Isolation of a Sequence Homolog to More Axillary Branches MAX2 Gene in Hibiscus rosa-sinensis and its Use as Genetic Marker

Introduction

Hibiscus rosa-sinensis is one of the most widely planted ornamental shrubs cultivated throughout the tropics and sub-tropics. Numerous varieties and hybrids are particularly appreciated in garden and landscape for their vigorous growth habit, but they are mainly employed as a pot plant when growth retardants for keeping a reduced plant size are applied. A compact basal branching growth habit is generally preferred.

Lateral branching structures exist in many forms throughout higher plants and even if plant architectures are influenced by environmental factors, their species-specific characteristics indicate the presence of widely conserved genetic regulatory mechanisms (McSteen and Leyser 2005, Johnson et al 2006).

In different plant species, pre-existing axillary meristems may either lie dormant for long periods or they may develop into branches instantaneously. This bud growth can be activated by intrinsic factors, and hormones play a crucial role in shoot branching control (Leyser 2009). It has been known for many years that auxin synthesize in the apex inhibits axillary meristems outgrowth, whereas cytokinin promote it efficiently by regulating the shoot branching phenomena (Liang et al 2010).

However, the involvement of a novel hormone-signaling pathway in the regulation of bud growth has been inferred by genetic analyses of mutants that have enhanced shoot branching phenotype in Arabidopsis, pea, petunia and rice (Booker et al 2005, Morris et al 2001, Snowden and Napoli 2003, Arite et al 2009). In Arabidopsis thaliana, a suite of mutants with More Axillary Branches encoded by the MAX1, 2, 3, and 4 genes has been analyzed by a combination of grafting and molecular techniques (Bennet et al 2006). The recessive mutations (max) cause premature and enhanced outgrowth of lateral shoots in combination with modest pleiotropic effects. These studies suggested that MAX1, 3 and 4 are involved in the synthesis of a mobile signal, whereas the MAX2 gene product mediates perception and response to the signal (Leyser 2009). Particularly, MAX2 has been shown to encode a nuclear localized F-box leucine-rich repeat (LRR) protein within the SCF (Skp1-Cullin-F box) complex that catalyzes the ubiquitination of proteins, and thus target them for proteasomal degradation (Xu et al 2009). In the case of MAX2, one or more proteins that activate bud growth are in the wild-type targeted for destruction by the MAX2 F-box LRR product (Stirnberg et al 2007). Presumably, these proteins, which would be stabilized in the absence of MAX2 activity, would in some way promote branching.

More recently, Wang et al (2013) demonstrated in Arabidopsis that the strigolactone hormone inhibits auxin transport, suggesting a complex interaction between these two hormones and the MAX2 F-box binding site in the protein degradation system.

Given the metabolic complexity of plants, there are probably more, perhaps many more, small molecules with signaling function. However, the discovery of regulatory mechanisms promoting the axillary branch proliferation could provide an environmentally independent, rapid and helpful tool for preliminary screening of genotypes characterized by a compact basal branching growth habit, suitable for pot plant cultivation.

With the aim to identify a gene-derived ‘perfect’ molecular marker associated to the compact plant architecture, we isolated conserved sequences for the MAX2 gene in the underinvestigated ornamental species Hibiscus rosa-sinensis. The knowledge gained through the previous AFLP characterization of a collection of H. rosa-sinensis cultivars (Braglia et al 2010) allowed us to develop a new strategy. Starting from plant MAX2 gene sequences, we followed a combined approach of degenerate primer PCR together with AFLP technique

Material and Methods

Genomic DNA was isolated from nineteen Hibiscus rosa-sinensis cultivars with different and contrasting plant architectures and thirteen Hibiscus botanical species (H. arnottianus G., H. boryanus H. and A,. H. calyphyllus Cav., H. cannabinus L., H. denisonii B., H. genevii B., H. kokio H., H. moscheutos L., H. panduriformis B., H. schizopetalus H.,H. storckii S., H. syriacus L., H. tiliaceus L.), selected from materials collected at the CRA-FSO in Sanremo (Italy). DNA from one hundred milligrams of fresh leaves was extracted using the DNeasy Plant Mini Kit (Qiagen, Germany) following the modified protocol reported for Hibiscus spp. by Braglia et al (2010).

According to the full-length cDNA sequence of Arabidopsis MAX2 gene (NM_129823), pea RAMOSUS4 gene (DQ403159), rice LRR-repeat MAX2 homolog (Oryza sativa Japonica group) (Q5VMP0) and poplar F-box family protein mRNA sequence (Populus trichocarpa) (XM_002320376) showing strong homology at the amino acid level, a consensus sequence by multiple sequence alignment was generated. A set of six degenerate primers were then designed and tested on three different Hibiscus rosa-sinensis genomic DNAs. PCR reactions were performed in a 50 µl containing 1X PCR Buffer (HotStartTaq®Plus Buffer Qiagen, Germany), 0.2 mM each dNTP, 2 mM MgCl₂, 150 ng of DNA template, 1.6 µM primer and 2.5U Taq DNA Polymerase (HotStartTaq®Plus Qiagen, Germany). The PCR conditions were 2 min at 94° C, 5 cycles 30 s at 94° C, 45s at 48° C, 2 min at 72° C, followed by 35 additional cycles 30 s at 94° C, 45 s at 58° C, 90 s at 72°. The reactions were held at 4°C after a final extension at 72°C for 10 min.

