMAX1b each other shared approximately 96% similarity and 90% identity at the amino acid level. The number of amino acids of PtrMAX1a (529 aa) and PtrMAX1b (529 aa) are comparable to that of AtMAX1 (522 aa). No other Populus proteins showed more than 43% similarity with AtMAX1 at the amino acid level.
The two closest Populus sequence homologues of Arabidopsis MAX2 (AtMAX2) were encoded by loci Potri.014G142600 and Potri.011G066700, designated as PtrMAX2a and PtrMAX2b, respectively ( Figure 2). A third homolog, Potri.002G214300 consisting of 316 aa is only half the size of AtMAX2 (693 aa). Thus, it was not included in our further analysis. PtrMAX2a showed approximately 76% similarity and 61% identity with AtMAX2 at the amino acid level. PtrMAX2b showed approximately 71% similarity and 54% identity with AtMAX2 at the amino acid level. PtrMAX2a and PtrMAX2b each other shared approximately 73% similarity and 57% identity at the amino acid level. The number of amino acids of PtrMAX2a (701 aa) and PtrMAX2b (671 aa) are comparable to that of AtMAX2 (693 aa). No other Populus proteins showed more than 32% similarity with AtMAX2 at the amino acid level.
Arabidopsis MAX3 (AtMAX3) has only one close sequence homologue in Populus encoded by loci Ptri.014G056800, designated as PtrMAX3 (Figure 3). PtrMAX3 showed approximately 77% similarity and 62% identity with AtMAX3 at the amino acid level. The number of amino acids of PtrMAX3 (613 aa) is comparable to that of AtMAX3 (629 aa). No other Populus proteins showed more than 35% similarity to AtMAX3 at the amino acid level.
Arabidopsis MAX4 (AtMAX4) has two close sequence homologues in Populus encoded by loci Potri.018G044100 and Potri.006G238500, designated as PtrMAX4a and PtrMAX4b, respectively ( Figure 4). Both PtrMAX4a and PtrMAX4b have two transcript variants based on the current annotation by Phytozome. The primary variant (e.g., Potri.018G044100.1 and Potri.006G238500.1) were used for subsequent analysis. PtrMAX4a showed approximately 78% similarity and 65% identity with AtMAX4 at the amino acid level. PtrMAX4b showed approximately 78% similarity and 64% identity with AtMAX4 at the amino acid level. PtrMAX4a and PtrMAX4b each other shared approximately 97% similarity and 93% identity at the amino acid level. The number of amino acids of PtrMAX4a (557 aa) and PtrMAX4b (557 aa) are comparable to that of AtMAX4 (570 aa). No other Populus proteins showed more than 42% similarity to AtMAX4 at the amino acid level.
In total, we identified two Populus sequence homologues each for Arabidopsis MAX1, MAX2 and MAX4, and one Populus sequence homologues for Arabidopsis MAX3.

Tissue and Organ Expression Patterns of Populus MAX Genes
Subsequently, we examined the expression patterns of each PtrMAX gene across various tissues and organs. We used qRT to examine the transcript level of each PtrMAX gene and normalized their transcript levels against the transcript level of PtrACT5. We made three interesting observations. Firstly, each PtrMAX gene was expressed at different levels with PtrMAX1a at the highest level and PtrMAX4b at the lowest level in most tissues and organs ( Figure 5). Secondly, differential expression in terms of transcript level was observed between PtrMAX paralogous genes. For example, PtrMAX1a was expressed over 10-fold higher than PtrMAX1b across most tissues and organs. Similarly, PtrMAX4a was expressed over 10-fold higher than PtrMAX4b across most tissues and organs. In contrast, PtrMAX2a and PtrMAX2b were expressed at similar levels in most tissues and organs. Finally, differences in tissue and organ expression patterns were observed between PtrMAX paralogous genes. For example, PtrMAX4a was highly expressed in roots but the transcript of PtrMAX4b was undetectable in roots ( Figure 5).

Genetic Complementation of Arabidopsis max Mutants with Populus MAX Genes
In order to further test whether PtrMAX genes may function similarly as AtMAX genes, we conducted genetic complementation studies by heterologously expressing each PtrMAX gene in corresponding Arabidopsis max mutants. The expression of each PtrMAX gene was driven by the constitutive 35S promoter [55]. At least 20 independent transgenic lines were selected from each transformation and two independent transgenic lines were selected for further analysis. We used the number of primary rosette-leaf branches as a phenotype to determine whether heterologous expression of each PtrMAX gene could rescue the corresponding Arabidopsis max mutants.
