Tk seeds were obtained from the United States Department of Agriculture-Agricultural Research Service National Plant Germplasm System, where they were deposited after collection in Kazakhstan in 2008 (accession No. KAZ08-014, ID W6-35169). Plants were germinated and grown in a greenhouse at temperatures between 28˚C (day) and 18˚C (night), with supplemental metal halide lighting to provide a 15-h-day length (1000 to 1250 µmol m⁻²・s⁻¹).

Total RNA from root, leaf and flower tissues of greenhouse-grown Tk plants was extracted using a PowerPlant^® RNA Isolation Kit with DNase (MoBio Laboratories Inc., USA). The total RNA was converted to cDNA using the Superscript III First-Strand Synthesis System for Reverse Transcriptase (RT)-Polymerase Chain Reaction (PCR) (Life Technologies, USA). Quantitative PCR (qPCR) on cDNAs of TkHMGR and 18S rRNA was performed using Applied Biosystems 7500 Fast Real Time PCR System. The following combinations of forward/ reverse primers were used for the qPCR reactions:

for TkHMGR1 5’-CTGGAACTATTGCTCCGACAA-3’/5’-GCCATGGCGGACCCTGTCAGG-3’ and

for Tk18S rRNA 5’-CGGCTACCACATCCAAGGAA-3’/5’-TGTCACTACCTCCCCGTGTCA-3’.

Other primer sets used in spatial and temporal analysis of Tk HMGR-like sequences are:

TkHMGR2 5’-TTTGAAATTCTTCTTGGAGGAA-3’/5’-ACCATGTTCATCCCCATTGCATC-3’;

TkHMGR3 5’-TTGCCACGGGACAAGACCCCGCT-3’/5’-GACTGTGAAGCTAGTTGAGTTCC-3’;

TkHMGR4 5’-CGGTCAGACCCTGCCCGGAATGTG-3’/5’-TCTATGGAAGGCATGGTGACAGA-3’;

TkHMGR5 5’-TTGCAGCACCGGTGATGCAATGGG-3’/5’-ACAGACTTTCCACGCCCTTCTATCCA-3’.

PCR amplification reactions were carried out in a volume of 25 µL containing 20 ng of cDNA, 0.5 µM each of forward and reverse primers, and 1x SYBR^® Green Master Mix (Applied Biosystems USA), using the following temperature regime: 95˚C for 20 s, followed by 40 cycles of 95˚C for 3 s and 60˚C for 30 s. The relative copy numbers of a gene were averaged over triplicates in the same 96-well micro chamber plate. Expression of TkHMGR was normalized to the expression of the constitutively expressed endogenous reference gene 18S rRNA. The method of Pfaffl (2001) [29] was applied to calculate comparative expression levels between samples. The qPCR experiments were repeated at least twice.

Hevea HMGR1 (X54659) was used as query for homology searches (tblastn) of a Tk root tissue expression library DNA sequences available at the National Center for Biotechnology Information (NCBI). Seventeen TkHMGR-like partial sequences were found with an average length of 750 bp (Table 1). Eleven of the sequences were homologous to the N-terminus of Hevea HMGR1, three to the C-terminus, and three to the middle portion of the protein (between amino acids 167 - 426). The large numbers of N-terminus homology sequences and the low representation of sequences with C-terminus homology suggest a technical problem during construction of the cDNA library. HMGR is known to be present in plants in several isoforms. Without full-length TkHMGR sequence information, it is impossible to determine the number of TkHMGR isoforms. However, spatial (organ specific) and temporal (age specific) expression analysis of these sequences can provide useful information about the presence of tissue-specific isoforms.

We aligned all 17 Tk sequences against the Hevea HMGR1 gene (X54659) and designed five sets of PCR primers. Computational analysis of the five primer sets predicted that each would amplify at least one of the following 6 sequences, GO661008, GO662289, GO668051, GO666861, GO665221, and GO670200. PCR amplifications were performed on cDNAs derived from total RNA extracted from leaf, root, and flower tissues of ~11 weeks old Tk plants. All primers sets amplified a product in all tissues (data not shown). However one primer set (Materials and Methods) consistently amplified an abundant product from cDNAs from root tissues. This primer set, herein called TkHMGR1 primers, was used for quantitative expression analysis. Computational analysis of TkHMGR1 primers predicted only one PCR product from only one sequence, GO666861, and, interestingly, no amplification from the cloned TkHMGR sequence (HQ857601).

*Compared to Hevea HMGR1 sequence (X54659); **Location in Hevea HMGR1 sequence (X54659).

To examine the relationship between rubber accumulation and TkHMGR1 expression, we further quantified TkHMGR1 transcript levels in root and leaf of young (~11 weeks old) and mature (~20 weeks old) Tk plants using qPCR technology. In our lab, rubber extractions from Tk roots typically yield 4 mg/g fresh weight from 12-week-old plants and 10 mg/g fresh weight from 20-week old plants). No rubber was detected in leaf tissues. As shown in Figure 1, we detected 11-fold or 3-fold higher of TkHMGR1 transcript levels in roots than in leaves from young or mature plants, respectively. The results indicated that the expression of TkHMGR1 was higher in root than leaf during the active growth phase of Tk plants. Although functional studies need to be carried out in order to determine the enzymatic activity and substrate specificity of TkHMGR1 in rubber metabolism, it is likely that TkHMGR1 contributes to rubber synthesis in Tk roots based on its expression profile.

Rubber is the end metabolite of the MVA biosynthesis pathway. The amount of rubber in mature root is the result of continuous synthesis and accumulation over the lifespan of the plant. Although we observed a decline in expression of TkHMGR1 in roots when plants became mature (Figure 1), the level of the transcript was probably still high enough to produce active TkHMGR1 enzymes for rubber biosynthesis. It is also possible that TkHMGR1 is a relatively stable protein, and thus supports continuous biosynthesis of rubber over development, leading to the high accumulation of rubber in root.

In summary, we have identified the TkHMGR1 isoform that is highly expressed in Tk root. The C-terminal part of TkHMGR1 sequences are represented by GenBank accession GO666861. The partial sequence provided the basis for the design of isoform-specific primers that can be used to isolate full-length TkHMGR1 and eventually its promoter sequences. The likely role that TkHMGR1 plays in rubber synthesis suggests that its over- expression in guayule or Tk could increase rubber production. In addition, the TkHMGR promoter can be used to drive gene expression for other key enzymes/proteins associated with rubber synthesis in Tk root, such as the genes for IPP synthase, cis-prenyltransferase, rubber elongation factor, etc. Thus, the sequence identified in this study could lead to other valuable tools for genetic engineering of rubber-producing crops.