The amplified products were separated using agarose electrophoresis (2.2 %). Only one primer pair (dF3 5’-TTYACNGARGGNTTCAAGTC-3’; dR2 5’-CCYTGRAAGTGNCCNAGCTT-3’) yielded the expected size fragment. This PCR product was subsequently cloned (TA Cloning® kit, Invitrogen) and sequenced, then analyzed using bioinformatic tools at the websites http://www.ebi.ac.uk/Tools/ and http://www.ncbi.nlm.nih.gov/.

Hibiscus specific primers (notable as Hsp_, Table 1) were designed on the isolated fragment sequence in forward and reverse. An AFLP (Amplified Fragment Length Polymorphism)-based approach was adopted to extend the Hibiscus DNA segment outside the boundary known sequence. Restricted/ligated fragments (EcoRI/MseI), hereafter R/L were generated according to the AFLP protocol reported by Vos et al (1995) from 300 ng of genomic DNA. The obtained R/L products were used to test different Hsp_ primers in combination with AFLP primers (Table 1). These latter, named Ead_pr and Mad_pr, had the 5’-region complementary to the adapter and the restriction site sequence without selective nucleotides at the 3’-end.

PCR reactions were performed in a 25 µl containing 1X PCR Buffer (HotStartTaq®Plus Buffer Qiagen, Germany), 1 µl R/L, 0.2 mM each dNTP, 1.5 mM MgCl₂, 0.5 µM for the Hsp_ primer and 0.1 µM for the AFLP primer with 2.5U of Taq DNA Polymerase (HotStartTaq®Plus Qiagen, Germany). The following PCR conditions were used: 5 min at 95 °C, 13 cycles of 30 s at 90°C, 30 s at 67°C, 60 s at 72°C with a decrease of 0.7°C of the annealing temperature carried out in each cycle followed by 27 additional cycles of 30 s at 90° C, 30 s at 56° C, 60 s at 72°. The reactions were held at 4°C after a final extension at 72°C for 10 min. The obtained PCR product was subsequently cloned (TA Cloning® kit, Invitrogen) and sequenced. One hundred additional base pairs were achieved from the AFLP-based approach and two new Hsp_ reverse primers were synthesized (Fig. 1, Table 1).

 Fig. 1: Hibiscus MAX2 Gene Fragment Schematic Representation. MseI Restriction Site is Reported at the 5’ Region. Black Arrows Indicate the Name and Position of Hsp_ Specific Primers.

Table 1: Hibiscus Specific Primers and AFLP Core Primers

The selected primer pair Hsp_Up1Hsp_Dw1 was tested on all genomic samples using the same PCR conditions reported above. Amplified fragments were sequenced to assess nucleotide polymorphisms.

Cluster analysis of Hibiscus MAX2-like isolated nucleotide sequences was performed using the Treecon program for Windows (Van de Peer and De Wachter 1994), with a bootstrap test (Hillis and Bull 1993).

The amino acid sequences were deduced and the sequence comparison was conducted through database search using UniProt (Universal Protein Research http://www.uniprot.org)

Results

The degenerate primer amplification allowed the identification of a DNA fragment (~350 bp) showing 65% similarity to the Arabidopsis MAX2 gene and the deduced amino acid sequence revealed a high degree of homology with those of other F-box subunit proteins from various biological sources: Populus trichocarpa (64%), Arabidopsis thaliana (64%), Pisum sativum (61%) and Oryza sativa (32%), most of them employed in the degenerate primer design.

Concerning to the AFLP approach, all tested specific primers in combination with the primer Ead_pr did not produce any amplification products (data not shown). Whereas, the primer combination Mad_pr/Hsp_Up1 allowed to extend theHibiscus MAX2-like sequence downstream the 3’ boundary known sequence of one hundred additional nucleotide base pairs. Unfortunately, the presence of a MseI restriction site at the 5’ region of the isolated fragment (named HibMAX2-like) did not permit to extend the sequence upstream (Fig.1).

The HibMAX2-like sequences could be amplified in all cultivars of Hibiscus rosa-sinensis and Hibiscus species except for H. cannabinus (kenaf). Control reactions on genomic DNA samples of Arabidopsis and pea did not produce any amplicons (Data not shown). The Hibiscus MAX2-like nucleotide sequences were submitted to NCBI database with the Accession Numbers JF813799-JF8137824. A neighbour-joining tree was then constructed (Fig. 2) for all Hibiscus isolated nucleotide sequences. In this tree, a main cluster (A, bootstrap value 98%) could be recognized, grouping all Hibiscus rosa-sinensis cultivars. Furthermore, some botanical species such as H. arnottianus, H. boryanus, H. denisonii, H. genevii, H. kokio, H. schizopetalus and H. storckii were spread throughout the main cluster. Conversely, H. calyphyllus, H. moscheutos, H. panduriformis, H. syriacus and H. tiliaceus, were grouped in separate clusters.

Fig. 2: Neighbor-joining Tree Built from HibMAX2 Nucleotide Sequences of 19 Hibiscus rosa-sinensis Varieties and 12 Hibiscus Botanical Species. The Arabidopsis MAX2 Sequence was Reported as an Outgroup. Numbers on Nodes Indicate the Bootstrap Values after 1000 Replicates.