For the test of PtrMAX1 genes, we used a homozygous T-DNA insertional max1 mutant allele designated as max1-4 as the transformation background. In this mutant allele, a T-DNA is insertion in the 1 st exon, 74 bp downstream of the start codon of MAX1 ( Figure S1). RT-PCR analysis indicated that the full-length transcript of MAX1 was absent in max1-4 mutant, suggesting that it is likely a loss-of-function allele ( Figure S1). Consistent with the RT-PCR result, max1-4 mutants displayed high number of primary rosette-leaf branches ( Figure S1). For the complementation, the full-length ORF of PtrMAX1a and PtrMAX1b was each cloned into the binary plant expression vector pGWB502V (26CaMV35SV) [50] and transformed into max1-4 mutant background via Agrobacterium-mediated transformation. RT-PCR analysis indicated that in the transgenic lines (lines #7 and #8 for PtrMAX1a, and lines #2 and #6 for PtrMAX1b), PtrMAX1a or PtrMAX1b transgene was expressed in the Arabidopsis max1-4 mutant background ( Figure 6A, B). We counted the number of primary rosette-leaf branches and compared it with that of Columbia-0 (Col-0) wild type and max1-4 mutants which were grown side-by-side with transgenic lines under identical conditions. We found that the number of primary rosette-leaf branches in the transgenic lines was largely reverted to a similar number as in the wild type. Under our growth conditions, Col-0 produced about two primary rosette-leaf branches whereas max1-4 mutants produced about nine primary rosette-leaf branches. The PtrMAX1a transgenic lines #7 and #8 produced approximately two primary rosette-leaf branches and PtrMAX1b transgenic lines #2 and #6 produced about four primary rosette-leaf branches ( Figure 6C). These results indicated that both PtrMAX1a and PtrMAX1b were able to complement or partially complement the shoot branching phenotypes of Arabidopsis max1-4 mutant.
For the test of PtrMAX2 genes, we used a T-DNA insertional max2 mutant allele, max2-4, as the transformation background. The max2-4 allele had been previously described by Umehara et al. [2] and Ha et al. [56] but the genotyping and branching phenotype were not reported. These data are provided here as supplemental materials ( Figure S2). In this mutant allele, a T-DNA insertion site was found within the 1 st exon. RT-PCR analysis indicated that the full-length transcript of MAX2 was absent in max2-4 mutant, suggesting that it is likely a loss-of-function allele. Consistent with the RT-PCR result, max2-4 mutants displayed more shoot branches ( Figure S2). For the complementation, the full-length ORF of PtrMAX2a and PtrMAX2b was each cloned into the binary plant expression vector pGWB502V and transformed into max2-4 mutant background. RT-PCR analysis indicated that in the transgenic lines (lines #4 and #8 for PtrMAX2a, and lines #3 and #6 for PtrMAX2b), PtrMAX2a or PtrMAX2b transgene was expressed in the Arabidopsis max2-4 mutant background (Figure 7). Compared with max2-4 mutants, the number of primary rosette-leaf branches in the transgenic lines was significantly reduced and in line #4 of PtrMAX2a, the number was reverted to a similar number in the wild type ( Figure 7C). These results indicated that both PtrMAX2a and PtrMAX2b were able to complement or partially complement the shoot branching phenotypes of Arabidopsis max2-4 mutant.
For the test of PtrMAX3 gene, we used a T-DNA insertional max3 mutant allele, max3-12, as the transformation background. max3-12 was previously described by Umehara et al. [2] and the genotypic and phenotypic data had been reported by Li et al. [47]. The full-length ORF of PtrMAX3 was cloned into the binary plant expression vector pGWB502V and transformed into max3-12 mutant background. RT-PCR analysis indicated that in the transgenic lines (lines #2 and #4), PtrMAX3 transgene was expressed in the Arabidopsis max3-12 mutant background (Figure 8). Compared with max3-12 mutants, the number of primary rosette-leaf branches in the transgenic lines was significantly reduced but was not completely reverted to wild-type branch number ( Figure 8C). These results indicated that PtrMAX3 was able to partially complement the shoot branching phenotypes of Arabidopsis max3-12 mutant.
For the test of PtrMAX4 genes, 35S:PtrMAX4a and 35S:PtrMAX4b transgenic lines were generated using Arabidopsis max4-1 mutant [31] as background. RT-PCR analysis indicated that in the 35S:PtrMAX4 transgenic lines, PtrMAX4a (lines #4 and #5) or PtrMAX4b (lines #3 and #9) transgene was expressed ( Figure 9). Compared with max4-1 mutants, the numbers of primary rosette-leaf branches were significantly reduced in these transgenic lines but were not completely reverted to wild-type branch number ( Figure 9C). These results indicated that both PtrMAX4a and PtrMAX4b were able to partially complement the shoot branching phenotype of Arabidopsis max4 mutants.

Discussion
SLs are a new class of plant hormones controlling shoot branching. Due to the importance of shoot branching in the determination of photosynthetic light use efficiency and biomass yield in woody perennial trees, we wanted to examine whether the strigolactone pathway is conserved in the woody model plant Populus. We report the identification of sequence homologues of four founding members of SL pathway genes, namely MAX1, MAX2, MAX3 and MAX4. Furthermore, we provide genetic evidence that these Populus MAX genes likely function similarly in controlling shoot branching.
