Differential Gene Expression in Papaver-Species in Comparison with Alkaloid Profiles

Papaver species are known to produce a large variety of benzylisoquinoline alkaloids, with each species exhibiting a specific alkaloid profile, but only a few genes involved in the biosynthesis or regulation of these complex pathways are known so far. Here we used a genomic approach to discover genes responsible for the determination of a specific alkaloid profile. A stem expressed sequence tag database of ~1100 unique genes from Papaver somniferum was created. Gene expression analysis of these sequences in P. bracteatum, P. somniferum and P. somniferum ‘Noscapine’ exhibited 39 cDNAs showing differential expression coincident with morphine accumulation. INTRODUCTION The tetrahydrobenzylisoquinoline class of alkaloids contains many physiologically active compounds, such as the narcotic analgesic, morphine, the antitussive, codeine, and noscapine, or the vasodilator papaverine. The tetrahydrobenzylisoquinolines can be divided into several classes such as pavines (isopavines, benzophenanthridines, rhoeadines) papaverrubines, protopines, phthalideisoquinolines, protoberberines, aporphines and morphinans (Preininger, 1986). Common to all these classes are the first steps in the biosynthesis, which includes the condensation of two tyrosine derived molecules to norcoclaurine and subsequent Nand O-methylations and hydroxylation steps leading to (S)-reticuline (Kutchan et al., 1998). This central intermediate enters different pathways for the generation of the individual tetrahydrobenzylisoquinoline classes. In contrast to the biosynthetic enzymes leading to (S)-reticuline, little is known about the following steps on the enzymatic and molecular level, but from the individual reactions one can assume that O-methyltransferases and P450-monooxygenases predominate. Similarly, knowledge about the regulation of the whole pathway and transport processes of intermediates is scarce. With a few exceptions, tetrahydrobenzylisoquinolines occur in the Ranunculales, specifically in the family of the Papaveraceae (Jensen, 1995). Among them, only species belonging to the genus Papaver are known to contain all classes of tetrahydrobenzylisoquinoline alkaloids, whereby strong variations in the alkaloid profiles are species specific (Waterman, 1998). We use the close genetic relationship but the diversity in the alkaloid profiles of Papaver species to correlate the gene expression with the alkaloid profiles in order to access cDNAs responsible for a distinct alkaloid profile. MATERIALS AND METHODS Plants were either grown in the field or in the greenhouse as indicated. Upper stems and latex of Papaver plants were harvested 4-6 days after petal fall. For alkaloid analysis, stems and latex were extracted with 80% ethanol and analysed by HPLC at 210 nm on LiChrospher 60 RP-select B columns (5 μm, 250 mm x 4 mm, Merck Eurolab, Darmstadt, Germany) using an acetonitrile gradient from 2% to 46% in 25 min, 46% to 60% in 5 min, constant 60% for 5 min and finally an increase up to 100% in 5 min in 0.01% phosphoric acid. Proc. WOCMAP III, Vol. 1: Bioprospecting & Ethnopharmacology Eds. J. Bernáth, É. Németh, L.E. Craker and Z.E.Gardner Acta Hort 675, ISHS 2005 174 RNA was extracted according to standard procedures (Sambrook et al., 1989) and the mRNA purified using an Oligotex mRNA purification kit (Qiagen, Hilden, Germany). The mRNA from stems of field grown P. somniferum plants was used to construct a directional SuperScript II RT cDNA library cloned in plasmid pZL1 (Invitrogen, Karlsruhe, Germany) according to the manufacturer’s instructions. The plasmid preparations of randomly chosen individual clones were sequenced from the 5`-end using the ABI PRISM BigDye Terminator v1.1 Cycle Sequencing Kit and run on an ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Weiterstadt, Germany). Database searches were performed using the blastX algorithm (Altschul et al., 1997) on the NCBI server (http://www.ncbi.nlm.nih.gov/). cDNAs chosen for arraying were PCR amplified using vector specific primers. The amplicons were purified using NucleoFast 96 PCR plates (Macherey-Nagel, Düren, Germany), adjusted to a concentration of 100 to 400 ng μl and spotted in duplicates on Biodyne B transfer membrane (PALL, Portsmouth, England) using a MicroGrid II (BioRobotics Ltd). The arrayed PCR amplicons were denatured by incubating the membrane on 0.5 N NaOH/1.5 M NaCl for 5 min, followed by neutralization with 1 M Tris-Cl pH 7.5/1.5 M NaCl and finally immobilized by UV crosslinking at 120 mjoules cm using a Stratalinker 1800 (Stratagene, Amsterdam, The Netherlands). Membranes were prehybridized in 5x SSC, 5 x Dehnhardt solution, 0.1 % (w/v) SDS, 125 μg ml denatured salmon sperm DNA for 3 hrs at 65°C, and hybridised overnight with labelled cDNA in the prehybridisation solution at 65°C. Filters were washed three times with 2 x SSC, 0,1% SDS for 15 min each at 65°C. Data were acquired with a Storm 860 Imaging System (Amersham Phamarcia Biotech, Freiburg, Germany) after the membranes were exposed to Storage Phosphor Screens. Data analyses was performed using the AIDA Image Analyzer Software (Raytest, Straubenhardt, Germany). About 40 μg of total RNA was reversed transcribed using M-MLV reverse transcriptase RNase H minus (Promega, Mannheim, Germany) and the resulting cDNA labelled with [alpha-P]dATP using the Megaprime DNA Labelling Kit as described by the manufacturer (Amersham Pharmacia Biotech, Freiburg, Germany). RESULTS AND DISCUSSION Expressed Sequence Tag Sequencing As a first step toward gene expression analysis, an expressed sequence tag (EST) project was initiated. Since P. somniferum is known to produce most of the tetrahydrobenzylisoquinoline alkaloids and is the only plant containing large amounts of morphine, on which compound our interest is mainly focussed, we decided to perform the EST project with this species. The stem-sections 2-4 cm below the capsule has been shown to possess high biosynthetic activity for alkaloids (Unterlinner et al., 1999; Huang and Kutchan, 2000; Grothe et al., 2001) and were therefore taken as a source for the RNA to construct a stem specific cDNA library. Sequencing of about 2000 randomly chosen cDNA-clones yielded more than 1000 unique sequences. More than half of the sequences showed homology to proteins with known function (Fig. 1). The largest groups of genes are represented by proteins involved in cell cycle, gene regulation, protein synthesis and turnover, as well as stress response proteins and proteins responsible for redox control. Another group of genes highly represented in our database codes for proteins participating in metabolism, mainly primary metabolism. To twenty sequences, a role in secondary metabolism could be ascribed, such as flavanol 3-hydroxylase and caffeic acid 3-O-methyltransferase of the phenylpropanoid pathway, farnesyldiphosphate synthase of the isoprenoid pathway or tyrosine/DOPA decarboxylase and codeinone reductase of alkaloid biosynthesis. The high percentage (40%) of sequences either coding for predicted proteins with unknown function or having no homologs in the databases suggests that P. somniferum stems might be an intriguing source for novel genes.