Due to recent genome duplication, in general, each Arabidopsis gene can have two paralogous genes in Populus. As such, we found two Populus paralogous genes for MAX1, MAX2 and MAX4 (Figures 1, 2 and 4). The exception here is MAX3; only one MAX3 homologous gene was identified in the Populus genome ( Figure 3). Our gene expression analysis indicated that Populus MAX paralogous genes are differentially expressed across various tissues and organs, in terms of expression level and expression patterns. Over 10-fold differences in transcript level were observed between PtrMAX1a and PtrMAX1b, and between PtrMAX4a and PtrMAX4b ( Figure 5). It remains unclear whether differences in expression level or pattern of Populus MAX genes may result in differences in their functioning. When their expression was driven by the constitutive 35S promoter, PtrMAX1 and PtrMAX4 paralogous genes could rescue the shoot branching phenotype of the corresponding Arabidopsis max mutants ( Figure 6, Figure 9), suggesting that both paralogous genes could function similarly. The expression analysis also indicated that PtrMAX3 and PtrMAX4a were expressed at the highest level in roots ( Figure 5). These results are consistent with the view that roots represent one of the major biosynthesis sites for SLs (reviewed by [25]). This has been demonstrated in a grafting experiment where grafting wildtype Arabidopsis rootstocks to max1, max3, or max4 scions were able to restore a wild-type branching pattern to the mutant shoots [30].
For the complementation tests, we found that all PtrMAX genes were able to rescue or partially rescue the shoot branching phenotypes of corresponding Arabidopsis max mutants (Figures 6,  7, 8 and 9). For PtrMAX1 and PtrMAX2, we could identify transgenic lines in which the numbers of primary rosette-leaf branches were completely reverted to wild-type level ( Figure 6, Figure 7), suggesting that PtrMAX1 and PtrMAX2 may function equivalently to Arabidopsis MAX1 and MAX2, respectively. On the other hand, although the numbers of primary rosette-leaf branches in PtrMAX3 and PtrMAX4 transgenic lines were significantly reduced when compared with Arabidopsis max3 and max4 mutants, respectively, the numbers were not completely reverted to wild-type level. There are several possible explanations. Firstly, the expression level of PtrMAX3 and PtrMAX4 in the transgenic lines may not be high enough to render complete rescue. Secondly, PtrMAX3 and PtrMAX4 may function similarly, but not equivalently, as Arabidopsis MAX3 and MAX4, respectively. PtrMAX3 showed approximately 77% similarity and 62% identity with MAX3 ( Figure 3) and PtrMAX4 showed approximately 78% similarity and 65% identity with AtMAX4 at the amino acid level (Figure 4). It is unclear whether differences in amino acid sequence between Populus and Arabidopsis MAX3 and MAX4 may have contributed to such partial rescue phenotypes in the transgenic lines. Finally, it is also possible that there may be other MAX3 and MAX4 homologues in Populus though no other Populus proteins showed more than 35% similarity with AtMAX3 and no other Populus proteins showed more than 42% similarity with AtMAX4 at the amino acid level. This deserves further investigation.
In addition to MAX genes, several other genes are involved in SL biosynthesis or signaling such as D27 [33,34] and D14 [36,38- 40]. It would be helpful to identify sequence homologs of each of those genes in Populus and conduct genetic complementation studies similar to what have been described in this study in order to strengthen the view that strigolactone pathway exists and operates in the woody perennial plant Populus. Our preliminary analysis indicated sequence homologues of AtD14 (encoded by locus At3g03990) are present in the genome of Populus. For example, two proteins encoded by locus Potri.002G118900 and Potri.014G016500 showed about 89% and 88% similarity with AtD14 at the amino acid level. Our preliminary analysis indicated that sequence homologues of AtD27 (encoded by locus At1g03055) are also present in the genome of Populus. For example, a protein encoded by locus Potri.005G216400 showed about 59% similarity with AtD27 at the amino acid level. In order to fully assign SL pathway in Populus, future studies should focus on the functional characterization of these SL pathway genes (e.g. generating Populus transgenic lines), the determination of SL molecules, biosynthetic intermediates and derivatives, and the determination of the physiological roles of SLs in Populus.
In summary, this study provided a preliminary characterization of strigolactone pathway genes in the model woody plant Populus. Sequence homologues of those four founding members of Arabidopsis SL pathway genes have been identified in Populus. Gene expression analysis indicated that Populus MAX paralogous genes are differentially expressed across various tissues and organs. Genetic complementation studies indicated Populus MAX genes are able to complement or partially complement the shoot branching phenotypes of corresponding Arabidopsis max mutants. These findings lay a foundation for further characterization of SL pathway and its functions in woody perennial plants.

Acknowledgments
Disclaimer: This manuscript has been authored by UT-Battelle, LLC, under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes.